CLINICAL TRIALS-DESIGN CONDUCT AND ANALYSIS PART - II
Item
- Title
- CLINICAL TRIALS-DESIGN CONDUCT AND ANALYSIS PART - II
- extracted text
-
V.*'
&
n<x4
?y)
z
<L
o
b
o
<?
0
7
7j
•j\
/O
QZ
(1]
7
C4
J
<x
kJ
~z
H
<~Z
D
1
20.3 Treatment monitoring reports 209
20. Interim data analyses for treatment monitoring
meetings of the safety monitoring committee.
The staffs in the data coordinating centers were
responsible for alerting members of the safety
monitoring committees to unexpected changes
occurring between meetings. In fact, the data
coordinating centers in several of the studies
(e.g., Coronary Drug Project and Veterans Ad
ministration Cooperative Study No. 43) distrib
uted interim reports between meetings to allow
members of the committees to call special meet
ings when appropriate. The frequency of meet
ings can be expected to increase as a study nears
a decision point. For example, the Macular Pho
tocoagulation Study (MPS) required two extra
meetings of its safety monitoring committee be
fore it decided in favor of photocoagulation for
patients with senile macular degeneration (Mac
ular Photocoagulation Study Group, 1982).
The usefulness of the monitoring process de
pends on a timely flow of primary outcome data
from the generation site to the analysis center.
It will be reduced by delays in the flow (e.g.,
see Chlebowski et al., 1981). It can also be
diminished by delays in receiving or processing
secondary outcome data based on reading of
records, such as ECGs, X rays, or fundus photo
graphs.
20.2 PROCEDURAL ISSUES
The general steps oulined in Section 17.7 con
cerning preparation of the analysis tape pertain
Several questions must be addressed before any
to interim as well as final analyses. Each moni
treatment monitoring can be done. One has to
toring report should be based on a defined data
do with designation of the individual or group
set that is used to generate all tables in the
responsible for carrying out the analyses needed
report. The analysis tape(s) or disk(s) should be
for monitoring and for generating the treatment
retained for a time following review of the re
monitoring reports. Normally, the responsibility
port. Some tapes (disks), especially those used
is vested in the data center for the trial.
for generation of reports leading to a treatment
A second issue has to do with selection of the
change, should be kept indefinitely.
individual or group having responsibility for re
viewing the monitoring reports and for deciding
whether or not the trial should be allowed to
continue. This review may be carried out by the
20.3 TREATMENT MONITORING
same individual or group that was responsible
REPORTS
for generation of the reports in the first place, or
The discussion that follows assumes that the
by someone else. The latter is the case for all the
reports are generated for review by committees,
trials sketched in Appendix B and is the pre
as described in Chapter 23. Table 20-1 provides
ferred mode of operation (see Chapter 23 for a
a stylized outline of a “typical" report. The out
discussion of treatment monitoring committees).
line assumes that other tabulations needed for
A third issue has to do with the schedule for
performance monitoring are contained in a sepa
interim analyses. They may be done on a fixed
rate report (see Section 16.7 and Table 16-5).
lime schedule (e.g., after every six months) or on
Appendix G contains sample tables from the
one determined by occurrences in the trial (e.g.,
MPS treatment monitoring report and the list of
after a certain number of deaths). All of the
tables appearing in a Persantine Aspirin Rein
trials sketched in Appendix B had schedules (see
farction Study (PARIS) treatment monitoring
item 29.g. Table B 4, Appendix B) that called
report.
for generation of two or three monitoring re
The report should contain a table of contents
ports per year in conjunction with scheduled
reasons to the contrary. (See Table 22-1 for
classes of trials requiring safety monitoring.) Ar
guments concerning the logistical difficulties in
volved in carrying out the monitoring, or that
are based on the assumption that the treatments
are safe are not acceptable. The same is true for
arguments based on the assumption that the
treatment differences will be small.
All of the trials sketched in Appendix B in
clude provisions for treatment monitoring. The
picture appears to be different when viewed
through the published literature. Very few of the
papers reviewed in Chapter 2 contained any evi
dence of such monitoring, even those involving
fairly long periods of follow-up. Either none was
done or the investigators simply failed to men
tion it in their reports.
Interim analyses for treatment effects can be
useful even if not needed for treatment monitor
ing. They help to ensure the orderly develop
ment of methods and procedures needed for anal
yses when the study is finished. In addition, they
may reveal data deficiencies that can be cor
rected by modification of the data forms or
study procedures.
■ t"
Pigs is pigs, data is data.
• &
Jerome Cornfield (1975)
20.1 Introduction
20.2 Procedural issues
20.3 Treatment monitoring reports
20.4 Special statistical problems
20.4.1 The multiple looks problem
20.4.2 The mulitple outcomes problem
20.4.3 The multiple comparisons problem
20.5 Data dredging as an analysis technique
20.6 The pros and cons of stopping rules in
monitoring trials
20.7 Steps in terminating a treatment
Table 20-1 Content of treatment monitoring re
ports
Table 20-2 Ground rules for data dredging via
subgroup analyses
Figure 20-1 Ninety-five percent mortality mon
itoring bounds for the tolbutamide-placebo treatment compar
ison in the UGDP
20.1
INTRODUCTION
An interim analysis is any assessment of data
done during the patient enrollment or follow-up
stages of a trial for the purpose of assessing
center performance, the quality of the data col
lected, or treatment effects. The kinds of tabula
tions and interim analyses needed for perfor
mance monitoring and data quality control are
discussed in Chapter 16. Those discussed in this
chapter relate to the treatment monitoring (also
referred to as safety monitoring; see Glossary)
carried out during the trial.
Major ethical questions arise if investigators
elect to continue a medical experiment beyond
the point at which more prudent people would
have stopped. A case in point is the Tuskegee
Syphilis Study, initiated in 1932 and continued
into the early 1970s. The study involved enroll
ment and follow-up of 400 untreated latent syph
ilitic black males (and 200 uninfected controls)
in order to trace the course of the disease. Criti-
cism of the study stemmed from the fact that the
syphilitics remained untreated after penicillin,
an accepted form of treatment for the disease,
became available (see Chapter 14 for references)'
The need for treatment monitoring extends to
most trials, whether they are done to assess a
therapeutic, prophylactic, or diagnostic proce
dure. and whether they involve a fixed sample or
sequential design, crossed or uncrossed treat
ment structure, short-term or long-term follow
up, or single or multiple clinics. Further, it ex
tends over the life of the trial, beginning with
enrollment of the first patient and continuing to
the end of follow-up, regardless of how and
when treatments are administered and even if
patients are no longer being exposed to the study
treatments.
Investigators have a responsibility to notify
patients as well as the medical community of the
preferred course of treatment once the choice is
clear. Patients assigned to the inferior treatment
should be removed from it (and offered the supe
rior treatment if appropriate) as soon as the
choice is clear.
The general need for treatment monitoring
has been noted by the National Institutes of
Health (National Institutes of Health Clinical
Trials Committee, 1979). Published guidelines
specify that:
• Every clinical trial should have provisions for
treatment monitoring
• The mechanism proposed should be ap
proved by responsible Institutional Review
Boards
• Multicenter trials should have an indepen
dent treatment monitoring committee that
- Includes clinicians with expertise in the
disease under study, biostatisticians, and
scientists from other relevant disciplines
- Excludes physicians caring for patients
in the trial
A good rule of thumb is to design the trial with
treatment monitoring unless there are overriding
208
4
1
210
Interim data analyses for treatment monitoring
Table 20-1
A. Table of contents
b
20.4 Special statistical problems
Content of treatment monitoring reports
• List of tables and figures in report and associated page
numbers
B. Narrative section
• Summary of main findings
• Discussion of special problems influencing interpreta
tion of results
• Procedures used for preparation of report, including
cutoff date for analysis, editing rules, etc.
C. Design summary section
• Purpose of the trial
• List of participating clinics
• Location of data center and other resource centers
• Recruitment goal and sample size specifications
• Study treatments
• Level of treatment masking
• Randomization or treatment unit
• Summary of patient admission criteria
• Prerandomization and follow-up examination sche
dule
• Projected timetable for the trial, including time for
patient recruitment, follow-up. and final analysis
D. Data quality and quantity
• Number of patients randomized by treatment group
and clinic
T.bte 20-1
Number of patients classified as dropouts bs
treatment group and clinic
Content of treatment monitoring reports (commut'd)
. Descriptive tabulations for selected baseline laboratory and physiological measures (e g., cholesterol,
body weight, diastolic blood pressure, etc.) by treat
ment group
. Other summary tabulations of entry characteristics
needed to provide a baseline for evaluation of sub
sequent changes by treatment group, with particu
lar emphasis on known or suspected risk factors for
the disease or outcome of interest
Number of patients lost to follow-up by treat,
ment group and clinic
- Number of missing items of information on com
pleted data forms by treatment group and
clinic
• Distribution of patients by time of entry (used to
indicate the amount of follow-up information bemi
generated and for predicting the amount of dan
that will be available at some point in the future
e g., the number of patients who will have at least
two years of follow-up by the next time the report is
generated)
F Treatment administration summary section
• Number of patients assigned to each treatment group
• Number of ineligible patients enrolled by treatment
• Number of delinquent data forms by clinic (and bv
treatment group if there is concern regarding a dif
ferential delinquency rate, e g., as in unmasked
trials)
group
• Number of patients who refused the assigned treat
ment by treatment group
• Number and percent of deficient data items by clinic
(and by treatment group in unmasked trials)
• Number of patients who received a treatment other
than the one assigned by treatment group
• Inter-aliquot differences in laboratory tests by dime
(and by treatment group in unmasked trials)
• Summary tables describing the level of adherence over
the course of follow-up by treatment group
• Coding and data entry error rates by clinic with dis
tributed data entry systems (and by treatment group
in unmasked trials)
• Number of instances in which treatment assignments
were unmasked (in the case of masked trials) by
treatment group
• Enumeration of special data problems that may influ
ence interpretation of the treatment results
C. Trfitment effects summery section
Number of missed prerandomization and follow
up visits by treatment group and clinic
• Frequency distribution of selected baseline demo
graphic characteristics, such as age at entry, sex.
race, etc., by treatment group
• Number and percent of patients dead by treatment
group
• Percent of patients who experienced the primary out
come at, or before, a specified cut-off date by treat
ment group
(Section A, Table 20-1). Pages in the report
should be numbered and stapled or bound in
some other fashion. The tables and graphs in the
report should have titles that are self-explana
tory. Axes of graphs should be labelled. All
information in the report should be checked for
accuracy prior to inclusion.
Section B, the narrative section, should indi
cate who prepared the summary, the amount of
data included in the report (by indicating the
cut-off date for data), and should include a sum
mary of the key findings contained in the report.
This section should also be used to remind com
mittee members of any deficiencies in the quality
of data and of coding or editing procedures that
might affect the way in which results are inter
preted.
Section C should contain a digest of the key
design features of the trial. The section may not
be necessary if committee members have an inti
mate knowledge of the study and meet regularly.
It is useful for complicated trials and for com
mittees that meet only a few times a year.
Section D should provide data on the nature
of the database. Tabulations indicating the num
ber of missed follow-up visits, the number of
dropouts, and number of patients lost to follow
up are important indicators of the completeness
and adequacy of the database and should be
included in each report.
Section E serves two functions. It should indi
cate the baseline comparability of the treatment
groups and provide a description of the study
population. Knowledge of the study population
characteristics is important for generalization of
treatment findings.
Information on the treatment process is sum
marized in Section F. It should provide data on
patient and physician compliance to the treat
ment protocol. The treatment results are sum
marized in Section G. It is the most important
and largest part of the report.
A typical report may contain a number of
other tabulations distributed throughout the sec
tions already mentioned, or contained in a spe
cial section at the end of the report. Some of
them may be standard and appear in each re
port, whereas others may be prepared in re
sponse to a specific request and may appear only
once.
Reports, after they have been reviewed.
• Summary of missing information as reflected by:
E. Population description summary section
A
211
• Lifetable analysis of the primary outcome to provide
event rates by treatment group over the course of
follow-up
• Percent of patients experiencing an indicated second
ary outcome by treatment group
• Lifetable analysis of each secondary outcome of inter
est by treatment group
• Subgroup analyses by treatment group, using selected
entry characteristics as a means of adjustment for
baseline differences in the composition of the study
group and for identification of treatment effects
within subgroups
• Multiple linear or logistic regression and Cox regres
sion analysis (see Chapter 18) as a means of adjust
ing outcome data for differences in the baseline
composition of the treatment groups
• Treatment comparisons involving the outcome of pri
mary interest by treatment group and level of treat
ment compliance
• Summary table of percentages and rates for the pri
mary and secondary outcomes as contained in cur
rent report as well as corresponding values from
previous reports
• Summary tabulation of patients experiencing indi
cated side effects by treatment group
H. Special analysis section
• Listing of special problems not covered in other sec
tions of the report, especially any that may temper
interpretation of the treatment results
should be stored at a central repository (usually
the data coordinating center). The written rec
ord (minutes of the meeting) generated during
review of the report (also stored in the reposi
tory) should indicate when the report was re
viewed and the specific actions recommended, if
any. as a result of review.
• Special tabulations designed to provide information
on the natural course of the disease under study
20.4
SPECIAL STATISTICAL
PROBLEMS
The need to make periodic treatment compari
sons of the outcome data over the course of
patient enrollment and follow-up gives rise to
what is termed herein as the multiple looks prob
lem. Two other problems, termed herein the mul
tiple outcomes problem and the multiple com-
1
212
w1
J,
•=/
20.4 Special statistical problems
Interim data analyses for treatment monitoring
parisons problem, are likely to be encountered as
well. The first problem is unique to interim anal
yses. The other two can arise in conjunction with
any data analysis, whether done during, or at the
end of the trial.
20.4.1
The multiple looks problem
This problem has been addressed by various au
thors (e.g., Abt, 1981; Anscombe. 1953, 1954;
Armitage et al., 1969; Bailey. 1967; Brown, 1983;
Canner, 1977a. 1977b, 1983a. 1983b; Cornfield.
1966a. 1966b, 1969, 1976; Coronary Drug
Project Research Group, 1972, 1973b, 1981;
Dupont, 1983a, 1983b; National Cooperative
Gallstone Study Group, 1981a; Seigel and Mil
ton, 1983; O'Brien and Fleming, 1979; Royall,
1983; University Group Diabetes Program Re
search Group, l970e, 1971b. 1975). Some inves
tigators have ignored the problem by behaving
as if each look is the only one to be performed
and have followed conventional rules for inter
preting p-values (i.e., have behaved as if a test
result is statistically significant at the 5% level
if its p-value is < 0.05). This approach has ob
vious shortcomings, forcefully illustrated by Ans
combe (1954). He has shown that the probability
of obtaining a “significant" result approaches
unity when a test of significance is performed at
various points over the course of a study.
Cornfield (1976) has commented on the same
problem in more picturesque terms:
Just as the Sphinx winks if you look at it
too long, so, if you perform enough signifi
cance tests you are sure to find significance,
even when none exists.
He relied on the likelihood principle to address
the problem (Cornfield, 1969). Crudely stated,
the principle specifies that the information con
tained in a data set is independent cf the way in
which the set is ordered (Dupont, 1983a, 1983b).
Cornfield’s method of analysis yields two proba
bility calculations—one under the null hypothe
sis of no treatment effect and the other under a
specified alternative to the null hypothesis. The
ratio of the two probabilities has been referred to
as the relative betting odds (RBOs) by Cornfield,
since the resulting value provides a measure of
the support for the null hypothesis, relative to a
specified alternative. (See the University Group
Diabetes Program Research Group, l970e,
1971b, 1975 for illustrations of the method.)
Another approach involves use of simulation
techniques to produce monitoring bounds, such
as those used in the UGDP and CDP (University
Group Diabetes Program Research Group.
l970e, 1975-. Coronary Drug Project Research
Group, 1973b). Figure 20-1 is a reproduction of
the bounds used for the tolbutamide-placebo
mortality comparisons in the UGDP. Figure 18
2 is an illustration of the same concept, as used
in the CDP. The bounds pictured represent, in
effect, the 95% statistical limits of variability that
one would expect in the observed test-control
treatment differences for the method of compar
ison used if it were possible to repeat a trial
many times under the null hypothesis and as
suming a set number of data looks over the
course of the trial. Viewed as a decision-making
tool, a trial continues so long as the observed
test-control difference for the specified outcome
remains within the bounds. The test or control
treatment is terminated if the observed differ
ence crosses one of the boundaries.
213
(a) All cause mortality
TOlBUT*tllt)E VERSUS Pl*CEBO
to
10-
0
ZO
IC
6
U,?,)
Figure 20-1 Ninety-five percent mor
tality monitoring bounds for (he tol-
II
(\'A)
butamide-placebo treatment compari
YEARS OF STUDY
son in the UGDP.
(h) Cardiovascular mortality
?0i TOIBUTBMIOE VERSUS P14CEB0
10-
0-
20.4.2
The multiple outcomes problem
This problem arises whenever two or more out
come measures are used to assess the study treat
ments. The need to look at multiple outcomes
exists in most trials, even those designed to focus
on a primary outcome measure. Analyses are
rarely restricted to that measure alone.
Among the three problems listed, this one is
the most difficult to address. It is complicated by
the fact that the primary and secondary out
comes of interest are likely to be interdependent
(Cuppies et al., 1984). The usual approach is to
ignore the interdependence and to make com
parisons involving the different outcome mea
sures as if they were independent of one another
The practice can lead to erroneous conclusions
unless results are interpreted with caution. For
example, one might be impressed with a statisti
cally significant difference in nonfatal Ml rates
favoring the test treatment in a heart study with
mortality as the primary outcome. However, the
result is only of interest if there was no test
control difference in mortality or if the differ
ence favored the test treatment.
A common practice in trials involving death
as the outcome measure is to focus on cause
specific mortality, for example, cardiovascular
deaths in MR FIT (Multiple Risk Factor Inter
vention Trial Research Group, 1982). The tests
of significance obtained in such cases must be
interpreted in conjunction with those obtained
for overall mortality. For example, the lack of a
statistically significant tolbutamide-placebo dif-
-?0-
(1’)
2
s
4
5
6
?
C^)
8
9
10
12
C.7,)
YEARS OF STUDY
.Sourrr Reference citation 468 Adapted with permission of the American Diabetes Association. Inc.. New York.
ference for overall mortality in the UGDP (pvalue = O.I7) made the study investigators re
luctant to draw any conclusion about the excess
cardiovascular mortality observed, despite its
size (University Group Diabetes Program Re
search Group, l970e).
Certainly, a practice to be frowned upon is
one in which results for only a subgroup of
outcomes are reported, as noted in Chapter 18
in connection with discussion of analysis ground
rule 3. Readers should be provided with results
for the entire set of outcomes (e.g., all deaths)
from which the subset (e.g.. cardiovascular
deaths) was derived or, failing that, for comple
mentary subsets (e.g., cardiovascular versus noncardiovascular deaths).
20.4.3
The multiple comparisons problem
The multiple comparisons problem arises when
an investigator chooses to make several different
treatment comparisons all involving the same
outcome measure (and all done at the same time
point). It has been addressed by various authors,
including Begun and Gabriel. 1981; Dawkins,
1983; Duncan. 1955. 1975; Duncan and God
bold, 1979; Duncan and Brandt, 1983a; Duncan
and Dixon, 1983b; Dunnett. 1955, 1964; Miller.
1966, 1977; O’Brien, 1983; Scheffe, 1953; and
Tukey, 1951, 1977.
The need arises in two general settings. In the
first, the investigator is interested in determining
subgroups of patients within the test-treated
group that appear to be benefited (or harmed)
by the treatment. It can give rise to an indetermi
nant number of comparisons if the subgroups
are identified as a result of data dredging (see
Section 20.5). In the second setting, the investi
gator is interested in comparing each of several
different test treatments with the control treat
ment or with one another. It will give rise to a
minimum of t test-control comparisons—one for
each test treatment. Other comparisons will be
required if the investigator wishes to establish
the superiority (or inferiority) of one test treat
ment relative to other test treatments.
1
214
20.6 The pros and cons of stopping rules in monitoring trials
Interim data analyses for treatment monitoring
Various frameworks have been developed to
deal with the statistical problems involved in
making multiple comparisons. A particularly
simple one is based on Bonferroni’s inequality.
The inequality states that the probability of one
or more k independent events occurring simul
taneously is < kp, where p, the probability of a
given event, is the same for all k events (Abt,
1981; Feller, 1968). The statement can be used to
provide an upper bound on the combined type I
error for making k simultaneous comparisons.
The probability of rejecting the null hypothesis
when it is true with tests of significance for the k
comparisons is ko' if each of the k comparisons
is made at an a' type I error level. The combined
type I error level for all k tests will be less than a
if a' is set equal to a/k. The NCGS used the
inequality to adjust p-values for individual test
control comparisons presented in that study (Na
tional Cooperative Gallstone Study Group,
1981a).
20.5 DATA DREDGING AS AN
ANALYSIS TECHNIQUE
Data dredging is a term used to character
ize analyses that are done on an ad hoc basis,
usually without benefit of a prestated hypothe
sis. as a means of identifying differences of note
within specified subgroups of patients. The sub
groups are typically formed by subdividing pa
tients into mutually exclusive subgroups using
observed baseline characteristics, as illustrated
in Table 18-8.
The practice of data dredging is common and
is not unique to clinical trials. In fact, it is the
hallmark of most epidemiological research con
cerned with identifying etiological factors of dis
eases. Data dredging arises in clinical trials from
the desire to identify subgroups of patients who
are benefited or harmed by the study treatment.
It can occur during the trial or when it is fin
ished. CDP investigators spent a great deal of
time doing such analyses in an effort to under
stand the dextrothyroxine (DT4) treatment re
sults (Coronary Drug Project Research Group,
1970b, 1972). The same was true of MR FIT
investigators trying to decide if antihypertensive
drug therapy for hypertensive men with an
abnormal resting ECG is dangerous (Multiple
Risk Factor Intervention Trial Research Group,
1982).
The main concern with data dredging has to
do with the statistical interpretation of differ
ences found in this way. Table 20-2 lists general
Table 20-2
analyses
form of internal cross-validation in which only a
portion of the data (say half) are used to identify
subgroup treatment differences (e g., see Coro
nary Drug Project Research Group, 1981). The
remaining portion is used to replicate the analy
sis to determine if both portions of the data
identify the same subgroups.
The emphasis here has been on data dredging
involving different subgroupings of the patient.
A variation involves using different outcome
measures. The ultimate form of dredging is to
use the two forms in combination.
Ground rules for data dredging via subgroup
• Limit choice of subgrouping variables to baseline char
acteristics
• Present results for all subgroups defined with a subgrouping variable
• Distinguish between a priori and a posteriori selected
subgrouping variables
• Choose cutting points that are independent of observed
treatment differences
• Avoid conventional interpretation of significance tests
• When possible, validate findings before reporting on
subgroups identified via data dredging
• Report methods and procedures
20.6 THE PROS AND CONS OF
STOPPING RULES IN MONITORING
TRIALS
• Be cautious regarding conclusions
rules that should be followed. They are in addi
tion to those outlined in Section 18.1.
As noted in Table 20 -2, the choice of sub
grouping variables (see Glossary for definition)
should be limited to data collected before ran
domization (i.e., baseline data). Variables ob
served after randomization may be influenced hv
the study treatments and, hence, the subgroups
created using them may be subject to selection
biases, especially those formed using measures
of treatment compliance.
All subgroups formed with a subgrouping vari
able (two if a single cut-point is used as in
Table 18-8, more if multiple cut-points are used)
should be looked at and reported. Failure to do
so can lead to erroneous impressions if a differ
ence observed in one subgroup is offset by a
difference in the other direction in the other
subgroup(s).
The fourth point—independence of the choice
of the cut-point and observed treatment differ
ences—listed in Table 20-2 is basic. It may be
taken for granted when cut-points are set before
the start of data collection or when they are
dictated by the data collection process. It cannot
be if the cut-points are chosen after the start of
data collection.
Investigators should be wary of any “signifi
cant" differences that are found via data dredg
ing. Conventional rules for interpreting a pvalue do not apply to dredged results. Precau
tions are needed to avoid false proclamations
of significance. The precautions may take one
of two forms. The first involves use of some
method for “adjusting" the p-values for the fact
that multiple comparisons were done, as dis
cussed in Section 20.4.3. The second involves a
!
i
A stopping rule is one, usually established be
fore or shortly after the start of patient recruit
ment, that specifies a limit for the test-<ontrol outcome difference which, if exceeded, automatically
leads to termination of one or the other treat
ments depending on the direction of the observed
differences (see articles by Dupont, 1983a,
1983b, and related discussion by Brown, 1983;
Canner, 1983a, 1983b; Greenhouse, 1983; and
Royall, 1983). An example of a stopping rule,
using mortality as the outcome and based on a
standardized comparison of two proportions, is
outlined below. The steps are carried out at each
of a series of designated time points over the trial
until a stopping point is reached or until the trial
is completed.
Sample stopping rule
Step I. Calculate the proportion of patients
dead in the test and control treatment
groups, pt and pc, respectively, at the first
time point at which an interim analysis is
required, as specified when the rule was
constructed.
Step 2. Evaluate the test statistic:
t=
Pt - Pc
Var (pt - pc)
Step 3. Stop the test treatment and conclude
it is inferior to the control treatment if
t > Z. Stop the control treatment and con
clude the test treatment is superior to it if
t < —Z'. Continue the trial if —Z' < t < Z.
(The values for Z and Z' will be set by the
study investigators. Their size will be a func-
I
215
tion of the degree of statistical certainty
desired before stopping and the amount of
“adjustment” to be made for multiple looks.
The two values will be equal if symmetry in
the decision-making process is desired. Z
will be < Z' if more evidence is required to
accept the test treatment as beneficial than
for stopping it because of possible harmful
effects.)
Step 4. Repeat steps I through 3 for each
subsequent time point until the trial is
stopped or until it is finished.
Stopping rules have some appealing features.
They are easy to use and they force investigators
to think about the analysis process and to spec
ify the outcome measure to be used in evaluating
the treatments before the trial starts (see Sec
tion 8.3 and Section 9.3.2). However, they also
have serious limitations. A major one is that it is
virtually impossible to construct rules that deal
with all of the contingencies that can arise dur
ing the course of the trial. Sometimes it may be
necessary to terminate use of a treatment even
though the test-control difference is well within
the range specified by the rule. For example, just
three cases of chronic active hepatitis in the
NCGS were enough to raise serious questions as
to whether to continue the trial, even though the
test treatment showed promise with regard to
the primary outcome measure. Further, even if
one were clairvoyant enough to anticipate the
various conditions that would require stopping
the trial, it is not wise to use statistical tests of
significance as the sole decision-making tool in
the treatment monitoring process. Other factors
that will enter in involve judgments concerning:
• The merits of the treatment
• The availability and usefulness of alternative
treatments
• The seriousness of the conditions being treated
• The acceptability of the treatment to patients,
as evidenced by their willingness to use it,
and by the number of side effects it pro
duces
• The clinical importance of the observed dif
ference
• The consistency of the results with other find
ings in the trial and with other studies
The amount of evidence required for investi
gators to give up on an elective treatment for
which there are alternatives may be less than
that required for a treatment considered to be
life sustaining for which there are no alterna
tives. UGDP investigators terminated use of
1
216
I
Interim data analyses for treatment monitoring
both tolbutamide and phenformin simply be
cause the treatments were no better than the
placebo (University Group Diabetes Program
Research Group, l970e, 1971b, 1975). They did
not consider it appropriate to continue an elec
tive treatment that failed to show any promise of
benefit.
The judgment as to how long the trial should
be continued in the face of a positive result will
be influenced by the size of the difference, the
length of time it took to emerge, and the degree
of certainty investigators have as to the stability
of the results. Investigators in both the Diabetic
Retinopathy Study (DRS) and MPS continued
treatments in those trials for some time after
emergence of a positive result (Diabetic Retinop
athy Study Research Group, 1976; Macular Pho
tocoagulation Study Research Group, 1982,
1983a). They were concerned that the benefits
observed might be offset by subsequent adverse
effects. Long-term follow-up data were needed
before they felt comfortable offering photocoag
ulation for untreated control eyes.
20.7 STEPS IN TERMINATING A
TREATMENT
Once the decision has been made to stop the test
or control treatment a series of steps will be
required to implement it (see Section 15.4 for
details on patient close-out and Sections 23.6
and 23.7 for comments on procedures for recom
mending treatment changes). The first step will
be to present the results to the clinic staff respon
sible for implementing the decision. The presen
tation should be done as soon after the decision
to stop as possible and should be designed to
acquaint clinic staff with the findings and the
reasons for stopping. It may be done from slides
or handouts prepared from the treatment moni
toring report leading to the decision and should
include a discussion of the implications of the
results and of the advice to be given patients
affected by the change.
Clinic staff should be provided with guidance
as to how rapidly they are to proceed in imple
menting the change. Treatments regarded as dan-
gerous will require deliberate and immediate ac
tion. However, even if this is not the case, it is a
good idea to proceed with implementation as
soon as possible. It is not a good idea to wait
until the results are ready for publication, espe
cially if the delay entails continued exposure of
patients to a harmful or inferior treatment.
Records should be kept to indicate when each
patient was contacted regarding the change and
what he was told. Documentation of this sort is
important regardless of whether the patient is be
ing taken off an ineffective treatment or is being
offered a beneficial one.
Obviously, patients affected by the change
should be told of the reasons for the change
However, it is also a good idea to inform other
patients in the trial of the change, even though
they are not affected by it. They may need reas
surance and may be asked to give a new consent
for continuation in the trial (see Section 14.6.5).
Patients removed from a treatment but who
remain associated with the trial may or may not
be given alternative forms of therapy, depending
on the treatment options available. Patients in
the UGDP and assigned to tolbutamide or phen
formin therapy were not offered any other oral
hypoglycemic agent when these treatments were
terminated. Untreated eyes still eligible for treat
ment in the DRS were considered for photo
coagulation treatment when study investigators
were told of results in that trial (Diabetic
Retinopathy Study Research Group, 1976).
Patients should be told if they are expected to
continue under follow-up after a treatment pro
tocol change. The data may be of some value in
characterizing long-term treatment effects. How
ever, their usefulness in this regard will depend
on the extent to which patients are exposed to
other treatments after the change. Investigators
in both the UGDP and CDP elected to have
patients continue on the same clinic visit sched
ule they had before the change. They did so, in
part, for the reason mentioned above and also to
avoid the morale problems and disruption that
might have resulted if some patients had been
separated while others were required to con
tinue.
Part V. Management and administration
Chapters in This Part
21. Funding the trial
22. Essential management functions and responsibilities
23. Committee structures of mulitcenter trials
The first chapter in this Part details the nature of funding vehicles for clinical trials, with
emphasis on NIH grants and contracts. It also contains specific budgeting suggestions for the
various centers in a multicenter trial. Chapter 22 contains an outline of the general principles
and practices to be followed in managing a trial. The last chapter contains a review of
organizational structures used in mulitcenter trials. The chapter discusses a number of
practical issues concerned with the formation and operation of committee structures.
217
1
21. Funding the trial
The hypothesis is unencumbered by any supporting evidence. The budget is the only part of
the application which seems to have any substance whatsoever.
Anonymous NIH study section member
I
i
I
Table 21-4 Direct cost items, by budget cate
21.1 Introduction
gory
21 2 NIH grant proposals
Table 21-5 Direct versus indirect (consortium)
21.2.1 Deadlines and review process
funding for centers in multicenter
21.2.2 Application outline
trials
21.2.3 Content suggestions
Table 21-6 Factors influencing the choice be
21.3 NIH requests for contract proposals
tween direct versus indirect (con
21.3.1 Deadlines and review process
sortium) funding
21.3.2 Factors to consider when deciding
whether or not to respond
21.1 INTRODUCTION
21.3.3 The response
An essential step in the execution of a trial is the
21.4 The study budget
acquistion of funding to carry it out. The ap
21.4.1 Grants
proach taken is influencd by whether the investi
21.4.2 Contracts
gator or sponsor is responsible for initiating the
21.5 Budget breakdown
trial. In practice, some trials, at least the larger21.5.1 Personnel
scale trials, are initiated through the joint efforts
21.5.2 Consultants
of the sponsor and investigator(s). The Coronary
21.5.3 Equipment
Drug Project (CDP) is a case in point. A special
21.5.4 Supplies
committee was convened by the National Heart
21.5.5 Travel
Institute (now the National Heart, Lung, and
21.5.6 Patient care costs
Blood Institute, NHLBI) in early 1961 to ex
21.5.7 Alterations and renovations
plore the desirability, feasibility, and methods
21.5.8 Consortium/contractual costs
needed to initiate a large-scale trial to evaluate
the role of lipid-influencing drugs in the treat
21.5.9 Other expenses
ment of post-myocardial-infarction patients. A
21.5.10 Budget justification
group of investigators worked in concert with
21.6 Preparation and submission of the funding
staff at the Institute to design the study as envi
proposal
sioned by the committee. Funding for the trial
21.7 Negotiations and award
started in 1965, about four years after the initial
21.8 Grant and contract administration
meeting of the special committee (Coronary
21.9 Special funding issues
Drug Project Research Group, 1973a; Zukel,
21.9.1 Direct versus indirect funding for mul
1983).
ticenter trials
The trials sketched in Appendix B represent a
21.9.2 Work unit payment schedules
mix of investigator (7 out of 14), sponsor (5 out
of 14). and sponsor-investigator initiated (2 out
Table 21-1 Number and percent of NIH extra
of 14) trials. See item 6. Table B-4. Appendix B,
mural sponsored trials, by type of
support
for specifics.
No inference can be made as to how a trial was
Table 21-2 Grant application content sugges
initiated from the type of vehicle used to fund it.
tions for clinical trials
The Diabetic Retinopathy Study (DRS) was in
Table 21-3 Questions to be considered when de
vestigator-initiated but was contract-supported
ciding on the merits of a response
over most of its course. The Diabetic Control
to a Request for Proposal (RFP)
219
1
220
-f;
I
21.2 NIH grant proposals
Funding the trial
and Complications Trial (DCCT), initiated by the
National Institute of Arthritis, Metabolism, and
Digestive Diseases—NIAMDD (now the Na
tional Institute of Arthritis, Diabetes, and Di
gestive and Kidney Diseases, NIADDK) has
both grant and contract funding. Clinics are
funded via grants and the data coordinating cen
ter is funded via a cost-reimbursement contract
(National Institute of Arthritis, Metabolism, and
Digestive Diseases, 1981a, 1981b).
Table 21-1 provides information on the use of
grants and contracts for the National Institutes
of Health (NIH) extramural trials listed in the
1979 Inventory of Clinical Trials (National Insti
tutes of Health, 1980). See Section 2.1, for de
tails on how the Inventory is compiled.
and payment if they are funded. The assign
ments may be made in consultation with person
nel from the institutes in question, but the final
decisions are made by DRG staff. The DRG is
also responsible for assigning the applications
for initial review. The reviews are carried out by
the 80 or so chartered study sections,1 or by
special ad hoc study sections. This review struc
ture is in addition to reviews managed by the
various bureaus, institutes, and divisions (BIDs)
of the NIH.
The primary responsibility of the study sec
tions is to assess the scientific merit of research
proposals received by NIH. Meritorious propos
als receive a priority score based on scores as
signed by individual members of the study sec
tion (1.0 for highest scientific merit through 5.0
for lowest scientific merit). This score, along
with a written critique of the application (sum
mary statement), prepared by the executive sec
retary of the study section (from written com
ments provided by members of the review
group), is forwarded to the institute(s) desig
nated by DRG to administer the grant.
The recommendations of the study section arc
reviewed by the advisory council (board) of the
21.2 NIH GRANT PROPOSALS
21.2.1
Deadlines and review process
Deadlines for unsolicited new applications are
February I, June 1, and October I of each year.
Deadlines for unsolicited continuation and sup
plemental applications are March I, July I, and
November I. Deadlines for applications solicited
by the NIH via requests for application (RFAs)
are announced in the solicitations.
All applications are received by the Division
of Research Grants (DRG), where they are as
signed to specific institutes for administration
P
i
Table 21—1
I. A publication. NIH Public Advisory Groups, produced by the
Committee Management Staff of the NIH. lists the chartered
study sections and their membership.
Number and percent of NIH extramural sponsored trials, by type of support
Grant
Number
75
46.9
82
Multicenlert
340
88.3
38
All cancer
415
76.1
120
153
24
177
76.5
33.8
Single center
Multicenlert
All
592
%
%
Number
3
1.9
160
100.0
7
10
1.8
385
545
1000
23.5
0
0.0
I
1.4
34.3
I'
0.4
200
71
271
100.0
64.8
129
35.8
3
0.8
360
100.0
84
18.4
8
1.8
456
100.0
213
26.1
II
1.3
816
100.0
%
Number
51.2
9.9
22.0
47
46
65.3
93
228
63.3
364
79.8
72.5
Number
All
Mixed*
Contract
%
Type
A. Canter Institute
Single center
designated institute. The council is composed of
health researchers plus others from outside the
health field. Members are appointed by the Sec
retary of Health and Human Services (HHS) for
a specified term, usually four years. The meeting
of the council is held about three to four months
after the initial review of an application and
about six to eight months after the deadline for
receipt of the application.
As a rule, only applications recommended for
approval by a study section and approved by an
advisory council will be funded. Most institutes
have the authority to fund a small percentage of
approved applications in the absence of council
approval. However, such actions are rare. The
number of proposals that are actually funded by
any given institute will be a function of the prior
ity scores assigned during the initial reviews, the
size of the institute’s budget, and existing fund
ing commitments.
An applicant will receive a written summary
of the results of the initial review, complete with
priority score, as soon after the review as is
practical. He will receive written notification of
the action taken on his proposal after the council
has met. This notification will be accompanied
by a letter indicating the likelihood of funding in
the case of an approved application. An appli
cant with a proposal recommended for funding
that does not have a priority score above the
payline (see Glossary) will receive notice to this
effect and information concerning prospects for
funding in the future. All such applications are
kept under active consideration for three consec
utive council meetings. They are removed from
consideration if they have not been funded
within that time.
21.2.2 Application outline
1.8
100.0
The outline below is based on details provided in
the NIH grant application package (PHS 398,
revision 5/82).
B. All other institutes
Single center
Multicenlert
All other
100.0
100.0
C. Total (A + B)
•Includes trials with both grant and contract support and trials with both an intramural and extramural component
'f
♦The NIH definition of a multicenter trial is not as specific as the one used in this book and hence includes some
studies that would be classified as single center
Section I: General
• Face page containing project title and other
identifying information
• List of key professional personnel to be en
gaged in proposed project
• Abstract of proposed project (must not ex
ceed designated space)
• Table of contents
• Detailed budget for first 12 months of project
• Budget for total period of support requested
221
• Budgets pertaining to consortium or contrac
tual arrangements
• Biographical sketch of principal investigator/
program director (not to exceed two pages)
• Biographical sketches for other key profes
sional staff (not to exceed two pages per
sketch)
• Sources of salary support (including support
covered in pending applications) for the
principal investigator/program director, as
well as for all other key professional staff
listed in the proposal
• Description of available resources, facilities,
and general research environment
Section 2: Research plan
A. Specific aims (not to exceed one page)
B. Significance of the proposed research (not to
exceed three pages)
C. Progress report/preliminary studies (not to
exceed eight pages)
D. Experimental design and methods
E. Human subjects
F. Vertebrate animals
G. Consultants
H. Consortium arrangements
I. Literature cited
Section 3: Appendix
This section will contain supplementary mate
rials pertinent to the application. Documents
may include published papers, manuscripts still
in preparation, proposed forms for data collec
tion. procedure manuals, etc.
21.2.3
Content suggestions
The grant application kit. aside from the general
outline provided above, does not specify content
requirements. The suggestions contained in
Table 21-2 are those of the author. The appli
cant will have to decide how the material out
lined in Table 21-2 will be organized vis-a-vis
the general outline given in Section 21.2.2. Most
of the items listed in Table 21-2 relate in some
way or other to the research plan.
A well-written application will contain an out
line of the study design, its rationale, and the
procedures that will be used to carry it out.
While it may not be practical to provide a de
tailed protocol and a polished set of data collec
tion forms, sufficient details should be provided
to give reviewers an accurate assessment of the
data collection approaches to be used.
1
*
222
T«ble 21-2
Defects to be avoided include:
Grant application content suggestions for clinical trials
1. Aims and objectives
• Clear statement of the objective of the trial and
the outcome measure to be used to judge the
success of the treatment
• Secondary aims to be pursued in the trial
2. General design specifications
• Rationale for stated goal
• Statistical properties of the proposed recruitment
goal (e g., type I and II error protection provided)
8. Data Intake
• Level of treatment masking
• Specification of types of data to be collected, com
plete with sample copies of data forms, when
possible
• Outcome measure of primary interest
• Staff responsible for data collection
• Proposed length of patient follow-up
• Quality assurance procedures for the data intake
process
• Method of randomization
• General procedures to be used for bias control in
the data collection process
• Baseline and follow-up examination schedule and
rationale for the schedule
• Outline of data collection quality control proce
dures
3. Significance of the study
• Importance of the treatment evaluation proposed
• Potential impact of the trial on future patient care
procedures
4. Timetable
• Anticipated length of the trial, including start-up
period and final analysis
• Time required for protocol development, patient
recruitment, patient follow-up, and final data
analysis
• Method of data entry and for verification of the
accuracy of the data entry process
9. Data processing and analysis
• General methods for receiving, coding, storing.
and processing study data
• Quality assurance procedures used to detect defi
cient data and approach to be used in correct
ing deficiencies
• Approach to monitoring for treatment effects
• Methods for detecting departures from the study
protocol and for monitoring the performance
of participating clinical centers
• Outline of general data analysis plans
10. Study organization
• Description of the test and control treatments
• List of centers to be included in the trial and
description of responsibilities to be performed
by specialty resource centers, such as the data
coordinating center, central laboratory, etc.
• Rationale for choice of treatments, supported
with appropriate literature references
• Composition of the key leadership group and de
scription of its method of operation
• Summary of previous evidence on the safety and
efficacy of the proposed treatments
• Method of creating key committees, including an
outline of membership qualifications
5. Treatment specifications
I
21.3 NIH requests for contract proposals 223
Funding the trial
• Method of treatment administration and level of
masking
6. Study population
• Patient eligibility and exclusion criteria
11. Other procedures
i
I
• Realistic appraisal of ability to meet specified re
cruitment goal using the stated eligibility and
exclusion criteria, preferably done with counts
of eligible patients seen in the clinicfs) over a
specified time period
7. Sample size specifications
• Patient recruitment goal and anticipated time re
quired to achieve it
Investigators should be realistic regarding the
time required for patient recruitment. Expe
rienced reviewers are likely to be skeptical of
claims regarding patient availability and rate of
recruitment unless they are supported with ap
propriate data.
Care should be taken to make certain that
essential data intake and analysis functions are
covered in the proposal. A general discussion,
unrelated to the specifics of the proposal, is
likely to be perceived as a weakness. This is
particularly true if study section members per
ceive a lack of statistical input in the writing
effort.
organiza-
An area often overlooked is the organiza
tional structure of the trial. Organization is im
portant for any activity involving large numbers
of people, whether located at a single center or
multiple centers. The written proposal should
outline the leadership structure proposed and
the methods to be used for coordinating trial
activities.
• Outline of patient informed consent process
• Methods of protecting patient confidentiality
• Provisions for secure data storage
• Proposed source of study patients
• Methods of patient recruitment
• Vague and unsubstantiated claims regarding
patient recruitment
• Unrealistic timetable
• Absence of a rationale for the stated sample
size
,
.
• Clumsily written and fragmented proposal
that lacks cohesion and that conveys the
impression that it was written in haste by
several people who had different percep
tions of the work required
• Lack of organizational details concerning
methods for carrying out the trial
12. Facilities description
• Description of clinic facilities, data coordinating
center, and other resource centers
• List of special items of equipment required for
data collection and analysis
• Description of any other facilities key to execu
tion of the trial
13. Budget and justification
• See Sections 21.4 and 21.5.10
21.3 NIH REQUESTS FOR
CONTRACT PROPOSALS
21.3.1
Deadlines and review process
As a rule, NIH contract-supported projects will
be initiated by the sponsoring institute, via re
lease of a request for proposals (RFP). Unsolic
ited proposals for contract funding are usually
not accepted by the NIH.
Institutes within the NIH are required to ad
vertise their intention to release an RFP in the
Commerce Business Daily at least ten business
days in advance of the projected date of release.
It is also announced in the NIH Guide for
Grants and Contracts. In addition, solicitations
may be advertised in selected scientific journals
and periodicals.
The RFP will indicate the deadline for re
sponse. Responses received after the deadline
will not be considered, unless it is in the govern
ment’s best interest to do so. Requests for exten
sion of the deadline are unlikely to be granted
unless the extension applies to all applicants.
The RFP will indicate where responses are to
be sent—generally, in the case of NIH-released
RFPs, the contracting or review office of the
institute that released the RFP. The technical
merit review of the responses received by the
NIH are either managed by BID review person
nel or, for the smaller institutes, by DRG. The
review process is similar to that described for
grant applications.
21.3.2 Factors to consider when deciding
whether or not to respond
A prospective respondent must decide whether
or not to prepare a response to an RFP. This
decision must be made within a short time pe
riod because of the constraints imposed by the
deadline for response. Questions to consider
when assessing the merits of responding to an
RFP are listed in Table 21-3. The questions are
written from the perspective of an investigator
considering applying for a center in a multicen
ter trial. The questions in Part A are general and
are not related to any particular RFP. Those in
Part B are specific to the RFP in question.
A single, or even a few, negative answers to
the questions listed need not preclude respond
ing to an RFP, but negative answers to key
questions should. The same is true for any RFP
that yields a large number of negative or equivo
cal answers, even if they are not related to key
questions.
A major frustration in preparing a response to
an RFP can be the amount of time available for
response. Most NIH solicitations require a re
sponse within 60 to 90 days. The time between
the date of release and the deadline for response
was as short as 40 days for some of the proposals
reviewed by the Coordinating Centers Models
Project—CCMP (Coordinating Center Models
Project Research Group. 1979b). An investiga
tor should bear in mind that the actual time for
response is always less (sometimes a great deal
less) than the difference between the date of
release and the deadline for response because of
time needed to clear adminstrative channels in
his institution after the response has been writ
ten.
1
224
21.5 Budget breakdown 225
Funding the trial
Table 21-3 Questions to be considered when deciding on the merits of a response to a Request for Proposal (RFP)
Part A.
General questions
Career ftoals
• Is the role proposed compatible with your career
goals and interests?
• Will there be adequate space, office equipment, and
facilities to do the work if you are funded1
• Do you have sufficient time to carry out the study?
• Does your institution have staff with the required
expertise for execution of the study and will you
have access to them if you are funded1
• Do you enjoy collaboration with others'’
• Are your opportunities for promotion likely to be
adversely affected by participation in the project,
especially if there are few if any opportunities for
recognition as a key author on publications gen
erated from the study?
• Can you function in a committee setting, and are
you willing to accept the dictates of such a com
mittee or the sponsor for execution of the trial?
Environment
• Are the stipulations in the business portion of the
RFP compatible with the policies of your institu
tion?
• Is the institution in which you work likely to con
tinue in operation for the period of the trial?
Specific questions concernin|> the RFP
• Is there sufficient time to prepare an adequate rtsponse?
• Is the problem posed worthy of investigation'’
• Is the project likely to achieve its stated aim?
• Does the project have a realistic timetable and is it
subject to modification if necessary?
• Does the sponsoring institute desire strong investi
gator input in the operation of the trial (i.e., does
it desire more than a service role from appli
cants)?
• Will there be adequate lead time for development of
the study protocol and data forms before the trial
is started'’
• Are the suggested funding levels realistic?
• Is the business office of your institution capable of
administering the contract?
• Will it be possible to amend the design and pro
posed operating tenets of the trial, if necessary’’
• Is the trial compatible with the goals of your institu
tion?
• Are the duties of the project officer in the sponsor
ing agency compatible with your perceived role in
the trial?
• Will you be able to obtain the necessary signatures
from administrative personnel in your institution
if a proposal is submitted?
21.3.3 The response
■
Pirt B.
• Are the personnel recruitment practices, pay scales,
and promotion criteria of your institution com
patible with those needed for execution of the
triaP
• Would colleagues view your activities in the trial in
a favorable light?
6
• Will you have the active support of your chief if you
are selected to carry out the proposed work’
The RFP will contain an outline of required
workscope along with a list of general methods
and procedures to be used in carrying out the
work. It may indicate the level of staffing needed
for the study and whether the level stated is to be
considered as an absolute or suggested upper
limit. The limit may be exceeded if the latter is
the case.
The respondent should indicate how the work
outlined in the RFP is to be accomplished. Dele
tions or additions to the workscope as outlined
in the RFP and reasons for the changes should
be noted in the response. Minor changes may be
acceptable if they do not alter the main purpose
or aim of the study Major modifications arc
likely to cause the sponsor to reject the response.
• Are the suggested staffing guidelines realistic?
• Are there adequate provisions for data processing
and analysis outlined in the RFP?
• Is the reporting schedule for progress summaries
during the trial reasonable?
Instructional material accompanying the RFP
should be read before starting work on the re
sponse and should be reviewed during its prepa
ration. The material provided will indicate the
way in which the response is to be assembled, the
number of copies required, the deadline for re
sponse, and where it is to be sent.
21.4 THE STUDY BUDGET
21.4.1
Grants
The budget categories for NIH grant applica
tions are listed below. Indirect costs (see Glos
sary) associated with execution of an NIH grant
supported project are not included in the budget
request, except for indirect costs that are to be
paid to other institutions (e.g., contractual ar-
rangements with other institutions that are out
lined in the application).
MH grant application cost categories
1. Personnel
2. Consultants
3. Equipment
4. Supplies
5. Travel
6. Patient care costs
7. Alterations and renovations
8. Consortium/contractual costs
9. Other expenses
|0. Total direct costs
The budget proposed1 should be a realistic
needed to carry out the
appraisal of what is r.z
study. It should not conform to a preconceived
limit, unless a limit has been set by the sponsor.
Requests that extend over multiple years should
anticipate normal salary increases. The same is
true for anticipated increases in the cost of fringe
benefits for personnel. Some institutes of the
NIH have escalation ceilings that relate to salary
increases in the second and subsequent years of a
budget request (e.g., 6% for NHLBl-supported
projects).
NIH grant applications require a detailed
breakdown of costs for the first year of re
quested support and a summary of costs for
each subsequent year. The detailed breakdown
should include the planned time commitment
(listed as hours per week or as a percentage
based on a full-time effort) and projected salary
support for each person or position listed. De
tailed information is not required for subsequent
years; however, it may be included if the appli
cant wishes to do so. The added detail can be
particularly important if there are large cost in
creases in the second or subsequent years due to
staff additions.
Appendix H contains a sample set of budget
tables, as contained in the budget request for the
Data Coordinating Center in the Hypertension
Prevention Trial (HPT). Only Table H-2 was
required. Tables H-3 through H-7 were con
structed to facilitate the budgeting process and
to provide the reviewers with detailed budgetary
data.
Construction of the budget requires specifica
tion of an anticipated starting date for the pro
posed work. This will be stated by the sponsor in
the case of a sponsor-initiated study and by the
investigator in an investigator-initiated study.
The starting date selected should be at least nine
months after the submission deadline in the case
of investigator-initiated NIH grant applications.
This much time wili be required for the review
and approval process, as outlined in Sec
tion 2I.2.I.
The proposed expenditures should be justified
(see Section 2I.5.IO). While it is true that the
initial review, in the case of NIH funding re
quests, is designed to focus on scientific merit,
budget details and their justification cannot help
but influence the review.
21.4.2
Contracts
Most NIH RFPs contain suggested budget cate
gories. The categories below are from Optional
Form 60—a form produced by the General Ser
vices Administration of the federal government
and which is a standard part of most NlH-released RFPs.
NIH contract cost category
1. Direct material
2. Material overhead
3. Direct labor
4. Labor overhead
5. Special testing
6. Special equipment
7. Travel
8. Consultants
9. Other direct costs
10. Total direct costs and overhead
11. General and administrative expense
12. Royalties
13. Total estimated cost
14. Fee or profit
15. Total estimated cost and fee or profit
Most institutes of the NIH require respon
dents to separate the business and research por
tions of the response. The separation ensures
that the initial review focuses on the technical
merit of the proposal without regard to budget
ary considerations.
21.5 BUDGET BREAKDOWN
Table 21-4 provides a list of items included
under each of the categories listed in Sec
tion 21.4.1 for grant applications. The list is in
tended primarily as a reminder of the type of
items to be considered in the budgeting process.
i
226
Table 21-4
Direct cost items, by budget category
1. Personnel (Individuals with a direct involvement in
the trial and with a slated time commitment. Funds
requested should be for salaries plus fringe benefits.)
• Center director and co-director
• Study physicians
• Clinic coordinator
Direcl cost items, by budget category (continued)
Supplies for special items of equipment, such as
word processors
Photocopying machines
Telephone equipment
• Clinic
Drugs, syringes, etc.
Laboratory reagents and supplies
Miscellaneous office equipment, such as heavyduty staplers, paper cutter. 3-hole punches,
electric staplers, etc.
• Clinic equipment
Data forms
6. Patient care costs* (Funds in this category are used to
pay for procedures carried out on patients that are
done primarily for their research value and that are not
considered necessary for routine medical care. Hence,
they cannot he charged to the patient or his insurance
carrier.)
7. Alterations and renovations*
Furniture for examining and waiting rooms
Patient informational material
• Renovation of a clinic area
• Biostatisticians
Required items of equipment needed for data
collection such as a random-zero sphygmom
anometer or laboratory equipment for spe
cial readings or analyses
Mailers for laboratory specimens
• Air conditioning for computing equipment
Supplies for special items of equipment, such as
film for fundus camera, etc.
• Renovations to accommodate special items of
equipment needed in the trial
• Data coordinator
• Data entry personnel
• Research assistants
• Administrative assistant
• Secretaries
• Clerks
• Other personnel
2. Consultants (Individuals paid on a fee-for-service
basis and who are not part of any center in the trial.)
Consultants may be needed to:
• Provide expert advice in the diagnosis, classifica
tion. or treatment of patients in the trial
• Perform a specialty function, such as reading
ECGs, biopsy material, etc.
• Provide expert advice to a resource center in the
trial, such as to the data coordinating center for
data analysis
• Serve as an expert advisor to the study leadership
or sponsor of the trial
3. Equipment
(Purchased or leased)
• General office equipment
Typewriters
Word processors
41
Transcribing and dictating machines
Filing cabinets
1
Desks, chairs, and tables
• Data center
Computer supplies, such as paper, printer rib
bons. magnetic tapes, disks, etc.
Items of equipment needed for data collection
such as ECG recorder, fundus camera, etc
(Requests for standard equipment, regarded
as essential to any nonstudy clinic setting,
may not he allowed when the budget is re
viewed unless the requests arc adequately jus
tified. The justification should indicate whyexisting equipment will not meet the needs of
the study).
Data entry supplies, such as punch cards,
floppy disks, tape cassettes, etc.
Supplies for special items of equipment, such as
graphics terminal, plotter, microfilm camera,
etc.
8. Consortium/contractual costs (Funds in this cate
gory are used to cover payments to individuals or
groups outside the investigator's institution who have
formal agreements to perform specified functions in
the study.)
9. Other eipenses
• Patient travel to and from clinic
• Equipment maintenance charges
• Data center equipment
• Telephone installation and monthly usage charges
5. Travel
Data entry equipment such as key-to-tape or
key-to-disk units, intelligent terminals, etc
• Study staff
Local (for mileage charges incurred as part of
patient recruitment and home visits)
Computing and related hardware such as tape
and disk drives, printers, remote job entry
stations. CRTs, portable terminals, etc.
National (for travel and living expenses in
curred in conjunction with study-related ac
tivities. including clinic site visits and study
committee meetings, as well as for travel to
selected professional meetings, especially for
presenting study-related papers)
Computing software for database management
and analyses
Mailing equipment, such as postage meter, post
age scale, envelope opener, envelope stuffer
and sealer, etc.
Machines for assembling and binding reports
International (for travel and living expenses for
foreign travel required for study and related
activities, and for selected professional meet
ings* related to the needs and goals of the
study)
• Consultants (for travel to study center or to
study-related meetings)
• Committee members (for travel of members of the
advisory-review committee and the treatment
effects monitoring committee to study-related
meetings)
Paper shredder (for disposing of confidential
records)
4. Supplies
• General office supplies
Paper, pencils, notebooks, typewriter supplies,
dictation tapes, etc.
Postage
Photocopy supplies (e.g., toner, developer, etc.)
• Copying and reproduction charges
• Computing time charges
• Data entry charges
• Study insurance
• Books and journals
• Journal page and reprint charges
• Charges for printing and distributing study forms,
manuals, etc.
• Fee-for-service charges, such as for laboratory de
terminations. reading ECGs. etc., if not covered
under a consultant or contractual agreement
• Space rental
• Moving charges
• Indirect costs for associated contractual services
included in item 8
10. Total direct costs (Sum of cost in above nine catego
ries)
'Categories or items that may not be allowed, or that require special justification.
No single trial will necessarily include all the
items listed.
21.5.1
■J
Tsble 21-4
• Laboratory technicians
• Programmers
I
21.5 Budget breakdown 227
Funding the trial
Personnel
A major portion (from 50 to 80%) of the re
quested support will be for personnel. Actual
salaries expected to apply at the start of funding
should be used for personnel named in the
budget. Salary estimates, using prevailing figures
at the applicant’s institution, should be used for
positions to be filled during the study. The lime
commitments for personnel should be realistic.
That for the center director should be large
enough to represent a meaningful role in the
operation of the center. Padding the budget by
including unnecessary personnel or needlessly
large time commitments is unwise and may lead
reviewers to question the competence or integ
rity of the applicant. In addition, it may cause
them to make drastic cuts in the budget.
|
Ideally, applications should carry named in
vestigators in key support positions. This is espe
cially true for the position of deputy director. An
alternative, when a deputy is not named, is to
indicate the approach to be followed in replacing
the center director, should that become neces
sary. A list of qualifications the individual to be
recruited should have and the mechanism for
screening and selecting a replacement should be
outlined.
Another key support position, vital to data
collection, is that of clinic coordinator. The per
son who fills this position provides a link be
tween patients and physicians in the clinic and
between the clinic and the data center.
Large centers may also require a part-time or
full-time administrator. Generally, the adminis
trative services available through the investiga
tor’s business office will not be adequate to meet
the day-to-day administrative needs of the study.
The justification should indicate why the posi
tion is needed, and why the duties cannot be
1
228
21.5.2
Consultants
Normally, consultants are required to provide
services or fulfill functions that cannot be met by
salaried personnel in the study. They should not
be used to perform essential day-to-day tasks
because of their peripheral role in the study. By
definition, they should be located outside the
applicant’s institution. When they are not, they
should be listed in the personnel section of the
budget.
21.5.3
I
ii
!
si’
II
21.6 Preparation and submission of the funding proposal 229
Funding the trial
performed by personnel in the investigator’s busi
ness office.
Equipment
Items of equipment requested may be purchased
or leased. The approach proposed will depend
on the expected duration of the trial and the
anticipated useful life of the items in question.
Purchase is usually cheaper than lease if the item
is required for three years or longer. Leasing
should be considered for equipment needed for
less time or when there is a chance it may have to
be replaced before the trial is finished. Costs for
equipment maintenance and repairs should be
included in the “other expenses” category.
The request may include funds for office equip
ment, such as typewriters, transcribers, and of
fice furniture, except where such costs are pro
scribed by the sponsor. The request, especially
for large-scale trials, may also include equip
ment needed for data processing as well (see
Section 5.2.4 and Table 5-4 for more details).
<• '•
21.5.4
Supplies
The budget may include funds to cover trans
portation for patients who cannot provide their
own. Trials requiring long-term patient follow
up may even include funds to cover transporta
tion and related living expenses of patients who
must travel long distances to continue in the
study. Costs for patient travel are listed in the
“other expenses”category in NIH grant applica
tions. since they are not considered a part of the
cost for patient care.
21.5.6
Travel
All costs in this category should include direct
as well as indirect contractor or subcontractor
costs This is true for grant applications as well
as contract proposals. A detailed budget, using
the categories listed in Section 21.4.1 or 21.4.2,
should be provided if the contract or subcontract
represents a significant fraction of the total
funds requested.
21.5.9
The budget for the study should include funds
for experimental procedures that are of no direct
benefit to the patient and that are done simply
for research purposes. Ordinary patient billing
and collection practices should be used to re
cover costs for procedures that are considered to
be an essential part of a patient’s care (i.e., those
that would be required whether or not the pa
tient was enrolled in the study). It is prudent,
when in doubt as to whether costs for a proce
dure should be billed to a patient or his insur
ance company, to include costs for the proce
dure in the study budget.
21.5.7
Alterations and renovations
Budget requests for alterations or renovations
can be expected to receive close scrutiny by the
sponsor. Normally, funds are not awarded for
such purposes, at least via the NIH, unless they
are absolutely essential to the study and are well
justified. The guidelines stated for grant re
quests, or as listed in the particular RFP in
question, should be consulted before requesting
funds for this purpose.
21.5.8
The need for money for staff-related travel may
be nil in a trial carried out in a single institution,
but may be sizable in a multicenter trial. Review
groups may question the need for the proposed
travel by study staff. Hence, the applicant must
take pains to explain why it is necessary. Foreign
travel, unless directly related to the study, is not
likely to be approved. The budget of one center,
usually the coordinating center in a multicenter
trial, may carry funds for travel costs not cov
ered in the budgets of the other centers, such as
for study consultants and for members of the
treatment effects monitoring committee.
I
Patient care costs
See Part 4 of Table 21-4 for list.
21.5.5
n-
Consortium/contractual costs
The typical application may not require any
funds in this category. Funds should be re
quested only in instances in which the applicant
proposes to have certain functions fulfilled out
side his own institution (e.g., certain laboratory
tests). The group(s) proposed to perform these
functions and the reasons for selection should be
indicated in the budget request. A contractual or
subcontractual2 arrangement should not be con
sidered if the functions to be performed can be
done better or at a lower cost in the applicants
own institution.
2 The term used depends on whether the parent application i( a
grant or a contract
I
I
Other expenses
This category, as seen in Table 21-4, includes a
variety of items. Several of them, such as com
puting and laboratory determinations, may be
hilled either on a fee-for-service basis or under a
fixed sum agreement. The cost under fee-forservice agreements is determined by the amount
of service rendered, whereas it is fixed in ad
vance under a fixed sum agreement. Agreements
of the latter type are easier to administer than
fee-for-service agreements. However, the options
available in any given case may be limited. For
example, most general-use computing facilities
will be reluctant to provide computing for a
predetermined fixed sum.
Most offices will have photocopying equip
ment that can be used to meet the copying needs
of the project. If so, the budget may simply
include an item for copying charges incurred for
using that equipment. If not, funds should be
included for renting or purchasing needed photo
copying equipment.
Some of the budgets for large-scale multicen
ter trials, such as the CDP and the Hypertension
Petection and Follow-up Program (HDFP), in
cluded funds for study insurance. The protection
provided was over and above that available via
an investigator’s own institution and extended to
all centers in the trial, including the data center,
as well as all study committees, including the
advisory-review and treatment effects monitor
ing committees.
21.5.10
Budget justification
All categories and major items within those cate
gories should be justified. The need for some
items, such as general office supplies and some
of the items in the “other expenses” category,
will he self-evident. However, other items, such
as proposed renovations or alterations, purchase
of costly pieces of equipment, and most travel,
will need careful justification. 1 he personnel
budget, because of its importance, requires de
tailed justification. It should be supported with a
brief description of the duties and responsibili
ties of each staff member or position listed and
the rationale for the stated time commitment.
It may be useful to provide summary tabula
tions, such as illustrated in Tables H-6 and H-7,
Appendix H, to indicate the way in which funds
have been apportioned. The percentage distribu
tion of funds by category of expenditure can
help reviewers judge the appropriateness of the
allocations proposed. Budgets that are topheavy with funds for personnel, relative to funds
for other categories, should be re-examined be
fore submission. Similar tabulations that break
down personnel costs by function to be per
formed (e.g., data generation versus data analy
sis) may help to determine whether the proposed
distribution of personnel is adequate.
21.6 PREPARATION AND
SUBMISSION OF THE
FUNDING PROPOSAL
The preparation of the funding proposal in
volves a great deal more than simply writing the
application and assembling it. Some of the pre
paratory steps include:
• Contacting colleagues to determine if they are
willing to participate in the study and to
reach agreements with them on time com
mitments for the work outlined
• Preparation of updated biographical sketches
for each professional listed in the proposal
• Collection of salary and fringe benefit infor
mation for use in preparing the personnel
budget
• Collection of cost information for items
of equipment, supplies, travel, computing,
and the like
• Collection of letters of agreement from con
sultants and contractors or subcontractors
mentioned in the proposal
Once the application is completed, it should
be reviewed to make certain it is properly pagi
nated. that the table of contents is accurate and
complete, and that all essential materials, such
as biographical sketches and support letters, are
included. The budget should receive special at
tention. Figures should be checked and re
checked for accuracy before the application is
submitted.
The application will not be reviewed by the
NIH without the proper signatures and assur
ances. Grant applications must be signed by the
1
*
230
applicant as well as by the senior administrative
officer of the applicant’s institution. Written as
surance (provided by forwarding a properly exe
cuted Form HHS 596 to the NIH) from the
applicant’s Institutional Review Board (1RB) re
garding the adequacy of the proposed patient
consent procedures and methods for treatment
and follow-up must be received within 60 days
following the deadline for receipt in order for
review to proceed (see Chapter 13).
21.7
f'
tl
I'
I
1'1
I
NEGOTIATIONS AND AWARD
The size of the award will be determined by the
sponsor, using input received during the review
process. In rare cases it can exceed the amount
requested if the sponsor elects to add funds for
expenses overlooked by the applicant. It is more
likely that the sponsor will impose cuts, often
without consulting the applicant. Agencies, such
as the NIH. have formal appeal processes that
can be followed if the applicant feels the cuts are
unjust or that they jeopardize the success of the
project. The appeal may not result in a full resto
ration of funds, but some redress may be possi
ble if the applicant makes a convincing case.
A greater opportunity for negotiation exists
under the contract mode of funding. The peer
review process described in Section 21.3.1 is de
signed to assess scientific merit. Once the reviews
are completed, the proposals are ranked on the
basis of priority scores. Those with the best
scores will be singled out for a second-stage
review. These offerors will be given an oppor
tunity to make their best and final offer and
may be asked to respond (usually in writing) to
a series of questions raised during the initial
review. Final contract negotiations will be un
dertaken with the offeror(s) selected.
Expenditures cannot be made against a pend
ing grant or contract until all necessary docu
ments have been received and signed by the ap
plicant’s institution. These include the Notice of
Grant Award for NIH grants and a signed agree
ment bearing all required signatures for con
tracts. Job commitments should not be made
until all the required documents are in hand.
21.8 GRANT AND CONTRACT
ADMINISTRATION
1
21.9 Special funding issues
Funding the trial
The applicant’s business office is responsible for
receiving money from the sponsor and for mak
ing all payments under the award. Administra
tive questions concerning use of study funds
Tible 21-5 Direct versus indirect (consortium) funding
ti centers in multiccntcr trials _______________________
must be cleared through this office. Questions
that cannot be answered should be forwarded to
the sponsor for resolution. The day-to-day ad
ministrative needs of the project, such as prepa
ration of purchase orders, payroll entries, and
the like, will be met by general staff in the appli
cant’s department or by staff hired specifically
for this purpose using study funds, depending on
the size of the project.
21.9
V Direct funding (Individual awards to each center di
rect from sponsor)
Advantaae^
• Vests all fiscal responsibilities with the sponsor
and thereby helps to maintain a clear separation
of the fiscal and scientific affairs of the trial
• All centers have identical relationship to the spon
sor (i.e.. avoids the unbalanced relationship of
indirect funding where the lead center has direct
funding and all others receive funding via that
SPECIAL FUNDING ISSUES
center)
• Grant or contract administration done through
sponsor, usually by experienced personnel
21.9.1 Direct versus indirect funding for
multicenter trials
• May be perceived by recipients as a more desirable
mode of support than support prov.ded via
another center
A key issue that must be resolved in any multi
center trial deals with the dispersal of funds lo
the individual centers in the trial. In one case,
each participating center submits an application
containing design documents common to the
entire study, plus operational and budgetary in
formation specific for the center. The award is
made to each successful applicant, directly from
the sponsor. An alternative approach involves a
consortium award (see Glossary) in which one
investigator submits a proposal designed to
cover the budgetary needs of all the participating
centers. If the proposal is funded, that same
individual, in conjunction with his business of
fice, assumes responsibility for dispersing funds
to individual centers in the trial. Both ap
proaches are used by the NIH for grant- as well
as contract-funded trials. The applicants have
prime responsibility for choosing the method of
fund dispersal in investigator-initiated multicen
ter trials. The sponsor will have the primary say
in sponsor-initiated multicenter trials.
The advantages and disadvantages of the two
approaches are summarized in Table 21-5. The
factors influencing the choice of one approach
over the other are outlined in Table 21-6.
The main advantage of consortium funding is
the opportunity it provides for reallocation of
funds among centers during the trial. A second
advantage has to do with the mechanics of pre
paring the budget request during the application
process. It is usually much easier for one or two
key investigators to develop a composite budget
for the trial than it is to coordinate development
of a series of budgets needed when each center
is to be funded directly from the sponsoring
agency.
The consortium approach may be necessary in
an investigator-initiated proposal if it is not prac
tical to identify all centers to be included in the
Disadvantages
• Requires a detailed funding proposal from each
participating center
• May preclude the sponsor from redistributing
funds among centers once the awards have been
made (especially true for NIH grant-supported
trials)
• Difficult to coordinate the preparation of individ
ual budget requests, especially if there are only
limited opportunities for contacts among appli
cants when the budgets are being prepared, as in
some investigator-initiated trials
B. Indirect funding
(Awards to individual centers via a
lead center)
Advantages
• Simplifies logistics of preparing the funding re
231
A disadvantage with the consortium approach
has to do with the difficulties any investigator
has in directing both the scientific and fiscal
affairs of a trial. The investigator responsible for
administration of the consortium award must
make fiscal decisions affecting individual centers
in the trial, while at the same time playing a key
role in the scientific conduct of the trial.
The overall administrative cost of either ap
proach is probably about the same. In the one
case the sponsor assumes the major portion of
the administrative burden, whereas in the other
it is assumed by the director of the lead center.
21.9.2
Work unit payment schedules
Payments for clinics may be a function of the
number of patients seen or enrolled. Work unit
payment schedules, while not common in NIHsponsored trials, are used in industry-sponsored
trials. PARIS, one of the trials sketched in Ap
pendix B, had clinic payment schedules that
were based, in part, on the number of patients
seen (Persantine Aspirin Reinfarction Study Re
search Group, 1980a).
The danger with any payment schedule based
on patient load is the temptation it provides for
the enrollment of questionable patients, or even
for falsifying patient reports to maintain a cer
tain income level. Such payment schedules
should not be considered without a monitoring
plan designed to guard against such possibilities.
quest
• Provides flexibility in amount of funds that may be
dispersed to any given center
Disadvantages
• Places a heavy administrative burden on the lead
center
• Fiscal control exercised by the lead center may
have adverse effect on its working relationships
with other centers in the study
• Quality of administration provided by lead center
highly dependent on experience and competence
of the center's business office
• May be viewed by recipients as a less desirable
mode of support than support provided direct
from the sponsor
_____
Table 21-t Factors influencing the choice between direct
versus indirect (consortium) funding
Direct funding
Considered when:
• All centers are to be selected before funding is initiated
• The amount of support required for each center is
above some minimum
• A leadership structure exists to ensure proper coor
dination of the individual funding requests
Indirect funding
trial at the time the proposal is submitted. How
ever, review groups can be expected to have
trouble recommending funding for any study
with unnamed centers, unless they are satisfied
with the process proposed by the applicant for
center selection.
i
Considered when:
• Individual centers require only minimal levels of
funding
• It is not possible or practical to select all centers
before funding is initiated
• It is not practical to coordinate the preparation of a
series of individual funding requests
1
22.3 Patient safety monitoring: An essential function
22. Essential management functions and responsibilities
i.
f
You cannot manage soldiers into battle: you must lead them.
Source unknown
22.1 Management requirements
22.2 Management deficiencies
22.2.1 Failure to delegate authority with re
sponsibility
22.2.2 Inadequate provisions for personnel
backup
22.2.3 Ill-defined decision-making structure
22.2.4 Inadequate funding
22.2.5 Lack of performance standards
22.2.6 Failure to separate essential activities
22.2.7 Ill-defined communication structure
22.3 Patient safety monitoring: An essential
Table 22-1 Classes of trials requiring safetv
monitoring
Table 22-2 Guidelines for committee operalions
22.1
Any research activity that involves multiple
investigators requires a defined structure for per
forming necessary activities. The need is most
apparent in multicenter trials, but it exists in
single-center trials as well, especially if they in
volve various people performing different func
tions. A sound structure will provide:
function
I-
£
i
££
ft.r
MANAGEMENT REQUIREMENTS
22.4 Advisory-review functions
22.5 Committee procedures
22.6 Preferred separation of responsibilities and
functions
22.6.1 Separation of treatment administra
tion and data collection personnel in
unmasked trials
22.6.2 Separation of personnel responsible
for patient care and safety monitor
ing
22.6.3 Separation of investigative and advi
sory-review roles
22.6.4 Separation of sponsor and investiga
tive roles
22.6.5 Separation of data collection and data
processing functions
22.6.6 Separation of centers in multicenter
• Delineation and separation of responsibilities
• A communications structure for disseminat
ing essential information needed by per
sonnel to discharge their responsibilities
• Checks and balances in the decision-making
process
• Specified goals for measuring progress and
performance during the trial
• Ongoing quality assurance and performance
monitoring to detect and correct deficien
cies in the data generation and processing
procedures in the trial
• Appropriate administrative support to imple
ment and carry out functions needed for
execution of the trial
odic intervals over the course of the trial if the
problem is to be avoided.
22.2.2 Inadequate provisions for
personnel backup
Kev positions in the study must be backed up to
assure continuity of operations. Failure to do so
can jeopardize the entire study if a key position
IS vacated at an inopportune time. The backup
should be accomplished by designating deputies
who are empowered to act in the absence of the
persons they represent. The concept applies to
committee positions as well as to positions
within individual centers. The designations may
be via informal understandings for secondary
positions, but not for key leadership personnel.
Formal appointments are required m such cases
io avoid confusion as to the succession of au
thority. It is interesting to note in this context
that only 8 of the 14 trials sketched in Appendix
B had a designated vice-study chairman (line
28.b, Table B-4).
22.2.3 Ill-defined decision-making
structure
Ambiguity in the decision-making structure of a
group is almost always due to the failure of the
group leaders to empower specific members of
the group with the authority needed to perform
designated tasks. Structures with multiple com
mittees or centers that have overlapping do
mains of responsibility are at greatest risk of
haring obscure lines of authority. The reluc
tance of members of the steering committee of a
multicenter trial to vest a designated center or
defined set of individuals with the authority
needed to perform specified tasks in the trial can
lead to chaos. The need to obtain approval from
the committee before each new step is taken
increases the time required to perform a task and
is demoralizing to those who must perform it.
trials
22.2
22.7 Special management issues
22.7.1 Disclosure requirements for potential
MANAGEMENT DEFICIENCIES
22.2.1 Failure to delegate authority
with responsibility
conflicts of interest
22.7.2 Level of compensation for committee
members outside the trial
22.7.3 Review and approval of proposed an
cillary studies
22.7.4 Publication and internal editorial re
view procedures
22.7.5 Publicity and information access pol
icy issues
A common deficiency is one in which a member
of the research team is expected to perform a
specific function but is not given the authorityneeded to carry it out. This deficiency can result
in bottlenecks in decision making. Position-byposition reviews, with the goal of matching au
thority to responsibility, are necessary at pen232
22.2.4
Inadequate funding
Attempting to carry out a trial without adequate
financial support is a serious mistake, especially
if doing so leads to poor quality data or requires
patients to assume risks that could be avoided
with adequate support. Responsible investiga
tors will not start a trial without adequate finan
cial support. Failing that, they will scale down
their efforts to bring them in line with available
funding.
22.2.5
233
Lack of performance standards
There is no way to monitor performance in a
trial in the absence of standards for making such
assessments. A well-designed and well-managed
trial will have a timetable indicating when key
activities, such as patient recruitment, are ex
pected to begin and finish. It will have a patient
recruitment goal and standards of performance
relating to various aspects of the data collection
and analysis processes as well.
22.2.6 Failure to separate essential
activities
Sound leadership requires separation of essen
tial roles in the trial. Some of these separations
are discussed in Section 22.6. Failure to provide
needed separations can result in duplication of
effort, internal squabbles, and biases in the data
collection or analysis of results.
22.2.7 Ill-defined communication
structure
The information flow in the trial should proceed
through designated channels. Structures involv
ing two or more communication routes, such as
in trials with both a data and treatment coordi
nating center (see Chapter 5 for general discus
sion and Sketch 13, Appendix B, for specific
example), must take special pains to make cer
tain that the structure does not produce conflict
ing communiques or allow lapses in the commu
nication process.
22.3 PATIENT SAFETY
MONITORING: AN ESSENTIAL
FUNCTION
Patient safety monitoring, or simply safety mon
itoring. is any process carried out during a trial
that involves the review of accumulated outcome
data for groups of patients to determine if any of
the treatment procedures practiced should be
altered or stopped. This type of monitoring is in
addition to that which is done on a per-patient
basis by each patient's own study physician.
Safety monitoring may be done by a single
person, usually a statistician, in a small trial or
by a committee in a large one (see Chapters 20
and 23). It is essential in any trial of treatments
that carries risks (see Table 22-1 for classes of
trials requiring safety monitoring). Continuation
1
234
22.6 Preferred separation of responsibilities and functions
Essential management functions and responsibilities
Table 22-1
Classes of trials requiring safety monitoring
• Trials with a clinical event as the primary outcome mea
sure
• Trials involving treatments that carry potential short
term or long-term risks
• Trials involving treatments with the potential for pro
ducing serious side effects
• Any trial that has the potential of generating definitive
results before the scheduled conclusion of the trial
• Any trial involving data collection procedures or sched
ules that either entail risks or inconvenience to the
patients or that are costly to carry out
of such trials is justified only so long as there
is no reliable way to choose between the test
and control treatments. Once data have accumu
lated to indicate the superiority of one treatment
over another, the inferior treatment should be
stopped. Ethically, the need for safety monitor
ing is greatest so long as patients continue to be
enrolled and treated in the trial. However, the
need exists even after recruitment and treatment
is finished (e.g., as in a surgical trial where pa
tients are followed after surgery). Patients who
received the inferior treatment by virtue of ran
domization should have an opportunity to re
ceive the superior treatment (assuming they are
still treatable). In any case, the need to inform
the medical community of the finding is the
same, whether or not new patients are still being
enrolled and treated in the trial.
There are also practical reasons for carrying
out periodic data analyses during the trial, even
if the treatments are innocuous and pose no risks
to those receiving them. The costs necessary to
continue a trial are not justified once the accum
ulated results are adequate to make a judgment
concerning the treatments. A trial that no longer
has a chance of producing any more useful infor
mation should also be stopped. By the same
token, use of a data collection procedure that
entails risk, discomfort, or inconvenience to pa
tients should be stopped once sufficient informa
tion has been obtained with it to answer the
question it was designed to address, or once it
becomes clear that no more useful information
can be generated from its use.
22.4 ADVISORY-REVIEW
FUNCTIONS
Most trials will require external reviews for vari
ous reasons over their course. The reviews will
Table 22-2 Guidelines for commitlee operations
be preparatory to the start of the trial or before
implementation of a protocol change during the
trial. Examples include:
• Review of patient recruitment experience in
order to offer advice to the study leadership or sponsor on the need for additional
V General
• Create no more committees than necessary
• Provide a written charge for each committee that out
lines the need for it and the function it is to perform
• Indicate the individual or group that has authority to
appoint or dissolve committees
• Avoid overlap of responsibilities with other commit-
I
clinics
• Review of the performance of a clinic in order
to offer advice to the study leadership or
sponsor on the desirability of continued
funding for the clinic
• Review of a recommendation to add a new
treatment to the trial or to terminate use of
an existing treatment because of adverse
effects
'
tees
• Outline the relationship of one committee to another
and the communications structure for committeeto-committee interactions
• Specify whether or not a committee has decision
making authority; if so. indicate the issues for which
it will serve as the final authority and those for
which it will serve only in an advisory capacity
The advisory and review functions may be
performed on an ad hoc basis by consultants
selected by the study investigators or on a regu
lar basis via a committee appointed by the inves
tigators or sponsor of the trial. Such commit
tees, especially when appointed by the sponsor,
are usually intended to provide advice to both
the sponsor and study investigators. Advice to
the sponsor will focus on policy and fiscal issues.
whereas advice to the study investigators will
center on operational issues.
22.5
B Chsirmtnship
• Specify the method of selection (e.g., election or ap
pointment) and the term of office
• Designate a chairman for each committee created; a
vice-chairman should also be designated for any
committee that is to perform essential ongoing func
tions in the trial
C. Membership
• Specify the membership criteria for each committee
COMMITTEE PROCEDURES
The general rules to be followed when forming
the committee structure in a trial are provided in
Table 22-2. The next chapter provides details on
committee structures for multicenter trials. How
ever, such structures are not unique to multicen
ter trials. They are required in large single-center
trials as well.
No committee should be created without a
written specification of its charge and duties. It
should indicate the composition of the commit
tee, criteria for membership, and voting rights
and rules. Some studies, such as Ml LIS, have
actually written bylaws detailing the committee
structure and governance of the trial (Multicen
ter Investigation of the Limitation of Infarct
Size Research Group, 1983). Such documenta
tion, whether or not formalized to the degree in
MI LIS, is essential in avoiding disputes among
committees with overlapping or competing func
tions. Once created, a member of the study staff
should be charged with the task of maintaining a
running account of committee meetings and of
additions or deletions to the charge of the com
mittee and to its membership. A cumulative rec
ord of committee membership over the course of
235
j
the trial is useful when preparing credits for
manuscripts.
The committee structure may require major
revisions at intervals over the course of the trial.
Revisions may be needed as patient recruitment
is completed and again in preparation for close
out of patient follow-up and for final analysis of
study data. Committees that are no longer
needed should be disbanded.
22.6 PREFERRED SEPARATION
OF RESPONSIBILITIES
AND FUNCTIONS
Preferred separations inlcude:
• Separation of personnel responsible for treat
ment administration and data collection in
unmasked trials
• Separation of personnel responsible for pa
tient care from those responsible for safety
monitoring
• Separation of the investigative and advisory
review roles
• Separation of sponsor and investigative roles
• Specify the methods to be used for rotation of
members (if any), for filling vacancies, and for re
placing nonfunctioning members
• Indicate ex officio committee positions (e g., chair
man of the study, director of the data center, etc.)
• Specify conditions that disqualify individuals from
filling a committee position, including conflicts of
interest
D. Voting
• Specify quorum requirement for conduct of committee business
• Identify voting and nonvoting committee members
and ex-officio voting and nonvoting positions
• Specify committee voting rules
E. Documentation and maintenance
• Maintain an up-to-date list of committee members,
their respective terms of office, and voting rights
• Designate an individual to serve as committee secre
tary
• Carry out periodic reviews in which committee
charges are updated and committee-to-committee
communication structures revised, where appro
priate
• Dissolve committees that have completed their work
or are no longer functional
• Separation of the data collection and data
processing functions
• Separation of centers in a multiccnter trial
Some of the above separations are desirable sim
ply for reasons of efficiency and proficiency, as
suggested in Chapter 5. Others are required for
scientific reasons as detailed below.
22.6.1 Separation of treatment
administration and data collection
personnel in unmasked trials
As noted in Chapter 8. masked administration
of the study treatments offers the best safeguard
against treatment-related biases in the data col
lection process. However, masked treatment ad
ministration is not always possible and therefore
other designs must be used. One approach is to
vest responsibilities for administering the study
treatments with one group, and those for data
collection with a different group (e.g., as was
done to some degree in the Macular Photocoag
ulation Study. Sketch 4, Appendix B. and in the
Hypertension Prevention Trial, Sketch 13, Ap-
1
236
pendix B). The need for separation is especially
important when the outcome measure is subject
to observer error. However, it is not always easy
to achieve, particularly when the patients them
selves may reveal the treatment they have re
ceived while outcome measurements are being
made. It is much easier to ensure when the out
come measure is based on some reading or deter
mination made by personnel who have no pa
tient contact.
22.6.2 Separation of personnel
responsible for patient care and
safety monitoring
1
r
I
22.7 Special management issues
Essential management functions and responsibilities
The ethical underpinnings for a trial rest on the
fact that there is no reliable way to choose
among the treatments being studied. A physician
is willing to allow his patients to be randomized
only so long as he remains uncertain regarding
the relative merits of the study treatments. He is
not, once he believes he knows which treatment
is best. Exposure to data and emerging trends as
the trial progresses may make him reluctant to
enroll new patients or to continue treating those
already enrolled once the data suggest one treat
ment is better than another. Separation of the
type suggested avoids this dilemma by shielding
the treating physicians from emerging trends
and by transferring responsibility for dealing
with them to a group not directly responsible for
patient care (see Chapter 23). A second and per
haps more convincing argument has to do with
the desire to reduce the possibility for feedback
bias in the way in which patients are treated and
in the data collection process. Biases can creep in
if study physicians know the direction of emerg
ing treatment trends.
22.6.3 Separation of investigative and
advisory-review roles
The advisory-review process, by its very nature,
should be performed by a group that is external
to the trial (see Chapter 23). The reasons are
obvious. Advice rendered by personnel in the
study may be self-serving, especially if rendered
by people who are emotionally involved in the
trial and who are dependent on it for salary
support. In addition, some of the issues requir
ing review may concern whether funding should
be continued in a clinic with a poor patient
recruitment or data collection record. Investiga-
tors from within the trial may find it hard to take
an adverse action against a colleague.
responsibility for the data intake and analyses
processes.
22.6.4 Separation of sponsor and
investigative roles
22.6.6 Separation of centers in
multicenter trials
This separation is desirable whether funding is
from government or industry. The preferred ap
proach is one in which all of the essential scien
tific activities for the trial are located outside the
sponsoring agency. This separation is particularly important whenever the sponsoring agency
has a proprietary interest in the product being
tested.
The best example of such separation for an
industry-sponsored trial is provided by the Persantine Aspirin Reinfarction Study (PARIS).
None of the investigative functions in that trial
were associated with the sponsor (Ptrsantine
Aspirin Reinfarction Study Research Group.
1980a). The structure of the Anturane Reinfarc
tion Trial (ART), another industry-sponsored
trial, represents a step in the right direction, but
it failed to provide complete separation. Data
center operations were housed in the firm spon
soring the trial (Anturane Reinfarction Trial Re
search Group, 1980).
22.6.5 Separation of data collection and
data processing functions
Every trial will have activities related to data
collection as well as those concerned with data
processing. These two activities should be car
ried out by different people under different ad
ministrative heads. Effective quality control pro
cedures for the data collection process will be
impossible to implement without this separa
tion.
Data processing operations should be on a
par with the data collection operations in the
trial. The two activities should be provided with
separate budgets via different awards direct from
the sponsor or via agreements with the chairman
of the study as to how funds are to be allocated
for the two activities when they are funded out of
the same budget.
Separation of these two activities is generally
assured in the multicenter trial by creation of a
dedicated data center with its own funding. It
may be more difficult to achieve in the single
center trial. However, it can be accomplished if
the clinic is able to establish a working relation
ship with another department willing to assume
'
237
22.7 SPECIAL MANAGEMENT
ISSUES
22.7.1 Disclosure requirements for
potential conflicts of interest
Participation in a trial as an investigator is a
Centers in a multicenter trial, in addition to
form of public trust. This trust is violated if the
being located outside the sponsoring agency,
investigator has conflicts of interest that bias the
should be administratively distinct from one
data collection, analysis, or reporting processes
another. The separation is usually assured by
of the trial. The mere suspicion of a financial
virtue of geographic location. However, there
1
conflict of interest can cause the public to view
are occasions when two or more centers are
results and conclusions from the trial as suspect.
located in the same institution. For example, the
A case in point, albeit outside the field of clinical
University of Minnesota housed a clinical center,
trials, is the report of the Food and Nutrition
nutrition coding center, ECG coding center and
Board (1980) of the National Academy of Sci
coordinating center for MRF1T (Multiple Risk
ences on dietary standards. The amount of cre
Factor Intervention Trial Research Group,
dence given to the report has been diminished in
1982).
the eyes of some because of relationships of
The relationship of the data coordinating cen
Board members to segments of the food industry
ter to clinics in the trial is of special importance
(Wade. 1980).
because of the key role it has in monitoring the
Some relationships and activities are clear con
data collection process (see Chapter 5). The
flicts of interest and should be avoided (e.g.,
ideal structure is one in which the center is both
working for a firm producing one of the drugs
physically and administratively distinct from all
being tested while serving as a member ot the
other centers in the trial. It may be difficult for
treatment effects monitoring committee). Cer
the center to maintain equity in the way in which
tainly, investigators involved in drug trials
'
it interacts with clinics if it is affiliated with one
should be free of all consulting and retainer
of the clinics in the trial. The actions of the data
arrangements with any drug firms that stand to
coordinating center in monitoring the perfor
gain or lose depending on the results of the
mance of clinics must be viewed by personnel m
trials. The same is true for perquisites (e.g.,
the clinics as being fair and without prejudice if
travel to an exotic place for an investigator and
they are to be effective.
his/her spouse ostensibly for a scientific meet
Structures that provide funding via a consor
ing) offered by manufacturers standing to gain
tium award (see Glossary) to the lead clinic (see
or lose from the trial.
Glossary) should provide a separate budget for
It is prudent to establish mechanisms to moni
the data coordinating center, even if it is located
tor for potential conflicts of interest within the
i
in the same department as the lead clinic. The
investigative group and related committees, in
'
center should not be headed by the director of
cluding the treatment effects monitoring and ad
the lead clinic, or by any other clinic director in
visory-review committees. Systems for disclosure
the trial for that matter. The budget and director
of confiicts are of little value if the persons cov
separation is essential if the center is to have the
ered by the systems are insensitive to the issues
independence it needs to operate effectively.
or activities that can be perceived as constituting
Ideally, all centers in a multicenter trial should
a conflict of interest. Any disclosure system, to
have the same administrative relationship to the
be useful, should be updated at periodic intervals
sponsor, and thereby to one another. The admin
over the course of the trial. Further, the state‘
istrative equality of centers will help to create a
ments filed by members of the study and related
collegial relationship among the investigators
committees must be reviewed by an appropriate
and facilitate interaction and communication.
body (e.g.. the sponsor in government-supported
The potential for administrative equity is greater
trials or the advisory-review committee) as they
when each center is funded directly by the spon
are received to identify conflicts that are serious
soring agency, or by a board as in PARIS (Perenough to disqualify a person from mvolvement
santine Aspirin Reinfarction Study Research
in the trial. All statements filed should be open
Group, 1980a), than when funding is provided
for public inspection once results have been pubvia a consortium award.
i
i
4
1
238
lished (e.g., as was done in the National Cooper
ative Gallstone Study, 1981b).
22.7.2 Level of compensation for
committee members outside the trial
Members of committees, such as the treatment
effects monitoring and advisory-review commit
tees, who are not associated with any center in
the trial, will usually be paid an honorarium or a
consulting fee for meeting activities. Excessively
large payments should be avoided since they
themselves have the potential of creating a con
flict of interest when members are required to
decide whether or not a trial should continue.
Governmental agencies, such as the National
Institutes of Health and the Veterans Adminis
tration, typically pay only travel expenses and a
standard fee of $100 or $150 per meeting day.
Generally, these agencies do not pay members
for time spent in travel to or from meetings or
for time spent in preparing for them.
22.73 Review and approval of proposed
ancillary studies
i
p
a
22.7 Special management issues
Essential management functions and responsibilities
An ancillary study is defined as an investigation,
stimulated by the trial and intended to generate
information of interest to it. It is designed and
carried out by investigators from one or more of
the centers in the trial and utilizes resources of
the trial, but is not a required part of the design
or data collection protocol of the trial. A largescale multicenter trial may spawn a number of
such studies. In fact, the opportunity to engage
in such investigations may represent an induce
ment for investigators to become involved in the
trial in the first place and may help them to
maintain their interest in the trial as it proceeds
(see also Sections 6.3.4 and 6.4.5 and Section
15.2.1).
The leadership of the trial (usually the steering
committee) is responsible for establishing the
general guidelines and policies concerning the
type of studies that may be undertaken, and
for review of proposals before they are imple
mented. The review should focus on:
• Aim and rationale of the proposed investiga
tion
• Type and amount of data to be collected
22 7.5
• Relevance in relation to the main aims of the
trial
• Extent to which investigations are likely to
interfere with patient enrollment and fol
low-up, or with established data collection
procedures
• Possibilities of biasing the data collection or
patient treatment procedures in the trial
• Amount of analytic help needed from the
data center
• Amount of study resources needed to earn
out the investigations
Investigations that have the potential of re
ducing a patient’s willingness to be enrolled into
the trial or to continue after enrollment should
not be undertaken for obvious reasons. Patients
approached for participation should be in
formed of the ancillary nature of the investiga
tions being proposed and of their right to refuse
without affecting their participation in the trial
Investigations in masked trials that entail col
lection of data that have the potential of un
masking treatment assignments in the clinics
should be proscribed, or should be done in such
a way so as to preserve the mask. The same is
true for analyses that have the potential for un
masking treatment assignments. Such analyses
may have to wait until the trial is over and the
treatment codes have been released.
Ancillary studies involving only modest time
commitments from a few study personnel may
be supported with funds from the trial. Large
undertakings, involving major time commit
ments from existing study personnel or the re
cruitment of additional staff, should be done
with independent funding.
One of the issues that should be covered dur
ing the review is the proposed authorship of
papers arising from the investigation and the
credit to be given to the trial in such publica
tions. The arrangement proposed should be com
patible with the general authorship guidelines
established for the trial (see Chapter 24).
22.7.4
Publicity and information access
policy issues
The leadership of the study is responsible for
guidelines concerning investigator-initiated pub
licity during the trial (see Chapter 24). Most
publicity, whether contemplated on a local (e.g.,
m relation to patient recruitment at individual
i
Publication and internal editorial
review procedures
A key issue in any trial has to do with mecha
nisms for the review and authorship of study
publications. The considerations related to this
issue are discussed in Chapter 24.
i
239
clinics in a multicenter trial) or national level
should be cleared through a central body in the
study. Similarly, as noted in Chapter 24, it is
prudent to develop general guidelines to indicate
how investigators are to deal with requests for
information from the news media or from
members of the scientific or lay community
while the trial is under way.
1
23.1 Introduction
23. Committee structures of multicenter trials
Ttble 23-2
Functions and responsibilities of the main organizational units of multicenter trials
A. Study chairman
• Serve as senior executive officer of the investigative
group
• Chair steering committee
• Serve as principal spokesman for the study
• Maintain communications within the study and with
We shall have long sittings, much fighting is anticipated.
Sir Robert Christensen
23.1 Introduction
23.2 Study chairman
23.3 Steering committee
23.4 Executive committee
23.5 Other subcommittees of the steering com
mittee
23.6 Treatment effects monitoring and advisory
review committees
23.7 Committee-sponsor interaction
23.8 Center-to-center communications
Table 23-1 Key organizational units
Table 23-2 Functions and responsibilities of the
main organizational units of mul
ticenter trials
Table 23-3 Functioning committees of the Cor
onary Drug Project (Sketch 6, Ap
pendix B)
Table 23-4 Characteristics of steering commit
tees and committees responsible
for safety monitoring in the 14
trials sketched in Appendix B
Table 23-1
g
1
Figure 23-1 Committee-sponsor interaction
models
23.1 INTRODUCTION
The three functions discussed in Chapter 22.
leadership, safety monitoring, and advisory
review, are usually met through committees in
the typical multicenter trial as considered in this
book. The main organizational units are listed in
Table 23-1. Table 23-2 provides a description of
the principal duties of each unit listed.
g. Steering committee (SC)
• Assume responsibility for general design and conduct
of the trial, including preparation of essential study
documents, such as manual of operations, data
forms, treatment protocol, etc.
• Review data collection practices and procedures, as
summarized in performance monitoring reports,
from visits to participating clinics, and other means,
to identify and correct remediable deficiencies
• Consider and adopt changes in study procedures as
necessary and desirable during the course of the
• Review progress of study in achieving its main goal
and take steps required to enhance likelihood o
success in achieving them
• Review and implement recommendations from the
ARC and TEMC (or ARTEMC) for a treatment
protocol change, such as termination of a treatment
because of lack of efficacy
• Review and react to other general advice or recom
mendations from the TEMC and ARC (or AR
TEMC)
C. Executive committee (EC)
• Act as the administrative and executive arm of the SC
• Make decisions on behalf of the SC on day-to-day
operational issues requiring immediate action
Function
Other designations*
• Head the investigative group and
chair the SC
• Study director
• Principal investigate.’
B Steering committee (SC)
• Leadership body of the investiga
tive group
• Director’s committee
• Executive committee
C. Executive committee (EC)
• Executor of SC
• Chairman’s committee
D. Treatment effects monitoring
committee (TEMC)
• Safety monitoring
• Data monitoring committee
• Data and safety monitoring com
mittee
• Ethics review committee
• Ethics committee
E. Advisory-review
committee (ARC)
• Advise sponsor and/or investiga
tors on conduct of trial
•
•
•
•
F. Advisory-review and treatment
effects monitoring committee
(ARTEMC)
• Advise sponsor and/or investiga
tors on conduct of trial and per
form safety monitoring
• Operations committee
Organizational unit
the sponsor
trial
• Appoint and disband subcommittees needed for exe
cution of the trial
• Make decisions on resource allocations and on priori
ties for meeting competing demands in the trial
Key organizational units
Study chairman
Si
1
Table 23-5 Do’s and don'ts for formation of the
steering committee
Table 23-6 Considerations leading to a separate
ARC and TEMC or a combined
ARTEMC
•Not used in this book.
240
Policy advisory board
Policy advisory committee
Advisory committee
Review committee
241
• Perform executive functions for the trial, including
scheduling meetings, preparation of SC and other
meeting agendas, etc.
• Coordinate preparation of progress reports requested
by the sponsoring agency in conjunction with fund
ing renewal requests and as needed at other times
• Perform other functions assigned by the SC
D. Treatment effects monitoring committee (TEMC)*
• Direct or carry out data analyses needed for assessing
treatment effects during the trial
• Review interim reports prepared by the data coordi
nating center for evidence of adverse or beneficial
treatment effects
• Recommend changes in the treatment protocol to the
ARC
• Provide advice to the SC on operational procedures
affecting the quality of the trial
E. Adriaory|rev*<w committee (ARC)*
• Advise the sponsor on performance of the trial and
whether funding for it should be continued
• Review and approve recommendations from the
TEMC for changes in the treatment protocol
• Recommend termination of support of centers when
warranted because of poor performance or for
other reasons
• Advise the SC and sponsor on important policy issues
• Review performance monitoring reports prepared by
the data coordinating center to detect deficiencies in
the data collection or intake processes and recom
mend corrective action when necessary
• Assume responsibility for external review of the data
coordinating center and other resource centers in
the trial
F.
• Assign priorities for activities in the trial, consistent
with the dictates of the SC
I. In this book, the individual chairing the steering committee is
considered to be chairman of the study.
mittee (ARTEMC)
• Committee has the combined functions of the TEMC
and ARC
•Functions assumed by the ARTFMC in structures not having a separate
The chairman of the study,1 in conjunction
with the steering committee (SC), or SC and
executive committee (EC) when the structure
includes both committees, provides the general
leadership for the trial. Membership on the committee(s) is generally limited to personnel asso
ciated with centers in the trial. Exceptions are
cases in which membership is augmented to in
clude consultants with expertise in areas not rep
resented within the study.
Adrisory-rerit" and treatment effects monitoring com
TEMC and ARC.
The advisory-review functions will be pro
vided through a specially constituted commit
tee. herein referred to as the advisory-review
committee (ARC). The same is true for the
safety monitoring function; generally, it will be
met through a committee herein referred to as
the treatment effects monitoring committee
(TEMC). or through a committee that fulfills
both the advisory-review and safety monitoring
functions, herein referred to as the advisory
review and treatment effects monitoring com
mittee (ARTEMC). The ARC and TEMC or
1
242
ARTF.MC are composed primarily of people
not associated with any of the centers in the trial.
All 14 trials sketched in Appendix B included
a SC. Several (5 of the 14) also included an EC.
Six of the 14 had separate committees for meet
ing the safety monitoring and advisory-review
functions. Four combined the functions into a
single ARTEMC and the remaining4 had only a
TEMC or an ARC, but not both (line 27, Table
B-4, Appendix B).
The committee structure of a trial is generally
more complex than suggested above, as indi
cated in Table 23-3 for one of the trials sketched
in Appendix B—the Coronary Drug Project
(CDP; see citation 104 for details). See also ci
tations 346, 375, and 476 for detailed listings of
Table 23-3
I
23.2 Study chairman 243
Committee structures of multicenter trials
committee structures in the National Coopera
tive Gallstone Study (NCGS), Persantine As
pirin Reinfarction Trial (PARIS), and University
Group Diabetes Program (UGDP). Other references pertinent to the tonic of this chapter in
clude citations 315, 318, and 479.
23.2
STUDY CHAIRMAN
The terms principal investigator and chairman
of the study may be synonymous in a single
center trial, but not in a multicenter trial where
there are. in effect, multiple “principal” investi
gators (see Glossary for comment). Someone
must be chosen or designated to head the inves
tigative group (IG) and to chair the steering
Functioning committees of the Coronary Drug Project (Sketch 6, Appendix B)
CDP committee*
Function
• Steering Committee
• As described in Part B of Table 23-2
• Data and Safety Monitoring
Committee
• As described in Part D of Table 23-2
• Policy Board
• As described in Part E of Table 23-2
• Criteria Committee
• Establish definitions and criteria used for determining
patient eligibility in the trial
* Laboratory Committee
• Review procedures and results produced by the central
laboratory for the trial
• Editorial Review Committee
• Review study manuscripts before presentation or submis
sion for publication
• Statistical Committee
• Advise the Coordinating Center on methods of data anal
ysis
• Natural History Committee
• Direct data analyses and paper writing activities con
cerned with the natural history of coronary heart dis
ease, as based on results obtained from the placebotreated group of patients
• Mortality Classification Commit
tee
• Classify deaths by cause
• Hepatology Committee
• Plan analyses and review data relating to liver function
tests
• Data Repository Committee
• Review ancillary study proposals requiring access to the
main study file and establish special data collection
procedures and respositories for results of ancillary
studies that threaten treatment masking
• Arrangements Committee
• Select site of semiannual investigative group meetings and
coordinate arrangements with host city
• Resources Committee
• Advise the SC on future studies and activities involving
CDP patients
• Newsletter Committee
• Prepare patient newsletter
•Members of all committees, except those serving on the Data and Safety Monitoring Committee and Policy Board,
were appointed by the Steering Committee. Members of the Data and Safety Monitoring Committee were appointed
by the Steering Committee or Director of the NHLBI. Members of the Policy Board were appointed by the Director of
the NHLBI
committee in such trials. This individual is re
ferred to as chairman of the study throughout
this book. Desired qualities of this individual
include:
• A keen intellect
• An understanding of and interest in the area
of study
• Research experience, preferably in other clin
ical trials
• Experience in collaborative research
• A respected research record
• Strong leadership capabilities
• Self-assurance but not arrogance
• The ability to make decisions, but not capri
ciously
• Integrity
• An ability to listen to others and to modify a
stand in the face of convincing arguments
• The ability to compromise
• Evenhandedness and fairness in deailing with
others
• Respect for others and their ideas
• Sensitivity to the needs and feelings of others
• Patience and perseverance
Ideally, the individual selected should be
chosen with this list in mind. However, in actual
fact, there may be little room for choice, espe
cially in an investigator-initiated trial, in which
the individual who conceives the trial is the one
who heads it. Room for choice is greater in
sponsor-initiated trials. The approach used in
the Aspirin Myocardial Infarction Study
(AMIS) serves as a useful model. A temporary
study chairman was appointed by the National
Heart, Lung, and Blood Institute (NHLBI)
shortly after selection of the clinics and the coor
dinating centers for the study. A permanent
chairman was appointed by the Institute some
months later, after input was received from the
investigative group regarding possible choices.
The choice, when one exists, should be limited
to persons who do not have a strong emotional
commitment to any of the treatments being
tested and who have open minds concerning
their merits. For obvious reasons, the individual
selected should be devoid of financial interests in
the treatments under test.
The chairman is sometimes selected by a vote
of the investigative group. In this case, the
choice should be made from a slate of suitable
candidates proposed by a nominating committee
appointed by the investigative group, or that has
been screened in some other way. A popular
election, without any screening, can lead to an
unwise choice (e g., selection of a highly popular
but poorly qualified individual).
All of the trials listed in Appendix B were
headed by persons with M.D. degrees (line
28.a.i, Table B-4, Appendix B). However, only 5
of the 14 chairmen had responsibility for pa
tients in the trials, although several others were
located in institutions housing a study clinic.
The association with a clinic has advantages and
disadvantages. On the one hand, it helps to en
sure that the chairman has firsthand knowledge
of the data collection procedures in the trial. The
knowledge is useful when chairing discussions
concerning protocol changes or when writing
papers containing results from the trial. On the
other hand, the association may make it difficult
for the chairman to maintain a balanced and
evenhanded approach when dealing with other
clinical investigators in the trial, especially if the
clinic he is associated with is one of the more
inept clinics in the trial. Jn addition, the need to
treat study patients, if the chairman has such
responsibilities, may conflict with some of his
other responsibilities (e.g., see Section 23.6).
It is perhaps no accident that none of the
chairmen for the 13 multicenter trials repre
sented in Appendix B were from coordinating
centers or sponsoring agencies. The addition of
chairmanship responsibilities on top of those
normally assumed by the coordinating center or
sponsoring agency is unwise because of the sepa
ration requirements discussed in Chapter 22. In
addition, the added concentration of power at
the coordinating center or sponsoring agency,
through the study chairmanship, may make it
difficult to establish the checks and balances
needed for a robust structure.
As a rule, the chairman will be appointed or
elected to serve for the duration of the trial (see
item 28.a.iv, Table B-4, Appendix B). The ad
vantage of an appointment without term stems
from the continuity of leadership provided when
the same person presides over the trial from
beginning to end. The main disadvantage is that
difficulties can arise if the chairman proves to be
an ineffective leader and must be replaced. A
term appointment can provide a graceful way
out in such cases.
Most of the above considerations pertain to
the position of study vice-chairman as well.
Ideally, the chairman and vice-chairman should
be from different centers in the trial. Location of
both individuals in the same center may concen-
1
23.4 Executive committee 245
244 Committee structures of multicenter trials
trate too much power and influence in a single
center and may restrict the range of ideas pre
sented to the SC and investigative group.
23.3 STEERING COMMITTEE
7'.
^1<1
’■ g
The steering committee is the main leadership
committee of the study. It is the body which is
responsible for overall direction of the study.
Every multicenter trial must face two key
issues in the formation of this committee. The
first has to do with center representation on the
committee. The issue is easily resolved when the
total number of centers in the trial is small (say
ten or less) and where, as a result, it is practical
to have a position for each center on the com
mittee. However, this form of representation is
impractical if the number of centers is large. For
example, this method of representation would
have led to a steering committee of more than 60
members in the CDP. Clearly, a representative
form of government is required in such cases to
avoid the expense, to say nothing of the logisti
cal difficulties, involved in convening the com
mittee. The CDP operated with a steering com
mittee of 15 members by providing for a mix of
standing and elected members. The chairman
and vice-ehairman of the study, director of the
coordinating center, project officer, and direc
tors of the five clinical centers named in the
initial funding application were designated as
permanent standing members. In addition, there
were four elected members, chosen by the IG
from among directors of clinical centers not ac
corded permanent representation on the com
mittee. Elected members served for a three-year
term, with provision for re-election. Terms were
staggered to allow for an orderly rotation of
elected members.
The SC should not be created under the onecenter-one-member rule if there is any likelihood
of having to reconstitute the committee later on
in the trial under a representative form of gov
ernment. It is far better to anticipate the need for
such a form of government from the outset than
it is to attempt to switch to it once the trial is
under way.
Several of the trials listed in Appendix B pro
vided SC representation for each center director,
even though it led, in some cases, to steering
committees with 20 or more members. See item
28.c.iii, Table B-4, Appendix B and Table 23 4
for specifics.
A second issue has to do with the nature of
representation on the committee for key pro
fessional groups involved in carrying out the
study. Formation of the committee along center
lines automatically leads to overrepresentation
of some types of personnel (e.g., clinical investi
gators), underrepresentation of others (e.g.,
personnel concerned with data analysis), and ex
clusion of still others (e.g., junior personnel per
forming essential functions in the trial).
Some studies have attempted to rectify this
problem by reserving positions on the committee
for designated classes of personnel. The ap
proach offers two general advantages. First, it
helps to provide the SC with the expertise
needed to discharge its leadership functions. Sec
ond, it avoids the obvious morale problems that
can arise if an important group of personnel in
the trial has no voice in the way it is run.
Four of the SCs sketched in Appendix B had
clinic coordinators represented (see Glossary for
definition). The advantages of such representa
tion have been discussed by Overton (1980) from
the perspective of the Aspirin Myocardial In
farction Study. It was the only position repre
sented in the trials sketched in Appendix B.
other than the study chairman, vice-chairman,
center directors, and project officers. One reason
for the lack of representation may have to do
with the natural reluctance of any group of se
nior investigators to dilute their base of power
through the addition of members not in key
leadership positions in the study. A second rea
son may have to do with the potential for embar
rassment if a second representative from a center
speaks or votes against a position held by the
center director.
Table 23-5 provides a list of some of the
general rules for SC formation (see Table 22-2
for general committee rules). The term of office
should be designated when positions are filled.
When less than the duration of the trial, terms
should be long enough to permit individuals to
play meaningful roles on the committee. General
rules for filling vacancies should be spelled out
before any are encountered. In addition, it is
wise to indicate conditions that will lead to can
cellation of membership on the committee be
cause of conflicts of interest, lack of interest in
the study as expressed by attendance records at
committee meetings or in other ways. Termina
tion of inactive members, so that their positions
can be filled with new and more active members,
is important in maintaining the vitality of the
committee.
Table 23-4 Characteristics of steering committees and committees responsible for
safety monitoring in the 14 trials sketched in Appendix B
Steering
Committee
Safety
Monitoring
Committee*
• Primary degree
M.D.
PhD
14
0
12
4
• Term
For duration of study
For specified number of yean
14
0
16
0
6
8
I
15
A. Chairman
• Patient care responsibilities
Yes
No
B. Vice-chairman
• Number of trials with vice-chairman
8
2
• Primary degree
M D.
non-M.D.
7
I
2
0
• Term
For duration of study
For specified number of years
8
0
2
0
6
2
0
2
2
5
5
2
7
5
I
I
14
8
14
13
5
13
10
4
13
0
13
0
6
3
6
8
12
2
• Patient care responsibilities
Yes
No
C. Number of members (voting plus nonvoting)
<10
11-15
16-20
>20
D. Study positions represented
Study chairman
Study vice-chairman
Director of coordinating center
Project officer
Clinic coordinator
Clinic director
Nonhealth professional or lay representative
E. Nonstudy members
Yes
No
•Two studies had safety monitoring committees headed by co-chairmen.
23.4 EXECUTIVE COMMITTEE
Most steering committees, even for a trial in
volving as few as five or six clinics, will be too
large to deal with the day-to-day decision mak
ing needed for efficient operation of the trial. A
smaller, more compact committee will be needed
for this purpose (see Part C of Table 23-2). The
EC is usually headed by the study chairman and
includes the study vice-chairman (if there is one),
director of the coordinating center, project offi
cer (in the case of trials funded by the National
Institutes of Health), and perhaps a few other
members of the 1G as well. As a rule, it will meet
more frequently than the SC (either face-to-face
or via conference telephone).
The usefulness of the committee will be de
feated if it has more than a half dozen members
246
Table 23-5
committee
Do’s and don'ts for formation of the steering
DO
• Provide for input from the investigative group when
organizing the committee
• Consult with the sponsor about the functions and
proposed membership of the SC
• Listen to suggestions made by the sponsor regarding
organization and function of the SC
• Outline general membership criteria, methods for se
lecting members, and terms of office
• Provide for representation of all essential skills and
disciplines needed for effective operation of the SC
• Set an upper membership limit on the SC and stick
to it
• Provide for rotation of at least a portion of the SC
members by appointment or election, especially for
trials with large numbers of centers
• Designate the study chairman as chairman of the SC
• Designate the study vice-chairman, director of the
coordinating center, and directors of other key re
source centers as ex-officio voting members of the
committee
• Make the project officer an ex officio (voting or non
voting) member of the committee
• Outline rules for filling vacancies on the committee
• Specify disqualifying conflicts of interest and other
conditions (such as poor meeting attendance) that
will lead to termination of SC membership
DONT
• Limit membership on the SC to center directors or
senior investigators
' I
• Use appointment to the SC as a method of dispensing
rewards or favors
• Include people on the SC known to have conflicts of
interest in relation to the study treatments
• Limit voting rights for selection of elected members
simply to senior members of the investigative group
• Permit the sponsor to dictate the organizing tenets of
the trial
A
'I
I1
I
23.6 Treatment effects m<tonitoring and advisoryreview committees
Committee structures of multicenter trials
or so. The temptation to make the committee
a “mini" steering committee, by including a
number of elected representatives from the par
ent committee, should be resisted. The ability to
convene the committee (by phone or in person)
on short notice will become progressively more
difficult the larger it is.
The concept of delegating executive responsi
bilities to the EC should be established before
the SC is created. Members of the SC may resist
creation of the EC once the SC has been estab
lished, especially if they view the move as one
which lessens their influence in the study.
Only 5 of the 14 trials sketched in Appendix B
had formally constituted executive committees
(line 27. Table B-4, Appendix B). The number of
members ranged from 7 to 10.
23.5 OTHER SUBCOMMITTEES OF
THE STEERING COMMITTEE
The SC may commission a number of subcom
mittees, in addition to the EC, to perform de
fined tasks (see Table 23-3 for list of CDP stand
ing committees). Care must be taken to avoid
needless proliferation of subcommittees and
overlap of functions among the committees com
missioned. The larger the number of commit
tees, the more cumbersome the organizational
structure of the trial, and the greater the likeli
hood of overlap of functions among the commit
tees.
Only committees that are commissioned to
fulfill a continuing need over the course of the
trial should be created on a standing basis. Com
mittees that are commissioned to perform time
limited tasks should be designated as ad hoc
committees and should be disbanded once the
tasks are finished.
Each committee, whether created on a stand
ing or ad hoc basis, should have a defined
charge and should have sufficient authority and
resources to carry out its charge. It should have a
chairman who has responsibility for convening
the committee and for reporting to the parent
committee as needed. Its members should be
derived from the entire IG, not simply from the
parent committees, although each subcommittee
should have at least one member from the parent
committee. The overlap in membership helps to
facilitate communications between the two com
mittees.
23.6 TREATMENT EFFECTS
MONITORING AND ADVISORY
REVIEW COMMITTEES
The discussion in this section assumes the trial is
one that requires both safety monitoring and
advisory-review (see Chapter 22). A key design
question in this context has to do with whether
to vest both functions in the same committee or
in two separate committees. Table 23-6 provi es
an outline of the conditions under which a sepa
rate ARC and TEMC may be needed and where
a single combined ARTEMC may do.
The main advantage of separate committees
has to do with the separation of functions made
247
made, to ensure that the appointments proposed
arc acceptable to both the IG and the sponsor.
Investigators may have no choice but to pro
ceed with formation of their own treatment
A considerations for separate ARC and TEMC
monitoring and advisory-review structure if the
. When treatment monitoring activities require frequent
meetings and where each meeting requires a half
sponsor has no interest in establishing such a
day or more to carry out the necessary data reviews
structure or sees no need for it (e.g.. as in some
• When the TEMC meets other general analysis needs
investigator-initiated grant-supported trials).
of the study (e g., is responsible for developing ana
The lack of cooperation can lead to problems
lytic approaches for dealing with special analytic
later on if the sponsor concludes that the struc
problems)
ture is inadequate to serve its needs and there
• When the trial is investigator-initiated and grant-sup
fore elects to superimpose its own structure on
ported
top of one already in place. This problem oc
• When the sponsor and/or investigators desire sepa
curred in the Program on the Surgical Control
rate committees
of Hyperlipidemia (POSCH). The original struc
g. Considerations for combined ARTEMC
ture provided for both a TEMC and ARC, with
. When the time required for treatment monitoring is
members of both committees appointed by the
small relative to the time required to perform more
study chairman. Later on, as the study pro
general advisory and review functions normally as
gressed and the need in the NHLBI for an advi
sumed by the ARC
sory and review process independent of POSCH
• When there is little or no need for advice or guidance
concerning the analysis procedures used for assess
came to be recognized, the Institute requested
ing treatment effects
POSCH investigators to accept an expansion of
• When the trial is sponsor-initiated
the two committees via the addition of members
• When the sponsor and/or investigator desire a single
appointed by the Institute. This arrangement
combined committee_________________
was sufficient to satisfy the needs of the Institute
until well into the trial. However it ultimately
moved to create its own ARC. This move re
quired dissolution of the existing ARC and led
to a series of discussions (involving the study
possible in this way. The separation, among
chairman, chairman of the TEMC, and NHLBI
other things, helps to ensure that adequate time
staff) to define the domain and responsibilities
will be spent on the safety monitoring process.
of the new ARC in relation to the existing
This assurance is more difficult to achieve when
POSCH committee structure.
the safety monitoring function is only one of a
_____ and ARC, or ARTEMC.
The typical TEMC
an
ARlarger set of responsibilities assumed by
as
seen
through
the sketches in Appendix B, for
as
seen
t
TEMC. The use of separate committees also
the most part, is made up of experts from spe
makes it possible for one committee to serve as a
cialty fields of medicine and biostatistics, al
check on the other for key decisions involving
though, a few included a professional from a
termination of a treatment because of adverse or
nonhealth field (e.g., a lawyer or clergyman) or a
beneficial effects. The main disadvantage is the
lay representative as a means of broadening the
added complexity involved in creating and staf
perspective of the committee (see item 29.e,
fing two committees rather than one.
Table B-4, Appendix B). The virtues of mem
Whatever structure is chosen should be de
bership for a nonhealth professional are dis
signed to meet the advisory-review needs of the
cussed in a paper by Hamilton (1981).
sponsor and of the investigative group. These
Table 23-4 provides a summary of the infor
needs, while overlapping, are different. Coopermation tabulated in Appendix B (item 29, Table
i
ation between the sponsor and investigators will
B-4) for the committees that performed safety
be needed to develop a structure that satisfies
monitoring (TEMCs in nine trials. ARTEMCs in
both needs.
four trials, and the SC in one trial—the UGDr).
Appointments to the TEMC and ARC or ARAll of the committees were chaired by persons
TEMC may be made by the sponsor, or by the
with expertise in epidemiology or biostatistics
study chairman on behalf of the SC. They
and who had an M.D. or Ph D. The number of
should be made by the study chairman in cases
members ranged from 5 to 27. counting voting
where the sponsor has a proprietary interest in
as well as nonvoting members. Several ol the
the treatments being tested. However, it is im
committees restricted voting privileges to memportant, regardless of how the appointments are
T.bk 23-6 Considerations leading to a separate ARC
ind TEMC or a combined AK i t-.iviv.
1
23.7 Committee-sponsor interaction 249
248 Committee structures of multicenter trials
V’
?
I
fI
bers not affiliated with any study center. This
restriction is a good idea, especially in cases in
which study members are dependent on the
study for salary support. Most of the committees
included the director of the coordinating center,
either as a voting or nonvoting member. Rep
resentation from this center is essential in the
monitoring process because of its key role in
data analyses. All of the NIH-sponsored trials
sketched included representation of the project
office on the committee as well. The majority of
the committees also included the study chairman
or vice-chairman, again as either voting or non
voting members.
Inclusion of the study chairman on the TEMC
or ARTEMC is open to debate when that per
son has patient care responsibilities in the trial.
The emotional commitment needed to treat can
affect any person’s scientific objectivity thereby
reducing that person's effectiveness on the com
mittee. Further, it can be argued that a study
physician who has access to interim results of the
trial will be affected by them, thereby increasing
the risk of bias in the treatment and data collec
tion processes performed by that person. In ad
dition, having access to the results can create a
dilemma for any study physician still involved in
recruiting patients for the trial if they suggest
one treatment is better than another, even if the
trend is not large enough to justify stopping the
trial. Shielding study physicians from the in
terim results protects them from the dilemma
mentioned by transferring responsibility to the
TEMC or ARTEMC.
The virtues of inclusion have to do with the
special qualifications of the study chairman.
This person may have the best perspective on the
trial and its data collection and treatment pro
cesses—a perspective that may be invaluable
when the TEMC or ARTEMC is faced with a
major decision concerning the study. In addi
tion, the chairman’s presence on the committee
can be reassuring to other clinical investigators
in the trial. In fact, they may be reluctant to
delegate responsibility for safety monitoring to
any group without such representation.
The approach practiced in some studies has
been to include both the chairman and vicechairman of the study on the treatment effects
monitoring committee, whether or not they have
treatment responsibilities in the studies. For ex
ample, the CDP opted for this approach, even
though the vice-chairman of the study
had such responsibilities. The TEMC in the Hy
pertension Detection and Follow-Up Program
(HDFP) was reconstituted during the trial to
include both the study chairman and vice-chair
man.
Ideally, persons selected to serve on the
TEMC, ARC, or ARTEMC should have prior
experience with multicenter clinical trials. This is
especially true for the chairmen of these commit
tees.
All voting members should be screened for
conflicts of interest before appointment. In addi
tion, mechanisms should be established to alert
the appointing authority to conflicts of interest
that may develop during the trial (see Section
22.7 of Chapter 22).
Members of the treatment effects monitoring
and advisory-review committees are usually ap
pointed for the duration of the trial. None of the
TEMCs, ARCs, or ARTEMCs listed in Appen
dix B made any provision for term appoint
ments. Undoubtedly, this is due to the desire of
the study leaders to maximize continuity of func
tion in these committees via a stable member
ship. However, this approach, as suggested in
Section 23.3, can cause difficulty if members lose
interest in the trial. Hence, it is prudent to have
some means of dismissing inactive members in
the absence of term appointments in order to
maintain a properly functioning committee.
Committee-sponsor interaction models
Figure 23-1
Model
Characteristics
Model A. Sponsor Directive
• Trial usually sponsor iniliated
• Members of ARTEMC appointed by sponsor, sometimes with
little or no investigator input
• Advisory and review functions provided by ARTEMC, as pre
Sponsor
scribed by the sponsor
• Communication between SC and ARTEMC via the sponsor
• Little or no direct communication between SC and ARTEMC
ARTEMC
SC
the SC for implementation
• Trial usually investigator initiated and grant supported, or in
Model B. Sponsor Mondirective
itiated via a joint effort of the investigators and sponsor
• Members of the ARC and TEMC may be appointed by the spon
sor or chairman of the SC
• Appointments made to the ARC and TEMC generally limited to
individuals who are acceptable to both the sponsor and investi
gative group
Sponsor
• Advisory and review functions provided by ARC, as prescribed by
mutual consent of investigative group and sponsor
• ARC provides advice and review for both the sponsor and investi
gators
ARC
SC
n
TEMC
23.7 COMMITTEE-SPONSOR
INTERACTION
Smooth interaction of the SC with the TEMC
and ARC, or with the ARTEMC in the case
of a single committee with combined advisory
review and safety monitoring functions, is es
sential for operation of the trial. Figure 23-1
provides stylized diagrams of three types of in
teraction models, as viewed from the perspective
of the sponsoring agency. It also outlines the
main characteristics of each of the models.
Communications between the SC and the
sponsor in the models are concerned with design
and operation of the trial. Communications be
tween the ARC or ARTEMC and the sponsor
are concerned primarily with assessment of the
adequacy of the study design, the nature of the
treatment results, and with fiscal affairs. None of
the models in Figure 23-1 provides for flow of
treatment results to the SC during the trial. In
formation of this sort passes only in conjunction
with a recommended treatment protocol change.
Each of the trials sketched in Appendix B has
been classified (by the author) as to type of
communication model using the criteria given in
• Recommendations for treatment change made by the ARTEMC.
Those approved by the sponsor are passed, via the sponsor, to
• Primary communications of the TEMC directed to ARC; limited
communications from the ARC to TEMC
• Recommendation for treatment change originates with TEMC.
reviewed by ARC. and passed to SC, via the sponsor, for imple
mentation
• Primary communications from the ARC to sponsor and from SC
to sponsor. Only limited communication from the sponsor to the
SC. or from the sponsor to the ARC. No direct communication
between TEMC and SC or between TEMC and the sponsor
• Some direct communication between SC and ARC. but not with
regard to treatment results
• Trial usually small-scale, investigator initiated, and grant sup
Model C. Sponsor Passive
ported
• Advisory and review function provided by ARC. as prescribed by
the SC
• ARC has no advisory-review role for the sponsor
Sponsor
• Members of ARC and TEMC appointed by chairman of SC. Little
or no interest expressed by the sponsor in the selection or ap
pointment process
• Virtually no communication from sponsor to SC or from sponsor
SC
ARC
n
TEMC
to ARC
• Limited communications from SC to sponsor and from ARC to
sponsor
• Same communication structure as for Model B regarding ARC
and TEMC
• Major communications between SC and ARC
• Recommendation for treatment change originates with TEMC.
reviewed by ARC. and passed by the ARC to the SC for imple
mentation with knowledge of the sponsor, but without its ap
proval
Notr: Arrows indicate direction of communications. Solid lines indicate major communication pathways Dashed lines indicate
secondary communication pathways.
V
250
Committee structures of multicenter trials
Figure 23-1. Four of the 14 trials were classified
as sponsor-directive and the remaining 10 were
classified as sponsor non-directive (see item 30,
Table B-4, Appendix R for specifics).
23.8 CENTER-TO-CENTER
COMMUNICATIONS
The coordinating center (or data coordinating
center) is the primary communication channel in
most multicenter structures. The center will re
quire linkages with each clinic as well as with all
other organizational units in the trial to perform
its functions effectively. The volume of informa
tion flowing in other communications channels,
such as those associated with the chairman’s of
fice, project office, and other resource centers in
the trial, will depend on the way in which coordi
nation responsibilities are divided (see Chapter
5). Generally, it will be small compared with that
of the coordinating center or data coordinating
center.
Information flowing from the clinics to the
coordinating center (or data coordinating cen
ter) will consist primarily of data, either as con
tained on completed data forms or as tapes or
disks of data already keyed at the clinics from
the data forms. Information flowing from the
coordinating center to the clinics will relate to:
• Edit queries concerning completed data
forms
• Procedural memos concerning the data col
lection process
• Manuals of operation or parts or sections of
manuals and related revisions
• Approved data forms and related revisions
• Progress reports and clinic performance mon
itoring reports
• Minutes of study meetings
• Miscellaneous study correspondence
I
■
1
The flow into and out of the clinics should be
via a defined pathway. Multiple entry and exit
points for the clinic will make it difficult, if not
impossible, to control information leaving the
clinic and to keep track of information flowing
into it. The preferred structure is one in which a
single person, usually the clinic coordinator, is
designated to serve as the conduit through
which correspondence and materials flow into
the clinic and through whom data forms and
related materials flow out of the clinic. One per
son who should not serve as a primary channel
for routine information flow is the clinic direc-
23.8 Center-to-center communications
tor. The likelihood of a smooth and continuous
flow is low because of the multiple commitments
and general lack of discipline and interest of
such a person in handling routine flows.
The designation of a single individual in the
clinic to receive and distribute important proce
dural information arriving from the coordinat
ing center (or data coordinating center) simpli
fies the distribution task of the coordinating
center. Mailings from the coordinating center to
several persons in each clinic is expensive and
may not, in any case, work as well as local
distribution systems. However, some mailing re
dundancy is wise to protect against communica
tion breakdowns if the primary channel fails.
The HPT data coordinating center used a pri
mary (clinic coordinator) and secondary (clinic
director) mail contact for all communications
concerning data collection and study proce
dures. The primary mail contact received origi
nals and all accompanying attachments to mail
ings from the data coordinating center. The
secondary contact received, via a separate mail
ing. copies of all numbered memos and a list of
attachments received by the primary mail con
tact.
The address directory is an essential commun
ications aid in the multicenter trial. To be useful,
it must be up-to-date and should contain the
names, mailing addresses, and phone numbers
of all study personnel at each participating cen
ter. Other useful information in the document
includes:
• Manuals of operation
• Materials used by clinic staff and patients for
treatment administration
The ground rules for document numbering
and communications should be established be
• An indication of the functions of each person
listed, including areas of certification for
data collection
• The name of the primary and secondary mail
contacts at a center
• The deputy director of a center
• List of study committees and the names, ad
dresses, and phone numbers of committee
members
The information system of a trial will require
a numbering scheme to facilitate the identifica
tion of the various documents in the study. The
need for a form numbering scheme has already
been addressed (Chapter 12). However, the need
does not stop there. It extends to other docu
ments as well, such as:
• Committee minutes
• Procedural memos, etc.
MpHOC
251
fore data collection is begun. Rules should be
written and reviewed by the SC before they are
promulgated in the study. The rules should be
reviewed and, when necessary, revised at inter
vals over the course of the trial.
i
Part VI. Reporting procedures
Chapters in This Part
24. Study publication and information policies
25. Preparation of the study publication
26. Locating and reading published reports
Chapter 24 deals with general policy issues involved in the production and publication of
study manuscripts. Chapter 25 outlines the content requirements of a finished report and the
steps involved in the preparation of such reports. The last chapter contains a review of
methods for locating reports of trials in the published literature. It also contains a list of
points a reader should consider when reading a report. It closes with a discussion of the
responsibilities of persons who critique study publications.
<¥•
253
1
24. Study publication and information policies
Never be so brief as to become obscure.
I
24.1 Information constraints
24.2 Publication questions
24.2.1 When to publish?
24.2.2 Presentation or publication?
24.2.3 Where to publish?
24.2.4 What to publish?
24.2.5 Journal supplements versus regular
issues
24.3 Authorship and internal review procedures
24.3.1 Introduction
24.3.2 Individual versus corporate authorship
24.3.3 Writing responsibilities
24.3.4 Credit rosters
24.3.5 Internal review procedures
24.4 Information access policy issues
to study data during the trial by
24.4.1 Access
/
outside parties
24.4.2 Access to study data at the conclusion
of the trial
24.4.3 Access to study forms and manuals
24.4.4 Inquiries from the press
24.4.5 Special analyses in response to criticisms
i
*
f
4;
24.4.6 Outside audits
Table 24-1 Pros and cons of interim publica
tions not related to a treatment
protocol change
Table 24-2 Options for initial communication
of results
Table 24-3 Long versus short papers
Table 24 4 Pros and cons of individual versus
corporate authorship
24.1
INFORMATION CONSTRAINTS
The types of trials described in this book typi
cally require constraints on the flow of informa-
Tyron Edwards
tion while they are under way. Commonly im
posed constraints relate to:
• Randomization (e.g., by withholding details
concerning the randomization process
from clinic personnel to keep them from
predicting future assignments; see Chapter
10)
• Treatment masking (e.g., by constructing
methods for assigning and administering
treatments so that a patient and his doctor
remain masked with regard to treatment
assignment; see Chapters 8 and 10)
• Data collection and coding procedures (e g.,
through separation of treatment and data
collection responsibilities in the clinic so
that observations are made and recorded
by personnel who are kept ignorant of the
treatment received by the study patients;
see Chapter 8)
• Treatment monitoring where only selected
members of the study organization are
privy to interim treatment results (see
Chapter 20)
All of the trials listed in Appendix B imposed
constraints of these types. All of the coordinat
ing centers withheld details concerning the
method of randomization until recruitment was
finished or until the end of the study. Most had
structures that allowed only certain members of
the study group to see interim treatment results.
Clinic personnel, as a rule, were not allowed
access to outcome data by treatment group until
the study was concluded, or until a treatment
was terminated.
Information constraints, whether limited to
members of the community at large, or to se
lected studv personnel as well, should not be
imposed unless there is a good rationale for
doing so. Further, they should be lifted as soon
as the need for them no longer exists.
255
1
256
Study publication and information policies
24.2
PUBLICATION QUESTIONS'
24.2.1
24.2 Publication questions
Table 24-1 Pros and cons of interim publications not
related to a treatment protocol change
When to publish?
Any investigator who undertakes a trial has a
responsibility to make the results obtained from
it available for public scrutiny via a published
manuscript. The manuscript should be prepared
and made available as soon after the results have
been obtained as possible. Normally, the manu
script (or manuscripts) describing the results will
be produced after the trial has entered the termi
nation stage (see Chapter 3 for stages). Excep
tions are cases in which interim publications are
needed to report results related to a treatment
protocol change, as in the Coronary Drug Proj
ect (CDP), Macular Photocoagulation Study
(MPS), and University Group Diabetes Pro
gram (UGDP). See reference citations 102, 103,
105, 291, 292, 293,468,470.472 for publications
of this type.
Investigators in long-term trials should decide
whether or not to allow publication of interim
results not related to a protocol change. The
pros and cons of such publications are outlined
in Table 24-1. The preferred policy is one pro
scribing publication or presentation of treatment
results during the trial, except those related to
protocol changes. A permissive policy has the
potential of compromising the trial, especially in
cases where the results can affect subsequent
recruitment or treatment patterns in the trial. In
addition, it can open the study to criticism if the
schedule of publication is perceived as having
been designed to maximize the impact of the
study.
Pressures to relax the proscription can be ex
pected during the course of most long-term
trials. They are most likely to arise from publica
tion of related studies, especially if the results of
these studies are contrary to those observed in
the trial. Investigators in the Coronary Artery
Surgery Study (CASS) were exposed to such
pressures because of interim publications com
ing from a European sister study (European Cor
onary Surgery Study Group, 1979, 1980, 1982a,
1982b). Ultimately, the proscription was upheld,
but not without a considerable amount of de
bate.
Investigators in some of the larger trials have
elected to summarize details of the design, methI. The term publication, a* med throughout this chapter and in
the next one, relates to a public document Access may be via a
published periodical, book, or the like, or via a public repository
for unpublished manuscripts and documents, such as the one
maintained by the National Technical Information Service (NTIS).
Pros
• Provides access to study results as they emerge
• May simplify preparation of the final publication
age), with subsequent publication in a med
ical journal
• Publication of the results in a medical jour
nal, with no prior public presentations or
announcements
• Publicity accorded the interim results may reduce the
amount of attention investigators are able to devote
to the conduct of the trial, especially if resources
needed to carry out the remainder of the trial have
to be diverted to respond to criticisms
Table 24-2 contains summary comments con
cerning each option.
A presentation should not be made if it pre
cludes publication in the journal of choice. Some
journals may regard certain kinds of presenta
tions as tantamount to publication and, hence,
may not be willing to publish the results. In
addition, various journals, such as the New Eng
land Journal of Medicine, discourage authorinitiated press coverage of results in papers
under consideration for publication.
Publicity emanating from presentations may
not be in the best interest of patients or members
of the medical community responsible for their
care. especially if the results presented are con-
• Impact of the study and its final conclusions may he
diminished because of the way data were presented
and analysed in earlier publications
T«bte 24-2
• Helps to maintain investigator interest in the trial
• Keeps the study in the “public eye"
Cons
• Inconclusive and preliminary nature of the results
may lead to confusion
• May reduce investigator enthusiasm for continued pa
tient recruitment or treatment if the interim results
are viewed as "discouraging"
• Knowledge of an emerging treatment difference, espe
cially in the case of unmasked trials, may bias sub
sequent treatment and data collection
ods, and baseline results in a separate paper.
Such papers may be prepared any time after
patient recruitment is completed. Ideally, they
should be published before any results for the
trial have appeared in print, as in CASS (Coro
nary Artery Surgery Study Research Group,
1981), or in conjunction with the first results
publication, as in the UGDP (University Group
Diabetes Program Research Group, 1970d).
However, in some instances they may not appear
until results have been published, as in the CDP
(Coronary Drug Project Research Group.
1973a).
24.2.2.
Presentation or publication?2
A key issue to be addressed is the way in which
results are announced to the medical commu
nity. Options include:
• Making an announcement of the results to
the news media with subsequent publica
tion in a medical journal
• Making a presentation of the results at a
national meeting (may lead to media cover2. The term presentation, as used throughout this chapter and in
the next one. relates to a paper concerning a trial that is prepared
and read by study investigators before a national meeting of some
medical group, but that has not been published (with the exception
of an abstract summary appearing in the meeting program)
Options for initial communication of results
A. Announcement of results to news media with subse
quent publication in a medical journal
• Approach should be avoided, except where the
press release is timed to correspond with publi
cation
• Particularly undesirable when there is a large time
gap between initial publicity and publication
• Members of the medical community may resent
the advance publicity, especially if they are
called upon to respond to questions stimulated
by the publicity without benefit of a published
manuscript
B. Presentation of results at a national meeting with sub
sequent publication in a medical journal
• Presentations are often used for initial communi
cation of results to the medical community
• Presentation may provide authors with useful feed
back for preparation of the final manuscript
• Approach suffers from the same problems noted
in Part A above if presentation leads to news
media coverage
• Generally best to forego presentation unless it can
be timed to correspond with publication
• Approach should be avoided if presentation pre
cludes publication in the journal of choice
C. Publication of results in a medical journal with no
advance presentation or publicity
• Preferred approach, especially for results that are
likely to be controversial or (hat challenge the
value of an existing treatment
• Publication may be accompanied or followed by a
press release
257
troversial and there is a large time gap between
presentation and publication. The five-month
period between presentation and publication of
the tolbutamide results in the UGDP caused
difficulties for patients and diabetologists alike
(see Chapter 7 for chronology of events). Public
ity surrounding the presentation caused many
patients on oral hypoglycemic agents to question
their physicians regarding the usefulness and
safety of the treatment. Physicians had difficulty
dealing with their concerns in the absence of a
published report detailing the results.
24.2.3
Where to publish?
The choice should be limited to refereed journals
that are covered in Index Medicus. Unrefereed
journals, proceedings of meetings, and mono
graphs should be avoided both because of the
absence of a critical review process as a prerequi
site to publication and because of the difficulties
involved in identifying and retrieving any paper
that is not listed in Index Medicus.
The nature of the study will influence the
choice of the journal. Results with general impli
cations should be directed to a wide circulation
journal. A specialty journal should be consid
ered if the results are of primary interest to a
medical subspecialty. Both kinds of journals
may be used in some cases, as in the UGDP with
the phenformin results. The initial report ap
peared in the Journal of the American Medical
Association in 1971. A more extensive report
appeared in Diabetes in 1975 (University Group
Diabetes Program Research Group, 1971b.
1975).
24.2.4
What to publish?
The goal in any publication should be to provide
a clear and concise description of the study re
sults. This requires a manuscript that contains
carefully constructed graphs and tables that de
scribe the results, as well as a description of the
design and methods used in the study. General
content requirements are discussed in Chap
ter 25.
The typical trial may produce only one publi
cation on results. It will come at the end of the
study and should contain results on all treat
ments studied in the trial. The decision as to how
much treatment data to include is not so obvious
if the paper is generated in conjunction with a
protocol change made during the trial. The
paper should satisfy the same content require-
1
•5^
!
258
I
I
24.3 Authorship and internal review procedures 259
Study publication and information policies
ments as a paper published at the end of a trial if
it serves the same purpose as a final publication.
This will be the case in trials involving just one
test and control treatment where one of the two
treatments is discontinued because of lack of
efficacy. The same is true for trials involving
multiple test treatments but where one test treat
ment is considered superior to all others and
hence the use of all other treatments is termi
nated in favor of the superior treatment.
The decision as to what to publish is not so
straightforward when the study involves multi
ple test treatments and when only one of those
treatments is to be discontinued. In this case,
investigators must decide whether to limit results
presented to those for the control treatment and
the particular test treatment in question, or to
include results from all other test treatments as
well. Investigators in the CDP elected to follow
the former approach in each of the three papers
detailing protocol changes in that study (Coro
nary Drug Project Research Group, 1970b,
1972, 1973b). They followed this approach even
when summarizing results leading to discontin
uation of the 5.0 mg dose of estrogen. Results for
a sister treatment, involving just half this dosage,
were not presented in the report even though
they tended to support the decision reached for
the high-dose treatment. In fact, the low-dose
treatment was discontinued about three years
later. UGDP investigators followed the latter
approach. They elected to include results for the
two insulin treatments in manuscripts concern
ing terminations of tolbutamide and phenformin, even though the insulin treatments were
not affected by the terminations (University
Group Diabetes Program Research Group,
l970e, 1971b, 1975).
A complementary decision process is required
when preparing the final results for a publication
in which some of the treatments were discon
tinued before the end of the trial. In this case, the
investigators must decide whether to include an
updated report on the discontinued treatment
groups. The UGDP investigators did include
summaries for both tolbutamide and phenfor
min treatments in their final report (University
Group Diabetes Program Research Group,
1982). The CDP investigators did not provide
such updates in their final report for the three
treatments stopped during the trial (Coronary
Drug Project Research Group, 1975). However,
many of the patients affected by those changes
were enrolled and followed in a sister study (Cor
onary Drug Project Research Group, 1976).
24.2.5 Journal supplements versus
regular issues
Advantages
• Avoids the usual space restrictions imposed on
papers contained in regular journal issues
ing to write a series of small papers than one
large one. Further, many journals have limits on
the length of papers they receive. Editors may be
unwilling to consider papers that exceed those
limits and those who do may assess page charges
to cover the cost of publication. Moreover, they
may place them in supplemental issues of their
journals. An added disadvantage when publica
tion is via a journal supplement is that there may
be problems in locating the issue after it is pub
lished. Journal supplements may not be listed in
Index Medicus and MEDLINE, and even if they
are, they may be hard to find in the library if
they are not bound and stored with regular
issues of the journal.
The main virtue of a single large manuscript
rests in its completeness. It is usually easier for a
reader to grasp the significance of a study if all
pertinent design details and results are contained
in one journal issue than when they are scattered
across various issues of the same journal or
among issues of different journals. The best strat
egy may well be a mix of the two approaches, as
mentioned in Section 24.2.3 in conjunction with
the UGDP phenformin results.
• May provide a more coherent picture of the results
Disadvantages
• Choice of journals limited to those that are willing
to publish supplemental issues
24.3 AUTHORSHIP AND INTERNAL
REVIEW PROCEDURES
Investigators in large trials will have to decide
whether to concentrate their paper writing ef
forts at the end of the trial on a single large
manuscript or on a series of short manuscripts.
The pros and cons of the two approaches are
outlined in Table 24-3.
The trouble with any large manuscript has to
do with the time and effort to prepare it and get
it published. It is easier and often more satisfy
Table 24-3
Long versus short papers
A. Lon,! papers requiring publication as a supplemental
issue of a journal
Comments
• Generally not necessary except for large-scale
trials with complicated data sets
• Usually feasible only if study is prepared to cover
the page charges associated with journal supple
ments
• Manuscript is more difficult and time-consuming
to prepare
• May be harder to locate and retrieve published
papers for reasons mentioned in Section 24.2.5
(see also Chapter 26)
B. Short papers suitable for inclusion in a regular journal
issue
Comment
• Task of preparing a series of short manuscripts less
onerous than that of preparing one long manu
script
Advantages
• Articles appearing in regular issues of a journal
may receive more reader attention and may be
easier to locate and retrieve than those appear
ing in a journal supplement
• Need to generate a number of short papers for
submission to the same or various journals over
an extended time period may help maintain in
vestigator and public interest in the trial
Disadvantages
• Space limitations by journals may make it difficult
to present a coherent picture of the study results
in any one paper
• A series of short papers, scattered over time and
perhaps journals as well, may make it difficult
for readers to obtain a comprehensive view of
the study results
24.3.1
Introduction
A key issue in any research effort has to do with
authorship and writing responsibilities for the
papers produced. Basic guidelines should be
worked out well in advance of the start of any
writing effort. The guidelines that are developed
should be reviewed and discussed by the entire
investigative body before they are adopted. A
good deal of debate may be required before an
acceptable policy is developed.
24.3.2 Individual versus corporate
authorship
The conventional approach is to list contribut
ing authors in the masthead of the paper. All but
2 of the 113 papers reviewed in Chapter 2 had
listings of this type. One paper, citation 113 in
Appendix C, did not list any authors. The other
paper listed a committee as the author (Manage
ment Committee of the Australian Therapeutic
Trial in Mild Hypertension, 1980).
A conventional author listing works best for
studies that are carried out at a single center and
that involve a relatively small number of investi-
gators, it docs not work well for large trials,
especially those involving multiple centers. It is
common in such cases to resort to corporate
authorship, for example, as reflected in citations
102 through 109, from the Coronary Drug Proj
ect and most of the other citations in this chap
ter. However, this method of citation is not with
out problems. It obscures the contribution of
individual authors and may work to the disad
vantage of young investigators seeking promo
tion in university settings (Reiman, 1979; Rem
ington, 1979). Table 24-4 summarizes the pros
and cons of the two authorship approaches.
The list of papers appearing in Table B-3 of
Appendix B can be used to assess the authorship
policies associated with trials sketched in that
appendix. The 130 citations can be classified as
Table 24-4
authorship
Pros and cons of individual versus corporate
A. Convention^ authorship listing
A dvantages
• Commonly accepted form of authorship
• Provides explicit indication of individuals in
volved in manuscript writing effort
Disadvantages
• Can result in lengthy author listing in a large-scale
trial
• Can be an unfair method of dispensing credit,
especially if author listing is limited simply to
those involved in writing the paper
• Increases the likelihood that the study will be iden
tified with specific individuals rather than with
the entire investigative group
B. Corporate authorship
Advantages
• Avoids the interpersonal problems that can arise
when it is necessary to name specific authors for
key study publications
• Avoids the inequities of the conventional ap
proach to authorship when it is not practical or
feasible to list all key study personnel
• Helps underscore the collaborative nature of the
study; especially important for multicenter trials
• Makes it possible to retrieve all papers of a study,
via MEDLINE, under a standard corporate
name, provided that the name appears as part of
the title of each paper (see Section 26.2)
Disadvantages
• Corporate authorship may discourage preparation
of needed papers, especially by people interested
in establishing their research credentials
• Absence of named authors makes it difficult for
readers to identify individuals responsible for its
preparation
1
260
fi. •
if
24.4 Information access policy issues 261
Study publication and information policies
follows: Those listing only a corporate author
(55), those listing only individual authors (55),
with the remainder containing both the study
name and the names of individual authors. The
classification is based on author listing as pro
vided by the individual studies. As such, they
may differ from the citations appearing in Ap
pendix I (Combined Bibliography). Preference
has been given to use of corporate listings in the
body of this book to allow all citations for a
given study to appear together under the same
heading.
The listings in Table B-3 do not necessarily
correspond to those appearing in Index Medicus
or computerized versions of the Index. For ex
ample, the author listing for citation 346 (Com
bined Bibliography), as retrieved via MED
LINE, lists Lachin, Marks, Schoenfield, Tyor,
Bennett, Grundy, Hardison, Shaw, Thistle, and
Vlahcevic as authors. No mention is made of the
NCOS Protocol Committee or the National
Cooperative Gallstone Study Group in the au
thor field. Only Schoenfield and Lachin are
listed as authors for citation 347 in the MED
LINE file. The official study listing in Appendix
B (citation 5.4 in Table B-3) lists 16 other au
thors and the National Cooperative Gallstone
Study Group.
The authorship approach used for specific pa
pers produced in a trial may vary depending on
their relevance to the main aims of the trial.
Most of the studies sketched in Appendix B that
have published papers used a corporate listing
(with or without mention of individual authors)
for mainline papers, i.e., those containing origi
nal treatment results or basic information on the
design and methods of the study. Papers of sec
ondary importance to the trial, for example,
those related to ancillary studies or to secondary
aims of the trials, for the most part, were pub
lished using a conventional author format. How
ever, even here exceptions can be noted, such as
in the Coronary Drug Project. It used a corpo
rate format for nearly all of its publications.
that never get written. The entries on the list
should be ranked in order of their importance in
relation to the aims and needs of the trial. It is
also useful to prepare a timetable indicating
when work might begin for each paper on the
list. The list should be reviewed at periodic inter
vals over the course of the trial to reflect changes
in writing strategies as the trial proceeds.
Most writing efforts will involve a team ap
proach. The team should be designated by the
head of the investigative group (in conjunction
with the leadership committee). Each team
should have a designated chief and should be
composed of members with the expertise and
resolve needed to write the paper. Papers that
involve analyses of study results should include a
biostatistician. The number of papers commis
sioned for development at any one time should
be controlled so as not to exceed the manpower
and computing resources of the study.
The finished paper may or may not list the
writing team. Team members may be included in
the masthead of the paper, as in the NCOS
papers discussed above, in a footnote to the title
of the paper or on the credit page, as in the
Coronary Drug Project paper on design, meth
ods, and baseline results (citation 104), or may
not be revealed at all, as in CDP results publica
tions (citations 102, 103, 105, 107, and 108).
24.3.4
Completed papers should contain a general
credit roster that lists the centers in the study
and their key personnel. The roster should also
specify the membership of committees responsi
ble for operation of the trial. The roster serves
the dual purpose of documenting individual con
tributions to the study while at the same time
providing readers with information concerning
facilities and staff involved in the study. Exem
plary credit rosters are contained in citations
104, 346, and 376.
24.3.5
24.3.3
Writing responsibilities
The head of the investigative group, in conjunc
tion with the leadership committee of the study,
is responsible for stimulating the production of
manuscripts. The first step in the process is to
prepare a list of potential publications early in
the course of the trial. The list should be as
exhaustive as possible and may include papers
Credit rosters
Internal review procedures
The study investigators should subject manu
scripts to rigorous review before they are submit
ted for publication. The review may be carried
out by a standing committee of senior investiga
tors from the study or by an ad hoc group
appointed for the purpose of reviewing a given
manuscript. The CDP used the former approach
and had a standing committee of seven senior
investigators that was responsible for all reviews.
The chairman of the committee selected two or
three persons from the committee to review any
given manuscript.
It may be useful to supplement the review
process by circulating the penultimate draft of a
manuscript to the entire investigative group for
comments. However, this step should not be
used as a substitute for the review processes
mentioned above.
The types of papers requiring internal review
and the conditions that must be satisfied to clear
them for submission to journals should be
spelled out before any writing is done. The re
view and clearance processes will differ depend
ing on the nature of the manuscript. The content
and conclusions of papers containing key find
ings may have to be approved by the entire
investigative group before submission. The in
vestigative group may transfer approval author
ity to the chairman of the editorial review group
for more technical papers dealing with results
related to a secondary aim of the trial or with the
design and methods of the trial. Papers concern
ing ancillary studies that are published under the
names of individual authors may not go through
any formal approval process.
24.4 INFORMATION ACCESS
POLICY ISSUES
24.4.1 Access to study data during the
trial by outside parties
Requests for study results by parties from out
side the study can arise before any results have
been published. The requests may be politely
ignored in privately funded trials but they are
not as easily disposed of in those that are fed
erally funded, especially if the requests are
made under the Freedom of Information Act
(FOIA).3 This Act has been used in a few in
stances to force investigators to release data
against their will. For example, it was used by
the National Enquirer, a weekly tabloid, to ob
tain treatment results from an ongoing trial spon
sored by the Veterans Administration (Montgo
mery, 1979).
There is still a great deal of uncertainty re
garding the limits of the Act as it relates to
) See documents from the Ethics Advisory Bo«rd of the Depart
ment of Health and Human Services (I9K0) for a review of the Act
and of testimony concerning the Act in relation to clinical trials.
ongoing federally funded trials. Proposals to
amend the FOIA to exempt ongoing trials from
requests under the Act have been introduced
into Congress, but have not been enacted (per
sonal communication with Office of the Director
of the National Institutes of Health, July. 1983).
There are obvious dangers in allowing any ex
emptions to the Act. However, unlimited public
access to data before a trial is completed has
dangers as well. The value of a trial can be
compromised with a forced data release. The
ensuing publicity may hamper the enrollment of
additional patients and may make continued
treatment of those already enrolled difficult, if
not impossible. Other dangers may be more farreaching. A pattern of forced releases would al
most certainly render future investigators reluc
tant to undertake long-term trials. The end
result would be even less adequate evaluation of
treatments than that which exists at present.
Court rulings to date have not been of much
help in defining the limits of the Act. Even the
United States Supreme Court ruling concerning
public access to UGDP data was limited to the
specifics of that case (United States Supreme
Court, 1980).
Requests for data or analyses not citing the
FOIA should be considered on an individual
basis. Most of the requests will arise from col
leagues and researchers who are interested in
some aspects of the disease or treatment being
studied. Factors to be addressed in deciding how
to respond to requests include assessments of
the:
• Efforts involved in meeting the requests
• Medical and scientific importance of the data
or analyses requested
• Willingness of the requesters to abide by con
straints imposed by study investigators on
the uses that can be made of the data or
analyses requested
There should be an understanding of the way in
which the data or analyses will be used before
the request is filled. If the requested data or
analyses are to be part of a publication, there
should be an agreement as to how the trial will
be acknowledged and the level of review author
ity retained in the trial over analyses or state
ments produced by the requester. No informa
tion should be released that reveals the identity
of individual patients, nor should the release
permit the requester to carry out analyses by
treatment group if the trial is still under way.
1
262
24.4 Information access policy issues
Study publication and information policies
24.4.2 Access to study data at the
conclusion of the trial
Investigators involved in any trial have a respon
sibility to facilitate access to pertinent study data
once the trial is completed. Part of this responsi
bility can be met with a publication policy that
includes extensive data summaries and patient
listings (devoid of personal identifiers) for key
baseline and follow-up data, such as those pro
vided in appendixes to several of the UGDP
publications (University Group Diabetes Pro
gram Research Group, l970e, 1975, 1982). A
well-written paper will provide readers with suf
ficient detail to allow them to verify the accuracy
of key analyses. Tables and listings that are too
extensive to be published as part of the manu
script can be made available through other
means, such as the National Technical Informa
tion Service (NTIS, sec Glossary).
It is desirable to release the entire data file
(except for patient identifying information) dur
ing the termination stage of the trial or sooner in
trials involving treatment terminations. The
usual approach is to prepare a paper listing or
magnetic tape of pertinent baseline and follow
up data, which is deposited at a central facility,
as was done in the UGDP (University Group
Diabetes Program Research Group, 1977). The
repository may be the sponsoring agency, the
data center, or some other study center (pro
vided it remains in operation after termination
of the trial). The NTIS or some commercial
repository should be used if there is no center in
the study willing or able to assume the reposi
tory role.
24.4.3 Access to study forms and
manuals
Copies of data forms, manuals, and other design
documents used in the trial should be made avail
able to the public once they have been approved
for use in the trial, unless there are convinc
ing arguments to the contrary. The arguments
should indicate how the study would be harmed
if the documents in question were released. Any
release proscription should be lifted as soon as
possible and always by the time the trial is com
pleted. Multicenter trials should designate the
access point for design documents. Publications
from the trial should specify the types of docu
ments that are available and where and how they
may be obtained.
24.4.4
Queries from the press can arise at any point in
the course of the trial. However, they are most
likely at the start and when results are presented
or published. Press coverage can serve a useful
purpose when done in a responsible manner.
Publicity at the start of the trial can help with
patient recruitment. That arising in conjunction
with a presentation or publication of results can
help familiarize physicians and patients alike
with the key findings of the trial.
Requests for information should be handled
as forthrightly and expeditiously as possible. A
good reporter will indicate the purpose of his
request and will provide respondents with the
opportunity to review his copy for errors or
misstatements before it is aired or printed. Re
quests for information having to do with the
design and methods of the trial should be hon
ored regardless of when they arise in the course
of the trial, unless there are sound operational or
scientific reasons for withholding the informa
tion. Requests for interim results or other details
arising during the trial which, if honored, are
likely to compromise patient care or have an
adverse effect on the trial must of necessity be
denied. Most reporters, once given the rationale
for the denial, will appreciate the need for with
holding the information.
There should be only one individual autho
rized to speak for a center in the trial (e.g., the
center director or the public relations officer of
his institution). All queries received by that cen
ter should be referred to that individual for re
sponse.
The chairman of the study, director of the
coordinating center, project officer, or some
other person designated by the investigative
group should be chosen to respond to queries
concerning the study design or results in multi
center trials. The choice should be made early in
the course of the trial and should be made
known to all personnel in the various centers in
the trial. Investigators should agree on the types
of queries that may be answered locally and
those that must be referred for response.
Publicity concerning study results in prepara
tion for presentation or publication should be
avoided. Publicity arising from “leaks” at those
times can serve to divert the energies of study
personnel from the task at hand and may anger
members of the medical and lay community if
they read or hear of results in the news media
before they have been presented or published in
has the potential of shedding additional light on
the results should be pursued.
As a result, the cost of such analyses will have
to be borne by the study, except in cases where
they are done simply to satisfy the needs of the
requesting party. An intangible benefit from de
posit of a data listing or tape with an outside
agency, as discussed in Section 24.4.2, has to do
with outside requests for analyses. Once the de
posit is made, any request can be dealt with by
referring the requester to the repository for the
data needed to perform his own analyses.
A particularly vexing question has to do with
the resources that should be devoted to respond
ing to published critiques of a trial. Some re
straint is necessary because of the investment of
effort required if the criticisms are extensive. In
addition, the energy devoted to response may
limit that available for other more essential ac
tivities. Investigators in the UGDP were con
cerned enough
enough about
about energy
energy mss.puuun
dissipation that
cerned
u
which
they limited their responses to criticisms whi
appeared in
They declined
...refereed
_____ * journals.
to
reply to editorials and critiques appearing
^ppearing^ in
unrefereed publications such as the Medical D-ibnal of Medicine, while having policies against
une.
such practices, offer subscription packages that
There is another reason for restraint. Investi
include provisions for express delivery of jour
nals after they have been printed. Investigators gators run the risk of losing objectivity and dam
aging their credibility if they become too preoc
should assume that major newspapers and wire
cupied with defending their own work. The
services will have such subscriptions and that
T-r a sound investigation is at
they will have their copies several days before philosophy neeacdjfor
odds
with
that
needed
for advocacy of a posi
they appear on the desks of regular subscribers.
tion. For this reason, if none other, it is impor
Further, it is wise to assume that the program
and published abstracts of papers to be pres tant that responses be thoughtful and devoid of
ented at national meetings will be available to emotion.
members of the press before the actual meeting
date. The advance publicity concerning the
24.4.6 Outside audits
UGDP tolbutamide results arose from distribu
It may be necessary to provide for special audits
tion of the program of the American Diabetes
Association several weeks before the actual meet of the study results if there are questions regard
ing their accuracy. The data records and anal
ing date (see Section 7.4 for details).
yses of the UGDP were subjected to two inde
pendent audits (Committee for the Assessment
of Biometric Aspects of Controlled Trials of Hypoglvcemic Agents, 1975; Food and Drug Ad
24.4.5 Special analyses in response
ministration, 1978). Special clearances will be
to criticisms
required if auditors are to be provided access to
the medical records of specific patients. To be of
Once the results have been published, study in
any value, the audits should be done by parties
vestigators may be urged to carry out a number
who are independent of the study, the sponsor
of special analyses by friends and foes of the
ing agency, and firms or groups that stand to
study. They will have to decide how much time
gain or lose from the study results. The auditors
and effort they wish to devote to such activities.
should prepare a written report of the audit that
The approach taken will depend on the rele
is then published or placed on file for public
vance of the requests in relation to the aims and
access.
needs of the trial. Certainly, any analysis that
a scientific forum. Investigators need to avoid
actions that may attract unwanted attention.
Members of the study group, particularly those
privy to interim results, such as members of the
safety monitoring committee, need to be re
minded of the importance of silence until results
have been presented or published. This policy of
restraint should be coupled with defensive mea
sures that can be implemented if leaks occur.
The measures may include preparation of a state
ment concerning the study results that can be
released to the press if publicity occurs prior to
the scheduled publication or presentation. Inves
tigators in the Coronary Drug Project took this
precautionary step with each of their mainline
results papers (Coronary Drug Project Research
Group, 1970b, 1972, 1973b, 1975). Fortunately
none of the statements were needed.
Investigators should be mindful of the prac
tices of the organizers of meetings and journals
hat may lead to unexpected press coverage.
Some publishers provide members of the press
with copies of selected papers in advance of publicationP Others, such as the New England Jour-
Inquiries from the press
!
I
263
1
25.3 Content suggestions 265
25. Preparation of the study publication
• Event recording in a clinical trial of a new
medicine (89). What kind of event? What
kind of new medicine7
• Clinical metrology A future career grade?
(122). Metrology in what area? What is
meant by a career grade?
• Evaluation of toxicity: Clinical issues (175).
What kind of toxicity? What clinical
issues?
Revise and revise and revise- the best thought will come after the printer has snatched away
the copy.
Michael Monahan
Needlessly detailed
25.1 Introduction
25.2 Preparatory steps
25.3 Content suggestions
25.3.1 Title section
25.3.2 Abstract section
25.3.3 Introductory section
25.3.4 Methods section
25.3.5 Results section
25.3.6 Discussion section
25.3.7 Conclusion section
25.3.8 Reference section
25.3.9 Appendix section
25.4 Internal review and submission
25.5 Acceptance and publication
25.3
Table 25-1 Content suggestions for the study
publication
25.1
CONTENT SUGGESTIONS
Table 25-1 outlines general content suggestions
for the publication. The remainder of this sec
tion relates to this outline.
25.3.1
INTRODUCTION
This chapter focuses on the task of preparing the
study findings for publication. The outline in
Table 25-1 assumes a single publication that con
tains a summary of the main findings of the trial,
as well as information on its design and opera
tion. In fact, as noted in Chapter 24, a trial may
produce a number of publications.
25.2
• High-dose methotrexate with “RESCUE”
plus cyclophosphamide as initial chemo
therapy in ovarian adenocarcinoma. A ran
domized trial with observations on the in
fluence of C parvum immunotherapy (7).
Title could be shortened without much loss
of information by deleting the phrase with
observations on the influence of C parvum
immunotherapy.
• A clinical trial of alignment of teeth using a
0.019 inch thermal nitinol wire with a tran
sition temperature range between 31 de
grees C. and 45 degrees C. (115). Too much
methodological detail.
• Medical, ethical and legal aspects of clinical
trials in pediatrics. Summary of a forum
discussion held at the ‘International Work
shop on Perinatal and Pediatric Aspects of
Clinical Pharmacology,’ Heidelberg, Fed
eral Republic of Germany. February 27-29,
1980 (143). Details regarding the meeting
are not necessary.
investigators as to who in the study will have
review authority over the paper and how con
flicts between the authors and the review group
will be resolved (see Section 24.3.5).
An essential step involves preparation of an
outline of the paper. The outline should be
as detailed as possible and should include a
mockup of tables needed for the paper. It should
be reviewed and approved by the leadership of
the study before the writing starts and should be
revised as needed during the writing effort.
PREPARATORY STEPS
Most of the preparatory steps needed to write
the paper have been alluded to in previous chap
ters. An essential first step involves preparation
of the data for analysis by creation of an analysis
tape, as discussed in Section 17.7.
There must be agreement among investiga
tors as to how the paper will be authored and as
to who will head the writing team (see Sec
tions 24.3.2 and 24.3.3). The review steps that
must be completed before the paper is submitted
for publication should be delineated as well. In
addition, there should be agreement among the
Title section
The title is one of the most important parts of
any publication. It is the prime item used by
readers to screen for publications of interest. A
good title is neither cute nor cryptic. It conveys
its message in a crisp and succinct manner. It
should indicate the main thrust of the paper in as
few words as possible. Superfluous words add to
its length without making any contribution to
content.
Examples of good and bad titles follow, as
taken from citations listed in Appendix C. The
number in parentheses following each title refers
to the citation number in that Appendix.
Cryptic
Good
• Controlled trial of cimetidine in reflux esophagitis (27).
• Amoxycillin versus ampicillin in treatment of
exacerbations of chronic bronchitis (53).
• Cimetidine in the prophylaxis of migraine
i
I
(69).
• Postoperative epilepsy: A double-blind trial
of phenytoin after craniotomy (71).
,
• Blood pressure in the elderly (16). What kind
of blood pressure? High blood pressure?
Low blood pressure? There is no way of
knowingfrom the title if the paper contains
data pertinent to the assessment of differ
ent forms of antihypertensive treatments in
the elderly.
264
I
Titles should be written to include the term
trial to facilitate identification via title scans.
One way this can be accomplished is to include
the name of the study in the title if it includes
the term. Unfortunately, investigators often use
other less descriptive terms, such as study, pro
gram. or project, in place of trial (see Table B-6
of Appendix B for list of study names). The
term trial should be added to the title of the
paper when it is not part of the official title of
the study.
A titling convention in which papers are se
quentially numbered, for example, as in the Uni
versity Group Diabetes Program (UGDP), is
worth considering, especially if it is clear from
the outset that the study will generate a number
of publications. The numbering scheme alerts
readers to the existence of other papers in the
series.
Some journals require authors to list a few
key words that characterize the content of the
paper. The words usually appear below the title
or abstract of the paper. Key words serve two
purposes: They indicate the thrust of the paper
to readers, and they help indexers classify it
under the proper subject headings in Index Med
icus and MEDLINE.
25.3.2
Abstrict section
The abstract of the paper is second only to the
title in importance. It provides a summary of the
paper and, as a result, is usually the first and
often the only part that is read, other than the
title. In addition, inclusion of the abstract in the
MEDLINE data file (the computerized version
of Index Medicus) makes it possible for users of
that file to identify the paper by searching the
abstract for terms or phrases of interest. The
abstract should include the items of information,
listed in Part 2 of Table 25-1. A sample abstract,
taken from the Persantine Aspirin Reinfarction
Study (PARIS), meets most of the content re
quirements listed (Persantine Aspirin Reinfarc
tion Study Research Group, 1980b).
Summary. In the Persantine-Aspirin Rein
farction Study (PA RIS) trial. 2026 persons
who had recoveredfrom myocardial infarc
tion (Ml) were randomized into three
groups: Persantine plus aspirin (PR) A)
(n = 810); aspirin alone (ASA) (n = 810):
placebo (PLBO) (n = 406). The average
length of follow-up study was 41 months.
Results for the three specified primary end
points were: total mortality 16% lower in
PR!A and 18% lower in ASA compared
with PLBO: coronary mortality 24% and
21% lower; incidence of nonfatal Ml plus
fatal coronary disease 25% and 24% lower.
These differences were not statistically sig
nificant by the study criterion (Z > 2.6). By
life-table analysis, the rates of coronary mor
tality and coronary incidence were about
50% lower in the PR/A group than in the
PLBO group from 8-24 months, and for
1
k
25.3 Content suggestions
266
267
Preparation of the study publication
Table 25-1
Content suggestions for the study publication
I. Title Met Ion
• Descriptive title
• List of author-selected key words indicating general
content of paper (useful for readers and as an aid to
NLM indexers)
• Authorfs)
• Source(s) of financial support for the study
• Acknowledgments
• Credit roster (see Section 24.3.4)
• Address for reprint requests
2. Abstract section
• Purpose of study
• Primary outcome measure
• Test treatment(s)
• Control treatmentfs)
• Level of treatment masking
• Number of patients enrolled
• Method of treatment allocation
• Conclusion(s)
3. Introduction section
• Historical background of trial
• Rationale for a trial
• Objective(s)
• Rationale for choice of test and control treatmentfs)
• Literature review
4. Methods section
• Study population
Eligibility and exclusion criteria
Method of patient recruitment
• Treatments
Study treatments used
Method of treatment administration
Level of treatment masking
Treatment proscriptions
Methods of measuring treatment adherence
• Outcome measures
Primary and secondary outcome measures
Diagnostic criteria for outcome measurements
Methods for coding and classifying outcomes
• Design specifications
Method of randomization
Description of the safeguards used to ensure
the integrity of the allocation process
List of stratification variables
Blocking specifications
Table 25-1
Description of procedures for packaging and
dispensing study medications in the case of
masked drug trials
Primary outcome measure and rationale for choice
Planned length of patient follow-up and rationale
for specification
Planned recruitment goal
Type I and II error protection level for planned
recruitment goal
• Patient safeguards
Outline of steps for obtaining patient consent
Method of updating consent (especially for long
term follow-up trials)
Measures taken to protect patient confidentiality
Description of procedures used to monitor study
results for evidence of treatment effects
• Data collection schedule
Sequence of baseline and follow-up visits
List of data items collected
Definition of missed visits and dropouts
Name of person or agency to contact for copies of
data forms, study manuals, etc.
• Data processing
Cut-off date for data included in manuscript
Description of approach and supporting rationale
for dealing with missing data and departures
from the treatment protocol (statement especially
important if analysis method departs from pre
ferred approach described in Chapter 18)
Literature references for methods used
Description of any special analysis procedures not
already described in existing literature
Methods for judging statistical importance of dif
ferences observed (e.g., simple p-values, adjusted
p-values, RBOs. etc.)
• Quality control procedures
General data editing
Quality control of laboratory tests and for special
reading and coding procedures
Checks on data entry, programming, and analysis
Other quality controls, such as site visits to clinics,
training and certification, etc.
• Performance monitoring
Measures used for assessing performance of partici
pating clinics and resource centers
Frequency of performance assessments
Methods used for reviewing performance monitor
ing reports and for implementing corrective ac
tion based on those reviews
• Treatment monitoring
Frequency of interim analyses for treatment moni
toring
Methods used to carry out interim analyses
Content suggestions (or the Mudv publication (coniinucd)
• Treatment comparisons by selected baseline charac
teristics
• Multiple regression analyses using baseline character
istics to provide adjusted treatment comparisons
Individual or group responsible for carrying out
interim analyses
Procedures for implementing protocol changes
based on results from interim analyses
• Organizational structure
Number and location of participating centers
Location of data center
Location of other resource centers
• Treatment comparisons by level of adherence
• Treatment comparisons by clinic (in multicenter
trials)
• Other special analyses relating follow-up data for one
variable (e g., cholesterol level) to a primary or
secondary outcome measure (e g., death)
or consortium award)
Policy on investigator conflicts of interests and
method used to monitor for potential conflicts o
6.
interest
* ^NotS’and language conventions in man'KC"P‘
Listing of special actions taken during the trial in
• Discussion of the implications of the findings
cluding:
Addition or deletion of a treatment
Data purges because of questions concerning
data reliability or accuracy
Major modifications of data collection forms
or coding procedures during the course of
• Enumeration of questions or areas needing further
analysis or research
7. Conclusion section
• Statement of conclusion
• Limits on generalization of the conclusions, including
discussion of observed statistical power if no treat
ment difference is detected
the trial
5. Results section
• Number of patients enrolled by treatment group
• Number of deaths by treatment group
• Comparison of treatment groups for the primary and
secondary outcome measures using various analytic
techniques, including simple comparisons of pro
portions, as well as lifetable methods, etc.
• Indicators of the completeness of follow-up by treat
ment group, such as:
Number of missed examinations
Number of dropouts
Number of patients lost to follow-up
• Indicators of treatment adherence, such as.
Comparison of treatment groups using an adher
ence score or some laboratory test
Count ofT number of patients^ in each treatment
who received
group Vi.v
.— none
----- of the assigned treat
ment
Count of number of patients in each treatment
group who received an alternative treatment
• Assessment of the comparability of the treatment
groups with regard to important baseline character
istics
• Treatment group comparisons for differences tn:
Occurrence of serious side effects
Rate of hospitalization
Other general health indicators
„„u.n r„en„al
Discussion section
• Discussion of how reported findings relate to previous
studies, paying particular attention to those consid
ered to be new and those that are not consistent
with findings of previous studies
g. Reference section
• List of literature references in required journal tormat
• Suitable reference citations for.
References to previous work
Data analysis methods
Methods not described in the paper
Laboratory methods
.
Coding or reading procedures for abstracting infor
mation from special records or documents
Treatment methods
Study rationale
Discussion of results
• List of studv documents that may be obtained on
request, such as study manual of operations, study
data forms, data listings, data tapes, etc.
4. Appendix section*
• Descriptions of special procedures needed to und"stand results, but too detailed to be included in the
body of the publication
• List of definitions, codes, diagnostic criteria, etc.
• Special analyses, tabulations, and data listings
• Sample data forms
o, i( Ml«» l»ve raided
o.He. me.n, « wp!yinr
•Not required if previous publications
pubhc repositors or hv supplymg them upon written request!.
depositing documents containing details in a
« « • >■'
1
268
coronary incidence all Z values were > 2.6:
ASA rates were about JOT'c lower than
PLBO rates, and for coronary incidence. Z
values were >2.6 at two points. For these
end points, from 8-20 months. PR/ A rates
were about 30% lower than ASA rates, but
all Z values were <2.0. PR/A and ASA
patients entering within 6 months of last Ml
showed the largest percentage reductions in
mortality; only the difference between PR/
A and PLBO groups for 3-year coronary
mortality yielded a Z value of 2.63
25.3.3
Introductory section
This section may be short or quite long depend
ing on the nature of the literature review. Its
prime purpose is to set the stage for the remain
der of the paper. It should indicate why the
paper is being written, describe the rationale for
the study and its objectives, and should recap
research that led to initiation of the study.
25.3.4
I
Methods section
The details contained in most reports of clinical
trials are too sketchy to allow readers to make
informed judgments concerning their quality, as
noted in Chapter 2 and in Meinert et al. (1984).
The absence of essential details is a reflection of
the failure of authors and editors alike to recog
nize their importance in making these judg
ments.
The contents of the methods section must be
checked against some predefined list, such as
contained in Part 4 of Table 25 I, if reporting
lapses are to be avoided. Information found to
be missing when the check is made should be
added before the paper is submitted for publica
tion. Details that have to be omitted because of
space constraints imposed by the journal should
be provided via other means (e g., in another
paper devoted primarily to the design and meth
ods of the trial, or by depositing essential design
and operating documents from the trial in some
repository for access by interested parties).
25.3.5
Results section
This section is usually the longest one in the
paper. Suggestions for its content are outlined in
Part 5 of Table 25-1. The essence of a paper
I Reference citation 376 Reproduced with permission of (he
American Heart Association. Inc , Dallas. Texas.
A,
25.4 Internal review and submission 269
Preparation of the study publication
information contained in the
or expand upon
original article.
should provide all the authors’
The listing
’■
names in the case of conventionally authored
materials. The preferred approach is to list the
last name, followed by the initials of each au
thor. The author field should contain the ap
propriate corporate designation if the article in
question was written on behalf of some research
group, institute, agency, or committee. See refer
ence citations 104, 375, 376, 467, 468. 472 and
476 in the Combined Bibliography (Appendix I)
for examples.
The citations should include the full titles of
the articles being cited. They are useful to read
ers when scanning the references for articles of
interest. They should also include full journal
names or accepted abbreviations, such as those
used in Index Medicus and MEDLINE (Na
tional Library of Medicine, 1983). The volume
number of the journal, date of publication, and
beginning and ending page numbers of each arti
cle cited should be listed as well.
The citation listing should be checked for ac
curacy before the manuscript is submitted for
publication. The checking should be done from
the actual articles cited and not from MED
LINE printouts or citations listed in other bibli
ographies.
should be captured in the tables, charts, and
figures it contains. They should be interpretable
without reference to supporting text in the body
of the paper. The titles and legends accompany
ing them should be accurately and succinctly
written.
25.3.6
Discussion section
This section (see Part 6 of Table 25-1) should
highlight noteworthy findings appearing in the
results section. It should be used to discuss the
clinical implications of the findings and to indi
cate the extent to which they are considered lo
support or refute previous findings.
25.3.7
Conclusion section
This section (see Part 7 of Table 25-1) may
appear at the beginning or end of the paper. It
should contain a statement of the conclusions
drawn from the trial and of the limitations on
the generalizability of the findings. It should also
contain a discussion of the statistical power of
the study if the conclusion favors the null hy
pothesis (see Section 9.7).
25.3.8
Reference section
A well-written paper will contain a supporting
bibliography. The papers included should be
those that are needed to document data collec
tion and analysis methods used in the trial, as
well as those needed in the introduction and
discussion sections of the paper. The citations
should be listed at the end of the paper and
should either be arranged alphabetically or in
the order in which they are referenced in the
text. Most journals will indicate the referencing
style to be followed. A paper that is written
before a decision is reached on the journal that is
to receive it should be referenced using general
methods, such as outlined by editors of medical
journals (International Committee of Medical
Journal Editors, 1982) or in a general desk refer
ence, such as the Chicago Manual of Style (Uni
versity of Chicago Press, 1982).
Only papers cited in the text should be listed
in the reference section of the paper. Original
articles should be referenced whenever possible.
A secondary source, such as a textbook or re
view article, may be cited if the original article
appeared in an obscure or foreign language jour
nal, or if the secondary source helps to explain
!
i
such as discussed in Chapter 24 involving sup
plemental issues of a journal or deposit of key
documents at a public repository, will have to be
used when appendixes are ruled out by page
limitations or other policies imposed by the jour
nal.
25.4 INTERNAL REVIEW AND
SUBMISSION
The manuscript should be subjected to a series
of reviews and checks before it is submitted for
publication. The first review should be done by
the authors and should be designed to check for
inconsistencies in format or style, for redundant
statements, and for reporting deficiencies. The
later review should be made using a checklist,
such as represented in Table 25-1. The titles and
legends of tables, charts, and figures should be
checked for clarity of exposition and accuracy.
The numerical information presented in tables
and graphs should be checked for errors. The
text of the paper should be checked to make
certain that figures cited agree with the numbers
appearing in the tables. Key analyses should be
repeated, ideally by a second person, and the
results of the two analyses should be compared.
All discrepancies should be resolved before the
paper is submitted for publication.
Information taken from published literature
should be checked against the cited source. This
checking process will be simplified if all cited
25.3.9 Appendix section
documents are collected as the manuscript is
This section should contain materials that, while
developed. The resulting file will serve as a valu
important in understanding the paper, are too
able resource for future papers on the same sub
technical or detailed to warrant inclusion m the
ject and for checking reference listings and other
main body of the paper. Items that appear in the
information contained in the manuscript.
appendixes of publications (see 104, 375, 467,
The second round of reviews should be by
468, 472, and 476 cited above) include:
colleagues selected by the authors or the study
leadership (see Section 24.3.5). These reviews
• Details of the sample size calculations
will help to identify areas of the paper that are
• Baseline frequency distributions
confusing and that need additional work. Major
• Sample data forms
changes proposed during this round of reviews
• Data collection schedules
or any of the other reviews outlined above may
• Derivation of analytic procedures
require a total revision of the paper and another
• Special charts or figures
round of checks and reviews.
• Data listings
It is a good idea to allow some time for “matu
• Special analyses or tabulations
• Descriptions of coding and classification
ration" of the paper after it is drafted and before
it is submitted for publication. The checking and
schemes
review processes take time. They will lose much
• Consent statements
of their value if performed under duress because
• Organizational and administrative docuof the imposition of unrealistic deadlines.
ments
The final draft of the paper should be checked
The use of appendixes is possible only if the
to make certain that the format conforms to that
journal in question allows them. Other avenues.
A
270
Preparation of the study publication
specified by the journal selected to receive it. The
instructions supplied by the journal should be
reviewed to make certain that the correct
number of copies is submitted and that glossy
prints of all figures and charts are provided. The
paper should contain the address and phone
number of the corresponding (usually senior)
author. A copy of the paper and accompanying
glossy prints (if any) should be retained by the
corresponding author. The cover page of the
manuscript should indicate the date the paper
was submitted for publication. All previous
drafts of the paper should be removed from the
author’s file and stored elsewhere once it has
been mailed to the journal to avoid mixups if the
journal, a reviewer, or someone else requests
copies of the manuscript before it is published.
25.5 ACCEPTANCE AND
PUBLICATION
The journal will carry out its own reviews of the
manuscript. They will be used by the editor to
reach a decision as to whether to accept the
paper. They may also serve as the basis for addi
tional changes to the paper if it is accepted for
publication. Publication may take place shortly
after acceptance or months later, depending on
the backlog of manuscripts awaiting publication
and the publication schedule of the journal.
The corresponding author is responsible for
ordering reprints. The number ordered should
be sufficient to supply co-authors with an ap
propriate number, as well as all other people
listed in the credit roster of the paper.
26. Locating and reading published reports
The corresponding author (or one of the other
authors) should establish an archive that con
tains all documents related to the development
and publication of the paper. The initial steps
for this process should be taken long before
publication. The last steps in the process should
take place just after the paper has been pub
lished. The completed file should contain:
Be sparing of criticism, since the habit of trivial comment weakens the force of
• Copies of data tapes and computer programs
used for analyses included in the paper
• Copies of papers and other documents refer
enced in the paper
• Intermediate drafts of the paper, particularly
those containing major revisions
• A copy of the manuscript submitted for pub
26.1 Introduction
26.2 Bibliography development
26.3 Questions and factors to consider when
reading a report from a clinical trial
26 4 Valid and invalid criticisms
26.5 Desirable characteristics of a critic
Table 26-1 Selected printed and computerized
databases of published literature
and work in progress
Table 26-2 Questions to consider when assess
ing a published report
Table 26-3 Universal criticisms
Table 26-4 Characteristics of a responsible critic
lication
• Copies of written critiques of the paper, as
provided by the journal, and correspond
ence relating to the critiques
• A copy of the manuscript as accepted for
publication
• Page proofs
• The published manuscript
(used to identify authors working in a particu
lar field as an aid to building a bibliography
of papers related to that field)
• Review of bibliographies of published papers for
citations of interest
• Pursuit of leads offered by colleagues or from
other sources, such as the news media, regard
ing specific papers or pieces of work
Table 26-1 contains a list of databases of pub
lished reports and work in progress (see also Roper
and Boorkman, 1980; Sciotti et al., 1982). The list
represents a selection of existing files considered by
the authors to be useful in constructing bibliogra
phies related to the design, conduct, and results of
clinical trials.
Locating reports of clinical trials is complicated
26.1 INTRODUCTION
by the way in which they are titled (as discussed in
This chapter deals with a potpourri of topics related
Section 1.3 and in Section 25.3.1) and because of
to the identification and evaluation of reports rele
the absence of a subject heading for clinical trials in
vant to the design and conduct of clinical trials.
most existing indexes. A notable exception is Index
Section 26.2 focuses on a review of methods for
Medicus and MEDLINE, starting with 1980 (see
developing bibliographies of results from clinical
Chapter 2). The usefulness of title searches for iden
trials. Section 26.3 is concerned with issues to be
tification of trials is limited without such headings.
considered when reviewing a published report from
Only 21 of the 130 references (16%) in Table B-3,
a trial. The last two sections are written from the
Appendix B. had the term trial in the title. De
point of view of an individual who is responsible for
signers of trials appear to prefer terms such as
preparing a written critique of a report from a
study, program, or project. Only two of the 14 trials
clinical trial. Section 26.4 provides a discussion of
listed in Appendix B had names containing the
what constitutes valid criticisms. The last section
term trial (see Table B-l, Appendix B). Among the
outlines the characteristics of a responsible critic.
113 trials reviewed in Chapter 2, less than 40% (44
out of 113) of the titles contained the term trial or
the design words blind, randomized, or controlled.
26.2 BIBLIOGRAPHY DEVELOPMENT
The difficulty in identifying work in progress
The development of a bibliography is likely to
extends to methodological work as well. The best
include any of the following techniques:
that can be done at present is to rely on special
annotated bibliographies, such as the one produced
• Review of selected journals for papers of interest
by Fletcher and co-workers (Research Develop
• Search of classes of journals, via Current Con
ment Committee. Society for Research and Educa
tents or some other means, for titles of in
tion in Primary Care Internal Medicine, 1983) con
terest
cerning clinical research methods, and by Hawkins
• Systematic search of papers or computerized
(see citations 227 through 230), in her periodic
indexes, such as contained in Index Medicus
reviews of literature related to clinical trials.
or MEDLINE
Papers concerned with statistical issues in the
• Use of the Science Citation Index to identify
design, conduct, or analysis of clinical trials must be
authors who have cited a particular paper
The archive should be kept in a safe place and
maintained indefinitely. Key documents, such as
data tapes and related materials, should be du
plicated and stored in separate locations if they
are considered irreplaceable.
Errors in the paper detected after publication
should be noted by the corresponding author.
The journal editor should be informed of those
that are serious.
i.
I
271
1
272
Locating and reading published reports
Table 26-1
263 Questions and factors to consider when reading a report from a clinical trial 273
Selected printed and computerized databases of published literature and work in progress
Database
Comments
Table 26-1
Primed and computerized databases of published literature and work in progress I continued)
Comments
Database
A. Published literature
• Index Medicus
• MEDLINE (Medical Literature
Analysis Retrieval System: MED
LARS on Line)
• SCI (Science Citation Index)
• BI OS IS (Biological A hstracts)
• CATLINE (Catalog on Line)
• CLINPROT (Clinical Protocols)
• Listings of titles, authors, and abstracts of papers appearing
in some 2,700 medical journals and periodicals. Publica
tion started in 1879. Published under the title Index Me
dicus beginning in I960. Entries indexed by author and
subject. Has subject heading for clinical trials starting in
1980. Before 1980, articles on clinical trials appeared
under the more general heading clinical research.
• Computer file of Index Medicus. International Nursing
Index, Index to Dental Literature, and part of Hospital
Index. File contains titles, authors, and abstracts of pa
pers appearing in some 3.000 biomedical journals. File
may be searched by author or subject. Titles and abstracts
can be searched with user-selected words. Introduced in
1966; with abstracts since 1975. Contains subject heading
for clinical trials starting in 1980. Before 1980; articles on
clinical trials appeared under the heading clinical re
search.
9 Exists both as a paper and computer file (SCISEARCH).
The computer file contains all entries published in the
Science Citation Index plus additional entries from the
Current Contents series of publications. SCI is unique in
that it identifies papers cited in articles appearing in some
2,600 journals and periodicals. The Index allows users to
identify articles that reference a particular paper. May be
searched by author or title words. Started in 1961. Pub
lished on a continuing basis since 1964; computerized
version since 1970.
• Exists both as a paper and computer file. Includes publica
tions from journals, books, symposiums, reviews, notes,
and research communications from the life sciences. Does
not have subject headings, only broad headings called
concept headings. First publication of printed version of
the file: 1926; computerized version of file introduced in
1969. Contains citations for some statistical literature.
• Computer equivalent of National Library of Medicine Cur
rent Catalog. Includes listing of all serials, monographs,
and books (all languages), collected by the National Li
brary of Medicine, and published after 1801. The Catalog
was first published in 1966. It has a subject heading for
clinical trials beginning in 1980.
• Computer-based data file containing summaries of clinical
investigations of new anticancer agents and treatments,
identified by screening statistical journals or me
thods journals, such as Controlled Clinical Thais.
There are a number of indexes that contain cit
ations to statistical and methods papers pertinent to
clinical trials (e.g.. Biological Abstracts, Current
Contents, Psychological Abstracts. Chemical Ab
stracts, Public Affairs Information Service (PAIS),
Mathematical Reviews, and Excerpta Medico), but
they are not identified as such in the indexes.
263 QUESTIONS AND FACTORS TO
CONSIDER WHEN READING A
REPORT FROM A CLINICAL TRIAL
Table 26-2 lists questions to be considered when
reading a published report. The greater the number
of affirmative answers the better the reporting pro
cess.
The reader should form his own judgment on the
• CANCF.RLIT (Cancer Literature)
• EMED (Excerpta Medico)
• MATH FILE
views)
(Mathematical
Re
________
with emphasis on clinical trials. File may be searched
using an index of 300 clinical terms or via user-selected
words.
• Computer-based data file containing over 260.000 citations
and abstracts of published literature relating to cancer.
Created originally from Cancer Therapy Abstracts
(started in 1967) and Carcinogenesis Abstracts (started in
1963). Both ceased publication in 1980 Titles or abstracts
may be searched via user-selected words. Entries since
1980 have been indexed using NLM subject headings,
including one for clinical trials.
• Computer-based data file containing citations from over
3,500 biomedical journals. File consists of entries from 43
abstract journals and the two literature indexes that make
up the printed Excerpta Medico, plus selected entries not
appearing in the printed publications. Contains citations
from June 1974 forward. Has subject headings for clinical
trials and controlled clinical trials.
• Computer-based data file of references to mathematical and
statistical papers.
B. Unpublished work in progress
• CRISP (Computer Retrieval of In
formation on Scientific Projects)
• RAI (Research Award Index)
• NTIS (National Technical Informa
tion Service)
• Computer-based data file of information on research pro
jects currently funded via the NIH and other agencies of
the United States Public Health Service. File may be
searched by subject, project, agency of support, or inves
tigator. Introduced in 1971 Does not contain heading for
clinical trials.
• Paper listing of research grants and contracts awarded by
the National Institutes of Health, by fiscal year. Produced
from CRISP. Published in two volumes: Volume I is
arranged by research subject; Volume II contains sections
organized by project, by grant or contract number, and
by investigator. Produced since 1962. Does not contain a
subject heading for clinical trials.
• Both a paper-based and computerized data file of over
970.000 documents available through the NTIS. Docu
ments stored at the NTIS are government-sponsored re
search reports prepared by federal agencies of the United
States government or their grantees or contractors. NTIS
has been in operation since 1964. The computer data file
covers acquisitions at NTIS from 1975 forward.
basis of the merits of the study before considering
opinions and critiques of others. Reviews supplied
gratis by sales people from firms with a proprietary
interest in the treatments should be ignored when
making the judgment. The same is true for com
mentaries and editorials on the study appearing in
throwaway medical journals.
The reader should be conscious of the motivating
forces behind the study and of their possible influ-
ence on the conduct of and reports from the study.
The conclusions in the paper should be questioned,
if not ignored, if they appear to have been written
to support the preconceived notions of the sponsor
or investigators regarding the merits of the study
treatments.
The role of the sponsor in the trial should be
considered when reading the paper. Published re
ports of trials that are carried out by firms produc-
i
274
26.3 Questions and factors to consider when reading a report from a clinical trial
Locating and reading published reports
Table 26-2
Questions to consider when assessing a published report
I. General
• Does the manuscript indicate the purpose of the
trial and rationale for the treatments studied?
• Does the trial address a relevant question?
• Is the paper in a peer review journal-’
2. Investigators
• Have the investigators done any previous work re
lated to the trial being reported? If so, do you
consider the work to have been of good quality?
• Does the paper indicate the location and institu
tional affiliation of the various members of the
team responsible for carrying out the trial?
• Does the team include people with appropriate
training and expertise for conduct and analysis of
the trial?
• If multiple outcomes are used, is it clear which one
is of primary importance in the trial?
h. Treatments
• Is there a defined test treatment?
• Is the test treatment of any interest and does the
administration of it correspond roughly to the
way it would be used in general practice?
• Is there an appropriate control treatment?
c. Study population and sample size
• Are the eligibility and exclusion criteria for patient
entry into the trial stated?
• Is there a discussion of the type 1 and II error
protection provided with the observed sample
size?
d. Allocation
• Is the method of treatment allocation described?
• Does it appear to have been free of selection bias?
3. Spomonhip and structural
• Does the paper indicate how the trial was funded?
• Is the role of the sponsor in designing, directing, or
analyzing the trial indicated9 (Especially impor
tant in trials involving proprietary products.)
• Are the key investigators, especially those responsi
ble for analyzing the results and for writing the
paper, independent of the sponsor?
• Did responsibility for data collection and analysis
in the trial reside with a group of people who
were independent of the sponsor?
• Did the authors recognize the possibility of con
flicts of interest for study members (especially
important if the report concerns a proprietary
product) and do they indicate steps taken to
avoid such conflicts?
• If the trial involved multiple centers, does the paper
list all affiliated centers and the functions per
formed by each9
• For multicenter trials, does the paper list commit
tees, along with their membership and a brief
description of their functions?
4. Study design
a. Outcome measure
• Is the primary outcome measure identified'’
• Does it have clinical relevance?
ing the products being tested, or that fail to indicate
the source of the funding, should be viewed with a
healthy skepticism. The same is true for reports
produced by private foundations that derive their
funds from unnamed sources or from sources with
interests that stand to gain or lose financially, de
pending on the conclusions stated in the reports.
For example, the Kilo, Miller, and Williamson
critique (19X0) of the University Group Diabetes
It
• Does it meet the general conditions specified in
Section 8.4?
e. Data collection procedures
• Is the data collection schedule described?
• Are patients in the test and control-treated groups
enrolled and followed over the same time frame9
• Does the design include adequate provisions to pro
tect against bias in the administration of the treat
ment and in measurement of the outcome, as
evidenced by the use of appropriate masking
procedures or other safeguards?
5. Study performance
• Was a recruitment goal for the trial stated? Was it
achieved?
• Was the missed examination rate low?
• Was the dropout rate low?
• Was the dropout rate among the treatment groups
about the same?
• Was it possible to locate all patients, including drop
outs, at the end of the study to update key mor
bidity and mortality data'* If not, was the number
who could not be located small and about the
same for each treatment group?
• Did all the patients enrolled meet the eligibility
criteria of the (rial? If not, was the number who
did not small?
Program (UGDP) tolbutamide results, simply
states that the analyses were supported by the Kilo
Research Foundation. Evaluations of the authors
objectivity may very well be influenced by the ex
tent to which support for the Foundation depends
on money supplied by manufacturers of the oral
hypoglycemic agents.
The reproducibility of the results and the gener
alizability of the findings reported should be care-
Table 26-2
275
Questions to consider when assessing a published report (continued)
• Was the proportion of patients who failed to receive
(heir assigned treatment low9
• Was there a reasonably high level of adherence to
the treatment regimens over the course of the
study?
• Is there a description of the effort made to monitor
for departures from the study protocol and for
maintaining data quality? Do you consider the
procedure to have been adequate, given the needs
and goals of the trial?
• Does the paper indicate how laboratory analyses
and readings from ECGs and other similar proce
dures were done?
• Does the paper contain a description of the quality
control procedures used to monitor laboratory
analyses and readings such as ECGs9 Do you
consider those procedures adequate, given the
needs and goals of the trial?
• Did the laboratory or readers perform the indicated
analyses or readings in a masked fashion (i.e.,
without knowledge of patient treatment)?
• Do the tables and graphs have intelligible headings
and legends?
• Are treatment comparisons adjusted for baseline
differences9
• Have the authors used a variety of analytic ap
proaches to support their conclusions, and do
they yield consistent results?
• Are there tabulations that describe the treatment
trends over time, such as via the use of lifetables
or cohort analyses?
• Are data presented in sufficient detail to permit the
serious reader to carry out additional analyses?
• Are the results internally consistent9
• Is there a stated cutoff date for the data included in
the report, and is there a stated rationale for the
date used? (Especially important if the report is
based on interim results.)
• Do the authors display statistical sophistication by
minimizing the use of p-value and significance
testing as a means of data interpretation?
7. Discussion
4. Data analysis procedures
• Does the methods section of the paper include de
scriptions of the data analysis procedures used,
and are the descriptions supported with appro
priate literature references9
• Are the methods of analysis appropriate?
• Is the paper based on data from all study patients?
If not, does it contain a statement indicating the
rationale for the data selection presented in the
report? Is the rationale reasonable?
• Are the key analyses based on original treatment
assignment and do they account for all patients
enrolled in the trial? If not, is the number of pa
tients not accounted for small and about the same
for each treatment group?
• Are data presented to describe the baseline compar
ability of the study groups?
• Is there an analysis that summarizes primary out
come data by original treatment assignment9
• Are patients who failed to receive the prescribed
treatment or who had low adherence to the as
signed treatment counted in the treatment group
to which they were randomized?
fully examined. The reader should be skeptical of
any results that pertain to a selected subset of the
patients or outcomes observed. Unfortunately, it is
not always easy to determine if this is the case. The
fact that certain patients or outcomes have been
excluded will be apparent only if the report con
tains statements to this effect and data for all pa
tients randomized into the trial. Sometimes the
only clue that some patients have been omitted is in
• Have the authors provided a discussion of their
results?
• Are the authors familiar with other relevant find
ings for the treatments being evaluated?
• Do the authors support statements contained in the
discussion section with appropriate literature ref
erences9
8. Conclusions
• Are the conclusions supported by the analyses pre
sented9
• Have the authors exercised a sufficient degree of
caution and conservatism in stating their conclu
sions?
• Have the authors refrained from overgeneralization
of the findings9
• Do the authors limit their conclusions to the types
of patients studied and to the treatments investi
gated9
• If the authors have concluded in favor of the null
hypothesis, do they provide a discussion of the
type II error possible with the sample size used9
the use of a single word or phrase (e.g„ as in the use
of the term evaluable patients or the phrase analysis
by treatment received).
In general, results from all clinics in a multicenter
trial following a common study protocol should be
presented in a single publication. However, another
way selection can occur is when individual clinics in
such trials have the option of analyzing and pub
lishing independently of other clinics. An investiga-
276
ii
I '■
■p
iI
■I
26.5
Locating and reading published reports
tor at a clinic producing a “statistically significant"
treatment difference is more likely to publish than
his colleague at a clinic who failed to produce any
noteworthy treatment differences. There is no easy
way to know if this form of selection occurs unless
the authors so indicate.
A judgment should be formed regarding the level
of statistical sophistication of the authors. Slavish
use of hypothesis testing should be seen as a mark
of naivety in the authors. The same is true for
simple characterizations of results as significant or
nonsignificant, depending on whether or not asso
ciated p-values are below or above the magical 0.05
level.
The study design, particularly as it relates to
safeguards against biases in the data collection pro
cess. should be examined. Some feeling for this
may be obtained by observing the extent to which
the investigators have attempted to mask data col
lectors in the trial. Vague statements concerning the
method of patient selection and assignment to treat
ment should raise questions concerning the ade
quacy of the treatment allocation process. Any sug
gestion that the authors equate a haphazard (see
Glossary) method of assignment to formal random
ization (see Glossary for definition of random)
should raise doubts regarding the validity of the
study.
Part of the assessment should focus on questions
concerning the quality and integrity of the data
generated. The methods section of the paper should
contain sufficient detail to answer questions con
cerning methods used to edit the data for errors or
inadequacies. The absence of any discussion of this
kind should raise questions concerning the ade
quacy of the data collection procedures used.
The reader must decide if the results observed
can be explained by differences in the baseline
composition of the treatment groups, by a differen
tial dropout rate, or by major differences in treat
ment compliance. Failure to provide information
that allows the reader to address these issues should
be viewed as a deficiency in the report.
Finally, a good paper will indicate if the results
presented are from interim or final analyses of the
data, and if they are of the former type why they are
being presented now as opposed to later when the
trial is finished.
26.4 VALID AND INVALID
CRITICISMS
There is no such thing as a perfect study, only
varying degrees of imperfection. The professional
critic can always cite one or more of the criticisms
listed in Table 26-3 without fear of contradiction.
For example, he can always argue that the results of
the trial should be ignored because the investigators
studied the “wrong" population Or he can chal
lenge the choice of treatments or the way in which
they were administered. And it is always possible to
chide investigators because they failed to collect
“important" data—at least as viewed from the per
spective of the critic. The problem is not coming up
with criticisms, but in deciding whether or not they
are valid. The trouble with the criticisms listed in
Table 26-3 is that they are so broad and sweeping
as to be beyond debate.
A criticism, to be valid, should:
• Have some basis in fact
• Be buttressed with supporting evidence
• Make a difference in the interpretation of the
results
Ttble 24-3
1
Universal criticisms
• Sample size not large enough
• Treatment groups not comparable with regard to some
baseline characteristic
• Important data overlooked in the data collection or
analysis processes (key baseline data missing, analysis
of some secondary outcome not done)
• Treatment protocol not followed in all cases
• Treatments not properly administered
• Amount of follow-up inadequate
• Clinical implications of findings questionable
• Design of the study flawed (wrong design, inadequate
stratification, wrong method of randomization)
• Execution of the trial faulty
• Errors made in the data collection, coding, or classifica
tion processes
• Inappropriate data analysis
• Important subgroups of patients overlooked in the anal
ysis
• Results cannot be generalized to ordinary clinical prac
tice
• Results are inconsistent with previous experience
• Results not definitive
Table 26-4
differences in the results.
• Avoids dogmatic pronouncements
• Appreciates the danger of subgroup analyses and the
limitations of a straight significance testing approach
to data interpretation
• Refrains from flamboyant statements designed more for
effect than for enlightenment
• Persuades through the force of argument rather than via
clever debating techniques and rhetoric
A clinical trial is not designed to produce abso
lute truth. A good critic will recognize that its
main strength is in the framework it provides for
comparing one treatment with another and that
comparisons among the treatment groups re
main valid, given proper methods of treatment
assignment, even if the populations studied are
select. He will avoid criticism for the mere sake
of criticism, and will recognize that criticism, to
be useful, must be focused on specific issues
concerning the design, execution, or analysis of
the trial. He will avoid vague criticisms that are
beyond debate. He will formulate his own list of
criticisms after reading the original report and
related documents and submit each of those criti
cisms to the tests discussed in the previous sec
tion before promulgating them. He will avoid
parroting the criticisms offered by others unless
he has carried out sufficient analyses of his own
• Wrong study population (too old. too young, too sick.
too healthy)
• Wrong treatments studied (dosage too high, dosage too
low. test treatment studied is not used in real life)
for the results observed. The variability has to be
sizable and must occur in connection with an im
portant predictor of outcome to make any real
26.5 DESIRABLE CHARACTERISTICS
OF A CRITIC
All three tests should be met. Among the three, the
third is the most difficult one to satisfy. For exam
ple, it is fairly easy to criticize a trial because of
differences in the baseline composition of the treat
ment groups. However, it is quite another thing to
show how those differences might have accounted
1
Desirable characteristics of a critic 277
to support them.
A critic should recognize that his views may
be colored by preconceived notions regarding
the treatments studied or by his specific inter
ests, scientific as well as financial, and hence will
disclose those interests in his critique. Critiques
that are commissioned and supported by a busi
ness firm with a proprietary interest in the treat
ments being evaluated should be so labeled. Feinstein (1971) took pains to disclose his interests in
and incentives for doing the UGDP critique. His
critique stands in marked contrast to those pre-
Characteristics of a responsible critic
• Reserves judgment until he has personally reviewed and
read all pertinent study reports and documents
• Does not make unsubstantiated claims
• Does not impugn the integrity of others without factual
data to support the charge
• Knows the general design strengths and weaknesses of
clinical trials
• Understands the concept of randomization and its uses
in a research setting
• Concentrates criticisms on weaknesses that could have
been corrected by better design procedures, not on
weaknesses common to any clinic trial
• Knows his own limitations and seeks the help of others
for assessment of areas outside his domain of compe
tence
• Reveals any motivations and incentives (including those
of a financial nature) that may have inOuenced his
judgment regarding the trial
• Voluntarily discloses any interests that have the poten
tial of being viewed as a conflict of interest
pared by others concerning the study, such as
those by Schor (1971) and Seltzer (1972). Disclo
sure of motivations and interests is important in
that it permits readers to make their own judg
ment as to the degree to which they may have
influenced the objectivity of the critic. Table 26-4
lists characteristics of a good critic.
Part VII. Appendixes
A. Glossary
B. Sketches of selected trials
C. Year 1980 clinical trial publications
D. Activities by stage of trial
E. Sample consent statements
F. Data items and forms illustrations
G. Sample manual of operations, handbook, and monitoring report
H Budget summary for Hypertension Prevention Trial Data Coordinating Center
I. Combined bibliography
Appendix A contains terms and acronyms used in this book plus terms considered to be
common to the class of trials considered herein. Appendix B contains detailed design and
operational information on 14 long-term trials. Appendix C provides the list of papers
considered for review in Chapter 2. Appendix D relates to Chapter 3. It details the activities
of a “typical” trial as it progresses from beginning to end. Appendix E relates to Chapter 14.
It contains sample consent forms from the Hypertension Prevention Trial (HPT), Macular
Photocoagulation Study (MPS), and the Persantine Aspirin Reinfarction Study (PARIS).
Appendix F provides illustrative material related to the construction of data forms, as
discussed in Chapter 12. Appendix G contains illustrative materials from a manual, hand
book, and treatment monitoring report from the HPT, MPS, and PARIS. It relates to
Chapters 16 and 20. Appendix H relates to Chapter 21 and the budgetary process for
coordinating centers. Appendix I provides a combined bibliography of all references listed in
the various chapters and appendixes of this book, except Appendixes B and C.
1
I 'f
279
I
A. Glossary
A.l
PREFACE
This appendix sets forth terms and acronymns use. Terms with more than one definition, such
appearing in this book, plus other terms com as sequential analysis and treatment effect, may
be used in different ways depending on the con
mon to the class of clinical trials considered.
Terms from other fields, most notably statistics text. Italic print is used to denote terms that are
and epidemiology, are covered, but only to a defined elsewhere in the Glossary.
This Glossary, while extensive enough to
limited extent. Readers should see Last (1983);
Kotz et al. (1982); Kruskal and Tanur (1978); cover usage conventions in this book, is not
Kendall and Buckland (1960); and James and comprehensive enough to cover the entire scope
James (1959) for more comprehensive glossaries of clinical trials. The hope is that it will serve to
stimulate others to extend coverage to other
of terms for these two fields.
Appendix I contains a list of reference sources classes of trials and that it will lead to greater
uniformity of language conventions in the field
used in the Glossary. Citations 69, 125, 128, 263,
of clinical trials (Meinert, 1980a).
272, 309, 330, 332, 343, 431, 438, 491, 495, and
502 represent general reference sources. Sources
related to specific terms are cited in conjunction
A.2. GLOSSARY
with those terms.
The impetus for this Glossary arose from
A
work of the author in the Coordinating Centers
AAW Ask as written.
Models Project (Coordinating Center Models
achieved sample size Observed sample size.
Project Research Group, 1979a, 1979c). That
ACTH Adrenocorticotrophic hormone,
work required a vocabulary to facilitate compar
active control Active control treatment.
ative analyses of the design, organizational, and
active control treatment A control treatment that
involves use of a pharmacologically or medically
operating features of the trials reviewed in that
active substance. See inactive control treatment for
project.
opposing term.
Communication in clinical trials is confused
ad hoc review group A review group that is created
by use of different terms to designate the same
for the sole purpose of reviewing a specified appli
concept, detail, or practice. A case in point in
cation or set of applications. Also referred to as ad
volves the term outcome, defined herein as a
hoc study section, especially if the applications are
result, condition, or event associated with indi
for grant support.
vidual study patients and which is used to assess
ad hoc study section A study section created to
the efficacy of the study treatments. Other re
review a specified application or set of applica
lated or equivalent terms include event, response
tions, especially applications for NIH grant sup
variable, and endpoint.
port.
ADA American Diabetes Association.
Practice in this book varies with regard to use
adaptive allocation A treatment assignment process
of modifiers of base terms. They are frequently
in which the treatment allocation ratio is allowed
dropped when meanings are considered to be
to change as a function of the number of patients
clear without their use. For example, the term
enrolled, observed baseline data, or observed out
allocation is often used as shorthand for treat
comes (see Simon. 1977).
ment allocation, and trial is frequently used as a
adaptive allocation design Adaptive treatment allo
shorthand expression for clinical trial. Many of
cation design.
the more commonly used shorthand expressions
adaptive random allocation Adaptive allocation in
which the treatment assignments are made via a
appear in the Glossary.
Various terms in the Glossary are accompan
random process.
adaptive randomization A treatment assignment
ied by usage notes (e.g., see endpoint, coopera
process using adaptive random allocation.
tive clinical trial, and blind). These notes are
adaptive treatment allocation design A treatment
used to indicate the way in which a specific term
allocation design in which the treatment allocation
is used in this book, or reasons for avoiding its
281
1
282
8
I
I
I..-
'^1
2
A. Glossary
ratio is allowed to change over the course of pa
tient enrollment.
adaptive treatment allocation schedule A treatment
allocation schedule constructed using an adaptive
allocation scheme.
adherence Treatment adherence.
adverse drug reaction Any side effect associated
with use of a drug that has adverse health implica
tions.
adverse side effect Any side effect associated with a
treatment procedure that produces an adverse ef
fect or that has adverse health implications for the
patient receiving the treatment.
advisory-review committee (ARC) A committee in
the organizational structure of a trial that is re
sponsible for advising the steering committee and
the sponsor on operation of the trial. Usually com
posed of individuals neither directly involved in the
execution of the trial nor associated with any of the
participating centers or sponsor of the trial. A key
committee in the organizational structure of a multicenter trial. See policy board and policy-advisory
board.
advisory-review and treatment effects monitoring
committee (ARTEMC) A committee that per
forms the functions of both the advisory-review
committee and treatment effects monitoring com
mittee. A key committee in the organizational struc
ture of a multicenter trial.
allocation The process of making a treatment allo
cation.
allocation ratio Treatment allocation ratio.
allocation schedule Treatment allocation schedule.
allocation strata Treatment allocation strata.
alternative hypothesis I. An alternative to the null
hypothesis that specifies some true underlying dif
ference or set of differences between two or more
populations or groups with regard to some func
tion, trait, characteristic, or effect. It may be stated
in such a way so as to be concerned with a difference(s) in only one direction (one-sided alternative
hypothesis) or in either direction (two-sided alter
native hypothesis) relative to the null value.
2. Alternative treatment hypothesis.
alternative treatment hypothesis A hypothesis that
states that the true underlying effect of the test
treatment, as expressed by a specified outcome
measure, is different from that associated with the
control treatment.
AMIS Aspirin Myocardial Infarction Study.
analysis by intention to treat A method of data
analysis in which the primary tabulations and sum
maries of outcome data are by assigned treatment.
See also analysis by treatment administered.
analysis by treatment administered A method of
data analysis in which the primary tabulations and
summaries of outcome data are by treatment ad
ministered, not by treatment assigned (see Taylor
et al„ 1982, for usage example). See also analysis
by intention to treat.
C 283
analysis database The subset of data contained in
the study database that can be accessed for data
analysis. Generally limited to data from the study
database that have been coded, keyed, and stored
electronically for easy retrieval and manipulation.
ancillary study An investigation, stimulated by the
trial and intended to generate information of inter
est to the trial, that is designed and carried out by
investigators from one or more of the centers in the
trial and that utilizes resources of the trial (e.g.,
money, study patients, staff time, etc.), but that is
not a required part of the design or data collection
procedures of the trial.
applicant Anyone who makes an application to
carry out a designated research project, particu
larly under the grant mode of funding. See also
offeror and proposer.
ARC Advisory-review committee.
ART Anturane Reinfarction Trial.
ARTEMC Advisory-review and treatment effects
monitoring committee.
assigned treatment The treatment designated to be
administered to a patient, as indicated at the time
of his enrollment into the trial.
assignment probability Treatment assignment prob
ability.
assignment process Treatment assignment process.
assignment unit Treatment assignment unit.
award Funding award.
B
balancing interval Treatment block.
baseline A time point or set of data that serves as a
basis for gauging changes in subsequent measure
ments or observations.
baseline adaptive allocation A treatment assign
ment process in which assignment probabilities are
allowed to change over the course of the trial as a
function of observed differences among the treat
ment groups for one or more baseline variables.
Changes in the assignment probabilities are made
so as to achieve comparable study groups with
regard to the variable(s) used in the adaptive pro
cess.
baseline adaptive random allocation Adaptive ran
dom allocation based on one or more observed
baseline characteristics of enrolled patients.
baseline adaptive randomization A treatment as
signment process using baseline adaptive random
allocation.
baseline characteristic Baseline variable.
baseline data I. The set of data collected on a spe
cific patient or set of patients during the preran
domization and randomization visits. 2. The same
as definition I, except excluding data collected at
the randomization visit. In this book, data col
lected at the randomization visit are considered to
be part of the baseline data set.
baseline examination An examination that is car
ried out as part of a baseline visit and that is
designed to assess a patient’s eligibility for enroll
ment into the trial and to produce required base
line data.
baseline observation An observation or recording of
a baseline variable made on the observational unit.
baseline variable A variable that is measured, ob
served, or assessed on a patient at or shortly before
treatment assignment and the initiation of treat
ment.
baseline visit I. A visit that takes place either be
fore randomization or during the randomization
visit. 2. Any prerandomization visit, excluding the
randomization visit. Usage note: The visit at which
randomization occurs is considered to be a base
line visit in this book. Reasonable, so long as there
are no data collected after randomization, or so
long as data collected after randomization are free
of treatment effects.
Bayesian analysis A method of data analysis that
provides a posterior probability distribution for
some parameter which is a function of observed
data and a prior probability distribution for the
parameter (see Cornfield, 1966b, 1969).
Bernoulli random variable A random variable that
is capable of assuming one of two values, e.g., 0 or
I, with fixed probabilities, P and l-P, respectively
(see Feller, 1968).
Bernoulli trial A single replication of an experimen
tal procedure on a defined observational unit with
a Bernoulli random variable as the outcome.
BHAT Beta Blocker Heart Attack Trial.
bias I. A preconceived personal preference or incli
nation that influences the way in which a measure
ment, analysis, assessment, or procedure is per
formed or reported. From Old French biais,
meaning oblique. From Old Provencal, perhaps
from Greek epikarsios, meaning oblique. 2. A spec
ified instance of a preconceived preference or incli
nation.
biased coin randomization A method of randomi
zation in which treatment assignment probabilities
are modified as a function of the observed differ
ence in the number of patients already assigned to
the study treatment groups.
BID Bureau, institute, or division.
binary outcome Binary variable.
binary outcome measure An outcome measure that
can assume only one of two values, such as in a
trial with death as the outcome measure.
binary variable A variable that is capable of assum
ing one of two possible values, 0 or 1, or more
generally E\ or fy. The variable is equivalent to a
Bernoulli random variable if the probabilities of E\
and E2 are fixed.
BIOSIS Biological Abstracts (a literature database;
see Chapter 26).
blind Masked. (Usage note: Term not used because
of potential for confusion, especially when used in
conjunction with a trial where loss of vision is the
outcome measure, or in a trial involving patients
who have lost their vision.)
blinded Masked. (See blind for usage note.)
BLIPS Biometrics Laboratory Information Process
ing System.
block 1. A group, quantity, section, or segment
that is considered as a unit for some purpose, proce
dure, process, or action. 2. (clinical trials) Treat
ment block.
block size 1. The number of individual elements
making up a block. 2. Treatment block size.
blocking The process of establishing defined
groups, as in a treatment allocation schedule de
signed to provide prespecified treatment block
sizes.
BMDP Bio-Mathematics Data Processing.
business office The office in an investigator’s insti
tution with legal responsibility for receiving funds
from the sponsor and for expenditure of those
funds under specified ground rules.
C
CANCERLIT Cancer Literature (a literature data
base; see Chapter 26).
case-control study (epidemiology) A study that in
volves the identification of persons with the disease
or condition of interest (cases) and a suitable group
of persons without the disease or condition of in
terest (controls). Cases and controls are compared
with respect to some existing or past attribute or
exposure believed to be causally related to the
disease or condition. Also referred to as a retro
spective study because the research approach pro
ceeds from effect to cause. The term applies even if
cases and controls are accumulated in a prospec
tive manner (Last, 1983, Schlesselman, 1982).
CASS Coronary Artery Surgery Study, including
the Coronary Artery Surgery Trial (CAST).
CAST Coronary Artery Surgery Trial, see C/ISS.
CATLINE Catalog on Line (a literature database;
see Chapter 26).
CC Coordinating center.
CCD Committee for the Care of the Diabetic.
CCMP Coordinating Center Models Project.
CCU Coronary care unit.
CDC Centers for Disease Control (a part of the
United States Public Health Service), Atlanta,
Georgia.
CDP Coronary Drug Project
center An autonomous unit in the structure of a
clinical trial that is involved in the collection, deter
mination, classification, assessment, or analysis of
data, or that provides logistical support for the
trial. To be counted as a center, the unit must have
a defined function to perform, must be administra
tively distinct from other centers in the trial, and
must function during one or more stages of a trial.
Centers include clinical center, data center, cnordi-
A
C 285
284 A. Glossary
h ■ ■
l:
1
I
&
I
L
noting center, data coordinating center, treatment
coordinating center, central laboratory, procure
ment and distribution center, project office, read
ing center, and quality control center.
center director The administrative head of a center.
central distribution of funds A method for distribu
tion of funds in which one center in a multicenter
trial receives funds for execution of the trial and
which, in turn, is responsible for distribution of
funds to other centers in the trial.
central laboratory A center in a multicenter trial
responsible for performing specified laboratory de
terminations on specimens collected from patients
enrolled or considered for enrollment into the trial.
Not counted as a separate center if administered as
part of another center in the trial.
centralized database A database held and main
tained in a central location, especially in a multi
center trial. See also distributed database.
centralized data entry A system in which all data
generated in a trial are received at a central point
for keying.
chairman of the study Study chairman.
CHD Coronary heart disease.
check digit A single digit that is used to reveal re
cording errors in some numeric identifier in a rec
ord, such as patient identification number. It is
typically the last digit of the identifier and is as
signed when the identifying number is issued. The
assigned digit is compared with the one calculated
using the identifying number (devoid of the check
digit). The entire record is rejected for entry into an
existing data file if the assigned digit does not agree
with the calculated digit (see Selmer, 1967; Smythe,
1968; Fellegi and Sunter, 1969; and Anderson
et al., 1974 for discussion of check digits).
clinic Study clinic.
clinic coordinator 1. An individual in a study clinic
responsible for coordinating the data collection ac
tivities for that clinic and who expedites the flow of
data and related records from the clinic to the data
center, data coordinating center, or coordinating
center. 2. An individual in the data center, data
coordinating center, or coordinating center who is
responsible for coordinating the receipt of data
from study clinics and for communicating with
clinics regarding data flow. See data coordinator.
clinic director The administrative head of a study
clinic.
clinic monitor I. An individual located in the data
center, data coordinating center, coordinating cen
ter, or in the sponsoring agency who is responsible
for receiving data from participating clinics and for
initiating communications with those clinics re
garding data collection and data flow procedures.
2. Field monitor.
clinic visit Any patient visit to the study clinic dur
ing the enrollment or follow-up process that is
related to the data collection, examination, treat
ment, or patient care procedures of the trial.
clinical center I. A center in the organizational
structure of a clinical trial that is responsible for
recruiting, enrolling, treating, and following pa
tients in order to generate required data for the
trial. 2. Study clinic.
clinical coordinating center Treatment coordinating
center.
clinical event A change in a patient’s state of health
characterized by the occurrence of some discrete
event that is considered to have adverse health
implications (e.g., diagnosis of cancer, hospitaliza
tion for an MI, initiation of treatment for hyper
tension, death).
clinical research associate An individual, usually
having an advanced degree, typically in medicine,
employed by a drug firm to facilitate the initiation
and direction of clinical trials sponsored by the
firm.
clinical trial A research activity that involves ad
ministration of a test treatment (e.g., a drug, surgi
cal procedure, diagnostic test, or medical device) to
some experimental unit in order to evaluate the
treatment. The term is subject to wide variation in
usage. In some cases it may refer to the first use of
a new treatment in man without any control treat
ment. In other cases it may refer to a rigorously
designed and executed experiment involving a test
and control treatment and randomization. The
experimental unit in most cases is man (or a larger
unit involving man, such as a hospital ward), but
can be some other experimental animal. (Usage
note: In this book, the term clinical trial or simply
trial always refers to a controlled clinical trial in
volving human beings.)
CLINPROT Clinical Protocols (a cancer literature
database; see Chapter 26).
close of trial The point at which the trial is consid
ered to be finished. Marked by completion of the
patient close-out or termination stage of the trial,
depending on whether the closing point is asso
ciated with completion of regular follow-up visits
or with data analysis.
close-out The process of separating a patient from
the trial after completion of required follow-up.
close-out examination The final examination or
series of final examinations performed on patients
just prior to termination of regular follow-up in the
trial.
close-out follow-up visit A follow-up visit made by
a study patient to a study clinic that is used for data
collection and to carry out specified procedures
related to his separation from the trial.
close-out stage Patient close-out stage.
closed sequential design A sequential design that
allows the experimenter to terminate the trial after
a certain number of observations, even if the ob
served treatment difference is not large enough to
allow the experimenter to conclude for or against
the test treatment. Distinct from an open sequen
tial design, which requires continuation of the trial
until the difference is large enough to warrant a
conclusion for or against the test treatment (see
Chapter 9).
closed sequential trial A trial with a closed sequen
tial design (see Chapter 9).
cohort A group of people defined by a common
characteristic or set of characteristics. Middle Eng
lish, from Old French cohorte, from Latin cohors,
meaning enclosed yard, company of soldiers, mul
titude. One-tenth part of an ancient Roman legion.
common calendar date close-out A method of pa
tient close-out in which all patients enrolled in the
trial are separated from it at or about the same
calendar date, regardless of when they were en
rolled. See common period offollow-up close-out
for opposing term. See also Chapter 15.
common period of follow-up close-out A method of
patient close-out in which patients are separated
from the trial after a specified period of follow-up
(e.g., after two years). See common calendar date
close-out for opposing term. See also Chapter 15.
comparative clinical trial Any clinical trial involv
ing two or more treatment groups. See controlled
clinical trial.
comparative study A study involving two or more
defined groups of patients in which groups are
compared, one with another, in order to make a
judgment regarding the influence of some factor,
condition, trait, or procedure that is present or
applied to one group but not to the other(s). Syn
onymous with controlled clinical trial if the study
entails comparison of different treatments involv
ing patients enrolled and treated over the same
period of time.
comparison group The group of patients desig
nated or selected for comparison with all other
groups in a study. The control-treated group of
patients in a controlled clinical trial.
comparison treatment Control treatment.
compliance I. To be in a compliant state. 2. A
quantitative indication of the compliant state, as in
the sentence: Compliance to the protocol was low.
compliant Willing to carry out a set of procedures
or practices in accordance with established guide
lines or standards.
composite event An event that is considered to have
occurred if any one of several different outcomes
are observed (e.g., occurrence of an attack of an
gina pectoris, a transient ischemic attack, or a my
ocardial infarction in a trial using a composite
vascular event as the outcome measure).
computer A programmable electronic device that
can be used to store and manipulate data in order
to carry out some designated function.
computer terminal Any device (dumb or intelligent)
that can be used for data input or output. It may be
part of a network of terminals connected to a
larger computing facility or may operate indepen
dently of all other facilities.
concurrent control A control that is based on data
collected over the same period of time as that used
to generate all other data in the study. See also
historical control.
confounding variable I. (epidemiology) A varia
ble that is related to two factors of interest (e.g.,
disease state and degree of exposure to some agent
in a case-control study; treatment assignment and
outcome in a clinical trial) that falsely obscures or
accentuates the relationship between the factors
(see Breslow and Day, 1980; Last, 1983). 2. A base
line variable in a clinical trial that influences the
outcome and that has a different distribution in the
treatment groups being compared.
consortium agreement An agreement between the
sponsor and one of the centers in a multicenter
study in which the center agrees to receive funds
for its own operations and that of other centers in
the study and to disperse funds among the centers
on an as-needed basis or as specified in the agree
ment.
consortium award I. A grant or contract awarded
by the sponsor to a center for execution of a study
involving multiple centers. The center receiving the
award assumes responsibility for allocation of
funds to all other participating centers in the study.
2. Same as definition I except that the award is for
support of only certain other centers in the study.
Remaining centers are funded in other ways.
continuous variable A variable that is capable of
assuming any value over a specified range.
contract A legally binding written agreement be
tween the sponsor and the business office of the
investigator’s place of employment that outlines
the nature and schedule of work to be performed
and terms of payment for said work.
contract office The office in the sponsoring agency
or lead center whose members are responsible for
negotiating, awarding, and funding contracts.
contract officer The individual in the sponsoring
agency or lead center who is responsible for negoti
ating. awarding, and funding contracts for speci
fied projects.
control A standard of comparison for testing, veri
fying. or evaluating some observation or result.
control group Comparison group, control-treated
group.
control patient A patient assigned to the control
treatment.
control-treated group I. The group of patients in a
trial assigned to the control treatment. 2. The
group of patients in a trial who received the control
treatment, whether or not originally assigned to
that treatment (not used this way in this book).
control treatment The drug, device, test, or proce
dure administered in a clinical trial that serves as
the standard against which test treatments are eval
uated. The control treatment may consist of a
placebo medication, sham procedure, a standard
treatment regimen, or no treatment of any kind,
depending on the study design.
1
286
A. Glossary
D
control trial Controlled clinical trial. (See random
ized control clinical trial for comment.)
controlled I. Constrained, monitored, or watched.
2. A system of observation and data collection that
provides a basis for comparison, as with a compar
ison group.
controlled clinical trial A clinical trial involving one
or more test treatments, at least one control treat
ment, and concurrent enrollment, treatment, and
follow-up of all patients in the trial.
controlled trial Controlled clinical trial.
conventional author citation A method of citation
in which specific individuals are designated on the
title page or elsewhere in a manuscript as its au
thors. See corporate author citation for opposing
term.
cooperative agreement I. An agreement between
an institute in the National Institutes of Health and
a set of investigators that provides a defined struc
ture for sponsor-investigator cooperation in the
design and execution of a research project. 2. Any
written agreement between a sponsor and investi
gators) that provides a defined role for both par
ties in the design and conduct of a specified re
search project.
cooperative clinical trial Term frequently used to
denote a multicenter trial. The term is avoided in
this book. Cooperation is required for execution of
any trial, whether or not it involves multiple clin
ics.
coordinating center A center in the structure of a
study that is responsible for receiving, editing, pro
cessing, analyzing, and storing data generated in a
study and that, in addition, has responsibility for
coordination of activities required for execution of
the study. See also data center, data coordinating
center, and treatment coordinating center.
coordinating center director The administrative head
of the coordinating center.
coordinator I. The individual in the data center,
data coordinating center, or coordinating center
who is responsible for coordinating the receipt of
data from study clinics and for communicating
with clinics regarding data flow, clinic coordinator,
data coordinator. 2. The director of the coordinat
ing center. (Usage note: Term not used in either
definition 1 or 2 context in this book.)
corporate author citation A method of citation in
which authorship of a given manuscript is ascribed
to a corporate entity (e.g., as in a paper listed as
having been authored by the Coronary Drug Proj
ect Research Group).
cost-reimbursement contract A contract in which
the amount of money paid to the contractor by the
sponsor is dictated by reasonable and allowable
expenses for the work performed.
Cox proportional hazards regression model A
method of analysis developed by D. R. Cox (1972)
involving regression analysis which is used to ad
just observed event rates, such as those obtained in
a clinical trial where patients are enrolled over a
period of time and followed to a common calendar
date, for variables (usually observed at baseline)
which are believed to influence the rates.
CPHA Commission on Professional and Hospital
Activities.
CPPT Coronary Primary Prevention Trial, a trial
conducted by the Lipid Research Clinics.
CPU Central processing unit (of a computer).
CRISP Computer Retrieval of Information on
Scientific Projects (a database of ongoing work; see
Chapter 26).
cross contamination Treatment cross contaminalion.
crossed treatment design I. Crossover treatment de
sign. 2. Factorial treatment structure.
crossed treatments Two or more study treatments
that are used in sequence (e.g., as in a crossover
design) or in combination (e.g., as in a factorial
treatment structure).
crossover Treatment crossover.
crossover design Crossover treatment design.
crossover treatment design A treatment design that
calls for the administration of two or more of the
study treatments in a specified order to experimen
tal units in the trial.
crossover trial Crossover clinical trial.
crossover clinical trial A clinial trial involving a
crossover treatment design.
CRT Cathode ray tube.
cut-point I. The point or value in an ordered se
quence of values that is used to separate those
values into two subparts. 2. Subgrouping cut
point.
CV Curriculum vitae.
D
data (pl. of datum) Factual information, such as
measurements, observations, or statistics, which is
used as a basis for reasoning, discussion, or calcu
lation. (Usage note: In this book, the term refers to
information collected and recorded on patients con
sidered for enrollment or actually enrolled in a
trial.)
data audit The comparison of specific items of information ccr.tair.cd
contained in an original data form (or
some other kind of record) with that produced for
some transcribed version of that form or record
(e.g., as contained on a listing of the computer data
file of the form or record) as a check for discrepan
cies.
data center 1. A center in a study structure that is
responsible for receiving, editing, processing, ana
lyzing, and storing data generated in the study, but
that has few if any of the other general coordina
tion responsibilities assumed by a data coordinat
ing center or coordinating center. 2. A center in a
study structure that is responsible for receiving,
editing, processing, analyzing, and storing data gen-
erated in the study, regardless of whether or not the
center has other more general coordination respon
sibilities. Used in this sense in this book where
there is no need to distinguish between a data
center, data coordinating center, or coordinating
center, and where the emphasis is on data intake
and processing functions.
data collection visit Any visit by a patient to the
study clinic that is used for data collection in the
trial.
data coordinating center A center that has the du
ties of a data center as well as general coordination
duties for data collection. The modifier data is
sometimes used in structures having two or more
centers with specified coordination responsibilities
(e.g., in a structure with both a treatment coordi
nating center and a data coordinating center).
data coordinator An individual in the data center,
data coordinating center, or coordinating center
who is responsible for coordinating the receipt of
data from study clinics and for communicating
with clinics regarding data flow. Sometimes also
clinic coordinator, but not in this book.
data dredging A term used to characterize analyses
that are done on an ad hoc basis, without benefit of
prestated hypotheses, as a means of identifying
noteworthy differences.
data editing I. The process of reviewing data for
the purpose of detecting deficiencies or errors in
the way they are collected or recorded. 2. The
process of detecting deficient or erroneous values
on completed data forms.
data entry 1. The process of keying data, as con
tained on completed data forms, in order to render
information into an arrangement more suitable for
storage and subsequent use, usually for tabulations
and analyses, especially on a computer. 2. The
process of filling out a data form.
data field I. A space on a data form or in an elec
tronic record designated to contain, or that actu
ally contains, alphabetic or numeric characters of
information recorded in response to a specific data
item on the form. 2. The actual collection of al
phabetic or numeric characters used to denote in
formation recorded in response to a specified ques
tion or statement on a data form.
data file A collection of data records. The collection
may be of paper records or of electronic records
that are arrayed in some way.
data form A collection of data items all arrayed on
the same paper record.
data item I. A question or statement and related
area to be used by the respondent in answering the
question or completing the statement appearing on
a data form. 2. Data field.
data monitoring committee I. TYeatment effects
monitoring committee. 2. A committee with treat
ment effects monitoring responsibilities plus other
monitoring responsibilities, such as those needed
for assessing data quality or for assessing the per-
287
formance of clinics participating in a trial. 3. A
committee that is responsible simply for monitor
ing data quality and the performance of centers in
a trial (i.e., has no treatment effects monitoring
responsibilities).
data record A collection of data items that is
treated as a unit for some purpose or function.
data reduction I. The process of taking raw data,
as recorded on study forms, and of codifying and
classifying them in such a way so as to condense
them into a form suitable for data entry and elec
tronic storage. 2. The process of taking data al
ready contained in an electronic record and sum
marizing them, through the use of various
classification schemes and arithmetic manipula
tions, so as to condense them into a form suitable
for tabulations, analyses, listings, etc.
data and safety monitoring committee Treatment
effects monitoring committee.
data system A package of interrelated procedures
or routines that are performed by hand or com
puter to carry out some function or set of functions
(e.g., data management or data analysis).
database A collection of data files that are organ
ized in a specified manner and that are accessed by
designated personnel for designated purposes.
datum Singular of data.
DCC Data coordinating center.
DCCT Diabetes Control and Complications Trial,
dedicated computer A computer that is under the
exclusive control of a single user (or group of
users) and that is used for a specified project, func
tion, or activity.
DESI Drug Efficacy Study Implementation.
design unit The observational unit used for sample
size calculations in a trial. Usually a patient, but
may be some larger unit in special cases, such as in
trials using hospital wards, families, or the like as
the treatment unit. Always a patient in this book.
DHEW Department of Health, Education and Wel
fare (a department in the executive arm of the
United States government until May 1980; its func
tions are now met by the Department of Health
and Human Services and the Department of Edu
cation).
DHHS Department of Health and Human Services
(a department in the executive arm of the United
States government).
diagnostic clinical trial A clinical trial designed to
evaluate the usefulness of some diagnostic proce
dure, tool, or device.
diagnostic trial Diagnostic clinical trial.
dichotomous variable A discrete variable that has
only two possible values. Binary variable.
direct award An award of funds (grant or contract)
made directly from the sponsor to a study center.
See also indirect award.
direct distribution of funds Distribution of funds to
centers in a study directly from the sponsor, as in
direct awards.
1
288
j't.: ■
I:
|
|
I
|
j
I,.
F 289
A. Glossary
direct patient contact Patient contacts that are in
itiated by the study clinic for the purpose of patient
recruitment and that are directed at specified pa
tients without any reliance on interviewing per
sons. agencies, institutions, or generalized advertis
ing campaigns to make the contacts. See also
indirect patient contact and Chapter 14.
direct patient recruitment Any method of patient
recruitment that involves direct patient contact.
direct research cost The cost for salaries, equip
ment. supplies, and the like associated with the
actual design, conduct, and analysis of a research
project. See also indirect research cost.
director Center director.
discrete variable A variable that is capable of as
suming only certain values over a defined range, as
for dichotomous (binary) or polychotomous vari
ables. See also continuous variable.
distributed data analysis Any arrangement in a mul
ticenter trial whereby investigators in the various
centers have access to the analysis database, or
portions of it, for the purpose of carrying out data
analyses.
distributed data entry A method of data entry in
multicenter trials where data generated at the clin
ics are keyed on site.
distributed data system A data system that is estab
lished and maintained at the various clinics in a
multicenter trial in order to perform functions nor
mally carried out at the data coordinating center.
distributed database A database that is made up of
component parts which reside at geographically
diverse locations (e.g., in clinics in a multicenter
trial).
distribution center Procurement and distribution
center.
DMSO Dimethyl sulfoxide.
double-blind Double-masked in this book. See
blind for usage comment.
double-blinded Double-masked in this book. See
blind for usage comment.
double-blinded clinical trial Double-masked clinical trial.
double-mask Double-masked.
double-masked 1. A procedure in a clinical trial for
issuing and administering treatment assignments
by code number in order to keep study patients and
all members of the clinic staff, especially those
responsible for patient treatment and data collec
tion, from knowing the assigned treatments.
2. Any condition in which two different groups of
people are purposely denied access to a piece of
information in order to keep that information from
influencing some measurement, observation, or
process.
double-masked clinical trial A clinical trial with
double-masked administration of the study treat
ments.
DRG Division of Research Grants of the National
Institutes of Health.
drop-in A term sometimes used (not in this book)
to denote a patient in a clinical trial who, although
assigned to one study treatment, receives one of the
other study treatments in place of, or in addition
to, the assigned treatment. See treatment crossover
for related term.
dropout A patient enrolled in a clinical trial who is
either unwilling or unable to return to the study
clinic for regular follow-up visits.
DRS Diabetic Retinopathy Study.
drug trial A clinical trial in which the test treat
ments are drugs.
dumb terminal A computer terminal that can act as
an input or output device, but that does not have
independent processing capabilities (as opposed to
an intelligent terminal).
dynamic allocation Adaptive allocation.
dynamic randomization Adaptive randomization.
dynamic treatment allocation schedule Adaptive
treatment allocation schedule.
E
early stopping I. A condition or provision incorpo
rated into the design of a clinical trial that enables
investigators to terminate patient recruitment or
treatment if data accumulated during the trial sug
gest an adverse or beneficial treatment effect. 2. A
term used to characterize an action involving ter
mination of a study treatment in a trial because of
adverse or beneficial treatment effects.
early stopping rule Stopping rule.
EC Executive committee.
ECG Electrocardiogram.
ECOG Eastern Cooperative Oncology Group.
edit check The process of reviewing a data item on a
completed data form for deficiencies in the way it is
completed or in the value reported.
edit query A statement generated from a review of a
completed data form that draws attention to a
suspected deficiency in an item of information on
the form and that requires some action by person
nel responsible for generation of the data to clear
the query.
editorial review committee A committee created in
the organizational structure of an investigative
body that has responsibility for reviewing manu
scripts produced by that body.
effective sample size Sample size after reductions
due to dropouts and treatment noncompliance. See
expected effective sample size and observed effec
tive sample size.
EFM Electronic fetal monitoring.
EMED Excerpta Medica (a literature database; see
Chapter 26).
endpoint I. A primary or secondary event that,
when observed in a patient, leads to termination or
alteration of treatment or follow-up. 2. A primary
or secondary event observed in a patient during the
course of treatment or follow-up. 3. Outcome.
4. Early stopping. 5. Slopping rule. (Usage note;
The term, endpoint is not used in this book because
of the potential for confusion. Use of the term in
the sense of definitions 1, 2, or 3 can mean that
patients are no longer eligible for treatment or
follow-up once they experience a specified event.
This is obviously true where the event is death, but
need not be so for nonfatal events. In fact, the
design of the trial may require continued treatment
and follow-up of patients over the entire course of
the trial, regardless of the number of nonfatal
“endpoints" observed. See event, clinical event,
primary outcome, and primary event for preferred
terms.)
enrollment Patient enrollment.
enrollment process Patient enrollment process.
equal allocation Equal treatment allocation.
equal treatment allocation A scheme in which the
assignment probability in the randomization pro
cess for any one treatment is the same as for every
other treatment in the trial. See uniform treatment
allocation.
estimated sample size The number of patients re
quired for a study, as derived from a sample size
calculation or in some other way.
ETDRS Early Treatment of Diabetic Retinopathy
Study.
ethics committee 1. Treatment effects monitoring
committee. 2. Institutional review board.
ethics review committee I. Treatment effects mon
itoring committee. 2. Institutional review board.
evaluable patient Evaluable study patient.
evaluable patients Evaluable study patients.
evaluable study patient A study patient who is re
garded by investigators in the trial as having satis
fied certain conditions (e.g., developed a tumor of
a certain size during the trial, followed the assigned
treatment) and, as a result, is retained for analysis
purposes (a patient not satisfying the conditions is
not so retained).
evaluable study patients The subgroup of study patients considered by investigators in the trial to
satisfy certain conditions and, as a result, are re
tained for analysis purposes. Patients not satisfying
the conditions are not so retained.
event I. An occurrence, incident, or experience, es
pecially one of some significance. 2. Binary out
come measure. 3. Clinical event. 4. The actual oc
currence of a condition, trait, or characteristic that
is defined by a binary outcome measure.
event rate The number of events experienced by a
specified number of patients in a specified unit of
time.
examination Patient examination.
executive committee (EC) One of the key commit
tees in the organizational structure of a trial. Re
sponsible for direction of the day-to-day affairs of
the trial. Usually consists of the officers of the
study and perhaps others selected from the steering
committee. Headed by the chairman or vice chair=
man of the steering committee and reports to that
committee.
expected allocation ratio The allocation ratio ex
pected using a given set of treatment assignment
probabilities. See specific allocation ratio.
expected effective sample size The number of ran
domization units (usually patients) specified when
the trial was planned, less reductions due to antici
pated losses from dropout and treatment noncom
pliance.
expected power The power computed for a given
treatment comparison when the trial was planned
(see Chapter 9).
experimental unit Tkeatment assignment unit.
explanatory trial Term used to characterize trials
that are designed to explain how a treatment works
(see Schwartz and Lellouch, 1967; Sackett and
Gent, 1979; Sackett, 1980). Term not used in this
book. See also management trial.
F
factorial structure Factorial treatment structure.
factorial treatment structure A treatment structure
in which one study treatment is used in combina
tion with at least one other study treatment in a
trial, or where multiples of a defined dose of a
specified treatment are used in the same trial. See
partial and full factorial treatment structure.
FDA Food and Drug Administration (a regulatory
agency of the United States government, located in
Rockville, Maryland).
feasibility study A preliminary study designed to
determine the practicality of a larger study. See
pilot study.
field monitor An individual employed by the spon
sor or a center of a trial (e.g., coordinating center)
to visit participating clinics to monitor data collec
tion procedures. See clinic monitor.
final data analysis The term given to data analyses
carried out at the end of the trial, normally in the
termination stage, for characterizing results ob
tained from the trial.
final examination Final patient examination.
final patient examination I. The last examination
of a patient prior to close-out. 2. The last exami
nation of a patient prior to enrollment.
fixed allocation Any method of treatment assign
ment involving a fixed treatment allocation ratio.
fixed allocation design Fixed treatment allocation
design.
fixed allocation ratio Fixed treatment allocation
ratio.
fixed allocation schedule Fixed treatment allocation schedule.
fixed-cost contract A contract in which there is a
prior agreement between the sponsor and the in
vestigator’s institution on the amount to be paid
for work to be performed, regardless of the actual
costs incurred.
1
290
L
I 291
A. Glossary
fixed sample size design A design in which the
number of patients to be enrolled is considered to
be fixed in advance of the start of patient enroll
ment for the study. The number may be deter
mined from a sample size calculation, or via other
considerations (e.g., cost, patient availability). It is
conventional to consider any study that does not
involve a sequential design as involving a fixed
sample size design, even if the number is not deter
mined before the start of the trial. See sequential
design for opposing term.
fixed treatment allocation design A treatment allo
cation design in which the treatment allocation
ratio is fixed.
fixed treatment allocation ratio An allocation ratio
that remains fixed.
fixed treatment allocation schedule A treatment al
location schedule based on fixed treatment assign
ment probabilities.
fixed treatment randomization A treatment assign
ment process in which the treatment assignment
probabilities remain fixed.
FOIA Freedom of Information Act (see Chapter
24).
follow-up Patient follow-up.
follow-up cohort 1. A group of patients enrolled
into a trial during the same time period. 2. A
group of patients enrolled at different time points,
but who are followed for the same length of time
(e.g., each patient for two years).
follow-up data Data collected on a patient, or a set
of patients, after enrollment into a trial.
follow-up data collection visit Any data collection
visit that takes place after a patient is enrolled into
the trial.
follow-up examination A patient examination made
at a follow-up visit.
follow-up observation An item of data collected on
a patient (or larger treatment unit) after enrollment
in a trial.
follow-up study Prospective follow-up study.
follow-up variable A variable observed on individ
ual patients (treatment units) after enrollment into
a trial.
follow-up visit Any patient clinic visit that takes
place after the randomization visit for studyrelated purposes. See required and nonrequired fol
low-up visit for classes of follow-up visits. See also
treatment adjustment, regular, interim, close-out,
post-close-out, and post-trial follow-up visit for
specific types.
free treatment arm 1. A treatment that is selected
by the study physician or study patient. 2. A study
group that receives the treatment selected by study
physicians or study patients.
FTE Full-time equivalent.
full factorial structure Fullfactorial treatment struc
ture.
full factorial treatment structure A treatment struc
ture in which each study treatment is used in com-
bination with every other study treatment. The
treatments may involve different drugs or proce
dures, or different levels or doses of the same treat
ment.
funding agency The institution, organization, or
foundation that provides fiscal support for a given
study. Sponsoring agency.
funding award A grant or contract awarded to an
institution for a designated project.
funding office The office responsible for fiscal ne
gotiations with the centers of a study and for the
disbursement and administration of funds for use
in the study. Grants management office, contract
office.
funding officer The head of the funding office.
Grants management officer, contract officer.
FY Fiscal year.
G
general-use computer A computer that is used by a
variety of users working on unrelated tasks or
studies.
grant A funding award from the sponsor to an in
vestigator, via his institution, to support designated
work. A grant, as opposed to a cooperative agree
ment or contract, is generally made in anticipation
of relatively little involvement in the work by the
sponsor.
grants management office The office in the spon
soring agency whose members are responsible for
negotiating and awarding grants and for disburse
ment of funds for execution of grant-funded pro
jects. See contract office for corresponding term in
contract-funded work.
grants management officer The individual in the
sponsoring agency with legal authority to negotiate
grant awards and to disburse funds in connection
with those awards. See contract officer for corre
sponding terms for contract-funded projects.
group sequential analysis A method of interim data
analysis that is carried out after enrollment of a
specified number of patients, as discussed by Pocock (1977) and DeMets and Ware (1980).
H
handbook Study handbook.
haphazard A process occurring without any appar
ent order or pattern. Distinct from random, as
used in this book, in that there is no mathematical
basis for characterizing a haphazard process.
hard endpoint Hard outcome.
hard outcome Any outcome measure that is not
subject to serious errors of interpretation or mea
surement. Usually death or some other explicit
clinical event.
HDFP Hypertension Detection and Follow-Up
Program.
health scientist administrator An individual at the
National Institutes of Health who is responsible for
providing technical and scientific assistance to in
vestigators in a grant-funded study.
HEW (Department of) Health, Education and Wel
fare, DHEW.
HEX Committee Human experimentation commit
tee. See institutional review hoard.
HHS (Department of) Health and Human Ser
vices, DHHS.
historical control A control that is based on data
collected in a period of time previous to that used
for generation of data on the test-treated group of
patients. See concurrent control for opposing term.
historical control group A group of patients (may
be loosely or explicitly defined) considered to have
the same disease or condition as the study group,
but who were diagnosed and treated in a period of
time prior to that of the study group and who
received the conventional form of therapy for that
time. Historical control groups are generally only
useful for evaluations of treatments involving rare
diseases with highly predictable outcomes and
where it is considered impractical or unethical to
carry out a controlled clinical trial.
historical controls A collection of patients used as a
comparison group who were diagnosed and treated
for the disease or condition of interest in the past
and in a period of time that predates the period of
time covered for other study groups.
HPT Hypertension Prevention Trial.
human experimentation committee Institutional re
view hoard.
human volunteers committee Institutional review
board.
i
ID Identification.
ID check digit A digit that is part of the identifica
tion number of a record and that is used as a check
for transcription errors in that number. See check
digit.
IDE Investigational Device Exemption (see IND
for corresponding term for drugs).
IDEA Investigational Device Exemption Applica
tion (see INDA for corresponding term for drugs).
identification number Patient identification num
ber.
1G Investigative group.
IMPACT International Mexiletine Placebo Antiarrhythmic Coronary Trial.
inactive control treatment A control treatment that
is not considered to have any pharmacological or
physiological effect. A placebo treatment or sham
procedure. See also active control treatment.
IND Investigational New Drug (see IDE for corre
sponding term for medical devices).
INDA Investigational New Drug Application.
indirect award A funding award made to one center
via another using funds received from the sponsor,
as in a consortium award.
indirect patient contact Methods of patient contact
that are initiated by the patient, his own physician,
or some other intervening person, agency, or insti
tution, for the purpose of patient recruitment. See
direct patient contact for opposing term and Chap
ter 14.
indirect patient recruitment Any method of patient
recruitment that involves indirect patient contact.
indirect research cost The cost incurred by the insti
tution housing a research project for general admin
istrative support and for providing space, heat,
light, and the like in connection with the project.
See direct research cost for opposing term.
informed consent The voluntary consent given by a
patient to participate in a study after being in
formed of its purpose, method of treatment, proce
dure for assignment to treatment, benefits and
risks associated with participation, and required
data collection procedures and schedule.
initial design stage The first stage of a trial. Con
cerned with design and planning (see Chapter 3
and Appendix D).
institutional review board A committee or board,
as set forth in United States Public Health Service
guidelines for research involving humans and ap
pointed by authorities in a research institution,
constituted to review and approve studies to be
carried out on humans in that institution. The re
view focuses on the ethics of the proposed research
and on the adequacy of the proposed patient in
formed consent process.
intelligent terminal A computer terminal that can
be used to perform data processing independent of
the computer to which it is connected (as opposed
to a dumb terminal).
interaction I. (statistics, James and James, 1959) A
case in which one variable, y, is a function of
another, x, and in which variation in x associated
with a given change in y is affected by the value
assumed by a third variable, z. Interaction is said to
exist between y and z. 2. (clinical trials) A situa
tion in which the magnitude of the test-control
treatment difference for the outcome of interest
depends upon the value assumed by a third factor,
such as age or prior disease state of the study
patients.
interaction effect Tkeatment interaction effect.
interim analysis Interim data analysis.
interim data analysis 1. Any data analysis carried
out during the trial for the purpose of treatment
effects monitoring. 2. Any data analysis done be
fore the trial is finished, for whatever reason, but
usually concerned with assessments of treatment
effects. (Usage note: Strictly speaking, the term
applies to any fixed sample size or sequential
trial where such analyses are done. However, it is
conventional to reserve the term for use with fixed
A
292
M 293
A. Glossary
sample size designs. That convention is followed
herein.)
interim follow-up visit In this book, any visit by a
study patient to the clinic after randomization that
is not part of the required sequence of follow-up
visits and that is initiated by the study patient or
study physician because of some medical or treat
ment problem. Not counted as a requiredfollow-up
visit unless it takes place within the specified time
period for a required visit and all the required
procedures for that visit are carried out as part of
the interim visit. See nonrequired follow-up visit.
interim result I. Any test-control treatment differ
ence observed during the trial. 2. A test-control
treatment difference observed during the trial that
results in a treatment protocol change.
interim visit Interim follow-up visit.
intervention study A study in which there is an
effort to change the natural course of a disease or
condition by attempting to alter the risk factors or
precursors associated with that disease or condi
tion.
intervention trial Technically, any clinical trial,
since administration of any treatment in a trial
setting is a form of intervention. However, the term
is usually reserved for trials in which the test treat
ment entails life-style changes.
Investigational Device Exemption Application
(IDEA) An application directed to the Food and
Drug Administration by the manufacturer of a
medical device or independent investigator for per
mission to evaluate the device in humans (Food
and Drug Administration, 1983). See Investiga
tional New Drug Application for corresponding
term for drugs.
Investigational New Drug Application (INDA, also
IND) An application directed to the Food and
Drug Administration (made by submitting a No
tice of Claimed Investigational Exemption for a
New Drug) for permission to evaluate a drug (new
or old) for a new indication in humans. See Investi
gational Device Exemption for corresponding term
for medical devices.
investigative body Investigative group.
investigative group The entire staff involved in a
study. Includes center directors, representatives
from the sponsoring agency, members of all study
committees and key support staff.
investigative team investigative group.
investigator Study investigator.
I/O Input/output.
1RB Institutional review hoard.
IRSC International Reflux Study in Children.
IUD Intra-uterine device.
K
Kaplan-Meier product limit A nonparametric
method developed by Kaplan and Meier (1958) for
F-
estimating follow-up event rates using conditional
probabilities. The method is especially well suited
to situations, such as encountered in clinical trials,
where patients are enrolled over a period of time
and followed to a common calendar time point.
key committee In this book any of the following:
steering committee, executive committee, advisory
review committee, treatment effects monitoring
committee, or alternatively, the steering commit
tee, executive committee, and advisory-review and
treatment effects monitoring committee. Also
major committee.
L
label insert Package insert.
landscape-style page orientation A form of orienta
tion in which printed and visual information is
arrayed on the long axis of a page. See also por
trait-style page orientation.
lay representative A member of a committee, usu
ally the advisory review or advisory-review and
treatment effects monitoring committee, who is
chosen to represent patients in the trial and who
has no recognized research credentials.
lead center I. A center designated in a multicenter
study to take the lead in testing or performing
certain procedures in a study or that is designated
to assume a leadership role in the direction of the
study. 2. The center responsible for disbursing
funds to other centers in a study funded under a
consortium agreement.
lead clinic I. The clinic in a multicenter trial that is
responsible for testing patient examination and
data collection procedures to be used in a trial.
2. The first clinic funded, especially when that
clinic is responsible for developing and testing data
collection procedures to be used in a study.
lifetable An assembly of data in table or graph form
that summarizes the survival (or mortality) expe
rience of the observational units (patients in this
book) from some specified starting point. The start
ing point may be based on age, as in most lifetables
compiled by demographers, or on some event, such
as diagnosis of disease, or enrollment into a study,
in the case of a clinical trial.
lifetable analysis A method of analysis that relies
on a count of the number of events observed and
the time points at which those events occurred,
relative to some zero point. The event may be
death or some other event. In clinical trials, the
time to an event for a patient is usually measured
from the time of enrollment. Treatment effects are
assessed by comparing event rates in the different
treatment groups.
likelihood principle (statistics) A principle that im
plies that the magnitude of the probability asso
ciated with a given outcome of an experiment
under hypothesis A relative to the magnitude of the
probability associated with that outcome under hy
pothesis B contains all the information provided by
the data from the experiment in choosing between
the two hypotheses (Cornfield, 1966b; Dupont,
1983a).
log rank test statistic A test statistic used to com
pare the distribution of event times among differ
ent groups (usually with some censoring) in a clini
cal trial. See Mantel-Haenszel test statistic and
Chapter 18.
loss to follow-up Any loss of follow-up data on a
study patient after enrollment into a trial. The loss
may occur because of the patient’s refusal or inabil
ity to return to the study clinic for follow-up data
collection visits, or because of the inability of clinic
staff to locate the patient for collection of informa
tion not requiring a clinic visit.
losses to follow-up The sum total of information
lost because of loss to follow-up.
lost to follow-up A patient who can no longer be
followed for the outcome of interest, e.g., a patient
who is unwilling or unable to return to the clinic
for follow-up examinations in the case of a clinical
trial using an outcome measured at the clinic, or a
patient who cannot be located for subsequent fol
low-up in the case of a trial involving mortality or
some other outcome that can be measured outside
the clinic setting.
lost to mortality follow-up A person whose vital
status cannot be determined, either because the
person cannot be traced or because of insufficient
identifiers to query data files such as the National
Death Index. Losses to mortality follow-up in a
clinical trial arise from patients who drop out and
who cannot be located for subsequent contacts.
LRC Lipid Research Clinics.
LRC-CPPT Lipid Research Clinics-Coronary Pri
mary Prevention Trial.
M
mainline paper Paper detailing the design, meth
ods, or baseline results of the trial or containing
original results related to the primary objective of
the trial and written by study personnel commis
sioned by the investigative group or their repre
sentative.
major committee Key committee.
management trial Term used by some to character
ize a trial that is designed primarily to provide
information on the value of a treatment in normal
usage (see Schwartz and Lellouch, 1967; Sackett
and Gent, 1979; Sackett, 1980). Term not used in
this book. See also explanatory trial.
Mantel-Haenszel test statistic A test statistic devel
oped by Mantel and Haenszel (1959) to test for the
equality of proportions in two groups over a series
of independent 2X2 tables. In the case of compar
ing the probability of failure at different points in
time, the statistic is equivalent to the log rank test
statistic (see Chapter 18).
manual of operations Study manual of operations.
mask A condition imposed on an individual (or
group of individuals) for the purpose of keeping
that individual or group of individuals from know
ing or learning of some fact or observation, such as
treatment assignment. (Usage note: Term used in
place of blind in this book. See entry for blind for
reasons.)
masked The condition of having a mask in place,
e.g., as in a single-, double-, or triple-masked trial.
matching placebo A pill (capsule or tablet) that is
designed to resemble in shape, texture, size, taste,
etc., a therapeutically active drug and that is used
as the control treatment.
MATHFILE Mathematical Reviews (a literature
database; see Chapter 26).
mean priority score The mean of the priority scores
assigned by individual members of a review group.
medical device A diagnostic or therapeutic contriv
ance that does not interact chemically with a per
son’s body. Includes diagnostic tests, kits, pace
makers, arterial grafts, intraocular lens and
orthopedic pins (Food and Drug Administration,
1983).
medical liaison office Project office.
medical liaison officer Project officer.
medical research associate Clinical research asso
ciate.
MEDLARS Medical Literature Analysis Retrieval
System.
MEDLINE Medical Literature Analysis Retrieval
System on Line.
MeSH Medical subject heading.
MI Myocardial infarction.
MILIS Multicenter Investigation for Limiting In
farct Size.
monitoring Ongoing evaluation of a continuing pro
cess to determine when and if changes in that pro
cess are necessary for reasons of efficiency, data
quality, safety, etc.
Monte Carlo simulation A method of simulating
some stocastic process or procedure using random
or pseudo-random numbers.
MPS Macular Photocoagulation Study.
MRFIT Multiple Risk Factor Intervention Trial,
multicenter Having more than one center.
multicenter clinical trial I. A clinical trial involving
two or more clinical centers, a common study pro
tocol, and a data center, data coordinating center,
or coordinating center to receive, process, and ana
lyze study data. 2. A clinical trial involving two or
more clinics. 3. A clinical trial involving at least
one clinical center and one or more resource cen
ters. (Usage note: A trial, to qualify as multicenter
in this book, must satisfy definition I. Trials simply
involving two or more clinical centers, as specified
in definition 2, do not qualify as multicenter unless
294
I
h
O
A. Glossary
NIGMS National Institute of General Medical
Sciences (part of the NIH).
NIH National Institutes of Health.
NINCDS National Institute of Neurological and
Communicative Disorders and Stroke (part of the
NIH).
NLM National Library of Medicine (part of the
NIH).
noncompliance Not in compliance with a desig
nated procedure. Usually in reference to some treat
ment or data collection procedure in this book.
noncompliant 1. The absence of a compliant state
in relation to a designated procedure. 2. Term used
to describe a patient who is unable or unwilling to
follow the assigned treatment regimen.
noncrossover design A design for a clinical trial in
which a patient is assigned to receive only one of
the study treatments. See also crossover design.
nonfactorial treatment structure A treatment struc
ture that has no factorial structuring.
nonhealth professional A member of a committee,
usually the advisory-review or advisory-review and
treatment effects monitoring committee, chosen
for expertise in an area outside the health field
(e.g., philosophy, theology, law).
nonmasked clinical trial A clinical trial that does
not involve any treatment masking.
nonmasked trial Nonmasked clinical trial.
nonrandom Any method that does not conform to
the statistical definition of random. Used primarily
in this book in contexts where there is a need to
emphasize the nonrandom nature of a haphazard
or systematic process.
nonrandom clinical trial A clinical trial that uses a
nonrandom method of treatment assignment.
nonrandom trial Nonrandom clinical trial.
nonrequired follow-up visit Any visit by the patient
to the clinic after the randomization visit that is not
part of the required sequence of follow-up visits.
The visit may be initiated by the patient or by study
personnel, and includes interim follow-up visits,
nonrequired post-close-out follow-up visit, as well
as post-trial follow-up visits. Data generated at
such visits are not generally used to satisfy data
collection needs for required follow-up visits, un
less they take place within the time windows for
those visits and all necessary procedures are carried
out during the visits.
nonsequential design A design that does not involve
a sequential design. Fixed sample size design.
nonuniform treatment allocation A treatment allo
cation scheme in which the assignment probabili
ties for the various study treatments differ.
Notice of Claimed Investigational Exemption for
New Drug A notice filed with the Food and Drug
Administration by a drug sponsor or independent
investigator requesting permission to test a new
drug, or an existing one for a new indication, in
humans. See Investigational New Drug Applica
tion and phase I. H. HI, and IV trials.
outcome measures, each of which is used or is to be
they have a common study protocol and a center to
used to make treatment comparisons.
receive and process data from the study. A trial
involving a single clinic, whether or not supported
by other resource centers, is classified as a single
center trial in this book.)
N
multicenter trial Multicenter clinical trial.
National Death Index (ND1) A central registry of
multiple comparisons In this book, a term used to
deaths, started in 1979 and operated by the Na
refer to the fact that two or more treatment com
tional Center for Health Statistics of the United
parisons, each involving the same outcome mea
States Public Health Services (see reference cita
sure, are made or are to be made at a designated
time point in the course of the trial. The compari
tion 345).
National Institutes of Health (N1H) A group of
son may involve all members of the treatment
institutes and related support structures located in
groups or subsets (e.g., as Jn analyses involving
Bethesda, Maryland, that is part of the United
subgroups of patients defined by the presence or
States Public Health Service. Responsible for fund
absence of some baseline characteristic.)
ing basic and applied research in the health field.
multiple linear regression analysis (statistics) A
Also initiates and carries out medical research on
method of data analysis using a multiple linear
an intramural and extramural basis.
regression model. Often used in clinical trials to
natural history of disease 1. The course of a disease
adjust treatment results for differences in the base
when left untreated. 2. The course of a disease
line composition of the treatment groups.
when treated with standard modes of therapy.
multiple linear regression model (statistics) A math
natural history study A prospectivefollow-up study
ematical model in which the outcome variable, y,
designed to yield information on the natural course
for the ith patient is written as a function of a series
of a disease or condition. Such studies generally
of independent observations, X|f-..... Parame'
focus on the control-treated group in a clinical trial
ters, 00,
and an error term, e,. The usual
(especially one in which the control treatment is a
form of the model, when no interaction terms are
placebo or standard medical care).
required, is:
NCOS National Cooperative Dialysis Study.
y/ = /So + 01 Xu + 02 x2i +■■■+ PkXki +
NCGS National Cooperative Gallstone Study.
The model derives its name from the fact that all
NCI National Cancer Institute (part of the NIH).
parameters enter as linear terms (i.e., all raised to
NCR No carbon required (a type of paper).
unit power). The independent variables, Xjj, for
NDA New Drug Application.
j = i, ..., k, in the clinical trial setting are usually
NDI National Death Index.
baseline characteristics. The outcome variable may
negative control Inactive control treatment.
be a continuous measure or a binary outcome, such
negative control treatment Inactive control treattreat
as life-death.
multiple logistic regression analysis (statistics) A
ment.
NEI National Eye Institute (part of the NIH).
method of data analysis using a multiple logistic
New Drug Application (NDA) An application sub
regression model. Often used in clinical trials to
mitted by the manufacturer of a drug to the Food
adjust observed treatment results for differences in
and Drug Administration for a license to market
the baseline composition of the treatment groups.
the drug for a specified indication (see Pre Market
multiple logistic regression model (statistics) A
Approval Application for corresponding term for
mathematical model in which an outcome variable,
medical devices).
y; for the ith patient, is written as:
NHLB1 National Heart, Lung, and Blood Institute
y,= l/(l + eT')
(part of the NIH and previously the National Heart
where e is the natural constant, the quality r,, is a
function of a series of observations, X^, ■■■ , Xki,
Institute).
NIADDK National Institute of Arthritis, Diabetes,
made on the ith patient and that are independent of
and Digestive and Kidney Diseases (part of the
y,, the model parameters, and the error term e/. The
NIH and previously the National Institute of Ar
usual form for ry, when no interaction terms are
thritis, Metabolism, and Digestive Diseases).
required, is:
NIAID
National Institute of Allergy and Infectious
Tj = 0o + 01 *li + 02 Xli + - + Wki +
Diseases (part of the NIH).
The model is especially well suited for analyses of
NIAMDD National Institute of Arthritis, Metabo
event data since probability estimates derived from
lism, and Digestive Diseases (now the National
it lie between 0 and I.
Institute of Arthritis, Diabetes, and Digestive and
multiple looks In this book, a term used to refer to
Kidney Diseases).
the fact that treatment comparisons are made or
NICHD National Institute of Child Health and
are to be made at various time points over the
Human Development (part of the NIH).
course of a trial.
NIDR National Institute of Dental Research (part
multiple outcomes In this book, a term used to refer
of the NIH).
to the fact that a trial involves several different
'
if l
I
’
’
■
‘
295
NTIS National Technical Information Service, lo
cated in Springfield, Virginia, and affiliated with
the United States Department of Commerce.
null hypothesis I. (statistics) A hypothesis that
postulates no underlying difference in the popula
tions or groups being compared with regard to the
factor, trait, characteristic, or condition of interest.
2. Null treatment hypothesis.
null treatment hypothesis A hypothesis that states
that the true underlying effect of the test treatment,
as expressed by a specified outcome measure, is no
more or less than for the control treatment.
number adaptive allocation Adaptive allocation
using the difference in the number of patients as
signed to the various treatment groups as the basis
for adapting the treatment allocation ratio.
number adaptive random allocation Adaptive ran
dom allocation using the difference in the number
of patients assigned to the various treatment
groups as the basis for change of the treatment
allocation ratio.
number adaptive randomization A treatment assign
ment process using number adaptive random allo
cation.
--------------------------------------------------------------
0
observation variable A condition or characteristic
associated with individual patients (e.g., age, his
tory of myocardial infarction, blood glucose level)
that may assume different values and that is ob
served and recorded at one or more time points
over the course of data collection.
observational unit An identifiable unit, always a pa
tient in this book but may be a collection of indi
viduals in other contexts (e.g., as characterized by
household members, a hospital ward, or an entire
community), that forms the basis for data collec
tion and analyses. Usually synonymous with treat
ment assignment unit in a clinical trial.
observed allocation ratio The actual allocation
ratio in a completed trial.
observed effective sample size The observed sample
size after reduction due to dropouts and treatment
noncompliance.
observed power The actual power for detecting a
specified treatment difference, given an observed
sample size, observed outcome event rate, and ob
served losses to follow-up due to dropout and noncompliance.
observed sample size The number of patients en
rolled in a study.
observed treatment difference The actual treatment
difference observed either at the end of the trial or
at some designated time point during the trial.
offeror The party or individual who offers or pro
poses to carry out a designated research project,
normally indicated via submission of a formal pro
posal to the sponsoring agency. Term normally re
served in the National Institutes of Health lan-
296
P 297
A. Glossary
guage for proposals received in response to a
request for proposal (RFP) and funded via the
contract mechanism. See applicant and proposer.
officers of the study In this book, generally taken as
the chairman and vice-chairman of the steering
committee, the director of the data center, data
coordinating center, or coordinating center, and
project officer.
OMB Office of Management and Budget (office in
the executive arm of the United States govern
ment).
one-sided alternative hypothesis One-tailed alterna
tive hypothesis.
one-sided test (statistics) One-tailed test.
one-tailed alternative hypothesis An alternative to
the null hypothesis that specifies a range of permis
sible values of all which lie to one side of the null
value (e.g., Ho'p.\ = ^2 versus
> pfl- See
also two-tailed alternative hypothesis.
one-tailed test (statistics) A statistical test ofsignifi
cance based on the null value of no difference
versus the set of all alternative values that are either
to the right or to the left of the null value (e.g., the
set indicating a positive treatment effect in a clini
cal trial. See also two-tailed test.
open clinical trial I. A clinical trial in which a
study physician or study patient decides on the
treatment to be administered. Nonrandom clinical
trial. 2. A nonmasked clinical trial. 3. A clinical
trial with an open sequential design. (Usage note:
Term not used in this book. Trials satisfying defini
tion I are referred to as nonrandom trials. Trials
satisfying defintion 2 are referred to as nonmasked
trials.)
open label trial 1. Nonmasked drug trial. 2. Any
nonmasked trial. (Usage note: The term open label
not used in this book because of potential for con
fusion, e.g., with open clinical trial, and because
nonmasked is considered to be more descriptive.)
open sequential design A sequential design in which
enrollment of patients continues until the test treat
ment is shown, in a statistical sense, to be either
better or worse than the control treatment. Distinct
from a closed sequential design, which allows for
termination of enrollment after observation of a
specified number of outcomes, even if it is not
possible to draw a conclusion for or against the test
treatment.
open sequential trial A trial with an open sequential
design.
operations committee The term used in Veterans
Administration sponsored multicenter trials to des
ignate the standing committee that performs the
functions of the treatment effects monitoring com
mittee and some of the functions of the advisory
review committee.
outcome I. A result, condition, or event associated
with individual study patients that is used to assess
the efficacy of the study treatments. 2. Primary or
secondary outcome, event. 3. An observed event in
a particular patient.
outcome adaptive allocation Adaptive allocation
based on outcomes observed for enrolled patients.
outcome adaptive random allocation Adaptive ran
dom allocation based on outcomes observed for
enrolled patients.
outcome adaptive randomization A treatment as
signment process using outcome adaptive random
allocation.
outcome event The event of primary interest in a
trial, e.g., the one used for sample size calculations
and for key data analyses in the trial.
outcome measure An observation variable recorded
for patients in the trial at one or more time points
after enrollment for the purpose of assessing the
effects of the study treatments. See outcome vari
able.
outcome variable An observation variable recorded
for patients in the trial at one or more time points
after enrollment for the purpose of assessing the
effects of the study treatments. See outcome mea
sure.
outlier Any value, reading, or measurement that is
outside established limits and, for this reason, is
questioned or considered to be in error.
P
p-value (statistics) A value associated with an ob
served test statistic that indicates the probability
that a value as extreme or more extreme than the
one observed will arise by chance alone in repeated
replications of a study.
package insert A document approved by the Food
and Drug Administration and furnished by the
manufacturer of a drug for use when dispensing the
drug, which indicates approved uses, contraindica
tions, and potential side effects. See label insert.
PAHO Pan American Health Organization.
PAIS Public Affairs Information Service (a litera
ture database; see Chapter 26).
parallel design Parallel treatment design.
parallel treatment design A term sometimes used
(but not in this book) to refer to treatment designs
involving uncrossed treatments. See noncrossover
design.
parameter I. (statistics) A constant appearing in a
mathematical expression that characterizes some
population, process, or the like, whose true value is
generally unknown but that can be estimated.
2. (clinical medicine) Observation variable. (Usage
note: Not used in the latter context in this book.)
parent center I. The center to which satellite cen
ters report. 2. Lead center.
parent clinic I. The clinic to which satellite clinics
report. 2. Lead clinic.
parent institution 1. The institution that has admin
istrative responsibilities for a specified study cen
ter and associated investigators. 2. The center re
sponsible for disbursing funds to other centers in a
study funded under a consortium agreement.
PARIS Persantine Aspirin Reinfarction Study.
partial factorial structure Partial factorial treat
ment structure.
partial factorial treatment structure A treatment
structure involving some but not all possible com
binations of the treatments used in the trial.
partially masked clinical trial I. A clinical trial in
which some, but not all, of the study treatments are
administered in a single- or double-masked fash
ion. 2. A clinical trial in which some, but not all, of
the staff in a clinic are masked to treatment assign
ment.
participant Study participant.
patient Shorthand for study patient in this book.
From Middle English pacient, from Old French
patient, from Latin patiens, from the present parti
ciple of pati, to suffer.
patient close-out The process of separating patients
from a clinical trial at the end of the treatment and
follow-up stage (see Chapter 3 and Appendix D).
patient close-out stage The stage of a trial in which
patients are separated from the trial at the end of
the treatment and follow-up stage (see Chapter 3
and Appendix D). The fifth stage of a trial in this
book.
patient compliance The degree to which a patient
follows a prescribed set of procedures or routines.
Synonymous with treatment adherence when the
procedures or routines in question are those con
cerned with administration of a patient’s assigned
treatment.
patient enrollment The act of enrolling a patient
(treatment unit) into a trial. In this book, consid
ered to occur when the treatment assignment for
the patient is revealed to clinic staff, or when treat
ment is initiated when assignments are known in
advance of enrollment.
patient enrollment process The process of enrolling
patients into a clinical trial. The process includes
all the examinations and data collection proce
dures associated with the prerandomization and
randomization visits.
patient examination Any examination done to eval
uate a patient to determine eligibility for enroll
ment into a trial or to provide follow-up data.
patient follow-up A process involving periodic con
tact with patients enrolled in a clinical trial for the
purpose of administering the assigned treatments),
observing the effects of treatment(s), modifying the
course of treatment(s), or for collecting required
data.
patient identification number A unique sequence of
numbers, or numbers and letters, that are used to
identify a patient.
patient monitoring Patient safety monitoring.
patient population Study population.
patient recruitment The process of identifying suit
able patients for enrollment into a clinical trial.
patient recruitment goal The number of patients
scheduled to be enrolled into the trial. Usually set
before the trial starts, or shortly thereafter, via a
sample size calculation or via practical considera
tions.
patient recruitment quota A specification, usually
set before patient recruitment is started or shortly
thereafter, that indicates the mix of patients to be
enrolled with regard to some characteristic, trait,
or condition (e.g., the number of males versus fe
males).
patient recruitment stage The stage of a clinical trial
concerned primarily with patient recruitment (see
Chapter 3 and Appendix D). The third stage of a
trial in this book.
patient safety monitoring 1. Any ongoing process
of reviewing accumulated outcome data for groups
of patients in a trial to determine if a designated
treatment procedure should be altered or stopped.
Treatment effects monitoring. 2. The process of
watching for treatment effects in an individual pa
tient (term not ordinarily used in this context in
this book).
payline A term used in connection with National
Institutes of Health grants to indicate the priority
score required on an approved application to per
mit payment. The payline is a function of the
number of approved applications received by an
institute, the distribution of priority scores across
applications, and the amount of money available
for new research initiatives by the institute.
performance monitoring An ongoing process car
ried out over the course of a trial to assess the
performance of some center, group of centers, or
some other task-oriented group in the structure of
a trial.
phase I trial The first stage in testing a new drug in
man. Performed as part of an approved Investiga
tional New Drug Application under Food and
Drug Administration guidelines. The studies are
usually done to generate preliminary information
on the chemical action and safety of the drug using
normal healthy volunteers. Usually done without a
comparison group (see Food and Drug Adminis
tration. 1977c; Pines, 1980).
phase II trial The second stage in testing a new drug
in man. Performed as part of an approved Investi
gational New Drug Application under Food and
Drug Administration guidelines. Generally carried
out on patients with the disease or condition of
interest. The main purpose is to provide prelimi
nary information on treatment efficacy and to sup
plement information on safety obtained from
phase / trials. Usually, but not always, designed to
include a control treatment and random allocation
of patients to treatment (see Food and Drug Ad
ministration, 1977c; Pines, 1980).
1
298
I'-"
i r
11
A. Glossary
phase III trial The third and usually final stage in
testing a new drug in man. Performed as part of an
approved Investigational New Drug Application
under Food and Drug Administration guidelines.
Concerned primarily with assessment of dosage
effects and efficacy and safety. Usually designed to
include a control treatment and random allocation
to treatment. Once this phase is completed the drug
manufacturers may request permission to market
the drug by submission of a New Drug Application
to the Food and Drug Administration, assuming
the results of the phase I. 11 and /// trials are
consistent with such a request (see Food and Drug
Administration, 1977c: Pines, 1980).
phase IV trial Generally, a randomized controlled
trial that is designed to evaluate the long-term
safety and efficacy of a drug for a given indication
and that is done with Food and Drug Administra
tion approval. Usually carried out after licensure of
the drug for that indication (see Food and Drug
Administration, 1977c).
PHS Physicians’ Health Study.
PI Principal investigator.
pilot study A preliminary study designed to indicate
whether a larger study is practical. See feasibility
study.
placebo A pharmacologically inactive agent given
to a patient as a substitute for an active agent and
where the patient is not informed whether he is
receiving the active or inactive agent.
placebo-controlled clinical trial A clinical trial in
which patients assigned to the control treatment
receive a placebo.
placebo effect The effect produced by a placebo.
The effect in placebo-controlled clinical trials is
generally measured by comparison of the effect
observed in patients receiving the placebo treat
ment with the effect observed in patients receiving
the active treatment.
placebo reactor A patient who reports side effects
normally associated with the test treatment while
receiving a placebo.
placebo treatment I. A treatment involving the use
of a placebo. 2. A treatment that is harmless.
play the winner treatment allocation scheme An out
come adaptive allocation scheme based on work of
Robbins (1952, 1956) in which the next treatment
assignment is a function of the success or failure of
the test treatment, as assessed in the last patient
enrolled. A success would cause the next assign
ment to be made to the test treatment. A failure
would cause the next assignment to be made to the
control treatment. Modified by Zelen (1969) so as
not to require complete dependence on the out
come observed in the last patient enrolled. The goal
is to minimize the number of patients assigned to
the inferior treatment.
PMA Pre-Market Approval (PMA) application (for
a new medical device; see New Drug Application
for corresponding term for drugs).
P 299
PO 1. Project office. 2. Project officer.
policy-advisory board Advisory-review committee.
policy board Advisory-review committee.
polychotomous variable A discrete variable that
may assume two or more different values.
portrait-style page orientation A form of orienta
tion in which printed and visual information is
arrayed on the short axis of a page (e.g., as the
pages in this book). See landscape-style page orien
tation for opposing term.
POSCH Program on the Surgical Control of Hy
perlipidemia.
positive control Active control treatment.
positive control treatment Active control treat
ment.
post-close-out follow-up visit I. Any follow-up
visit of a patient that takes place after his separa
tion from the trial, as indicated by completion of
the close-out follow-up visit. 2. Any follow-up visit
which takes place after completion of the close-out
stage of a trial. 3. Post-trial follow-up visit.
post-close-out visit Post-close-out follow-up visit.
post-marketing surveillance Term used by the Food
and Drug Administration to characterize any
procedure, implemented after licensure of a drug
for a given indication, that is designed to provide
information on the actual use of the drug for that
indication and on the occurrence of related side
effects. The surveillance usually involves survey
techniques rather than controlled trials.
post-randomization examination Any patient exam
ination made by clinic personnel during a post
randomization follow-up visit for data collection.
post-randomization follow-up visit Any visit by a
patient to the clinic after the randomization visit,
required as well as nonrequired follow-up visits.
The former class includes treatment application
and adjustment, regular, close-out, and post-close
out follow-up visits.
post-randomization visit Post-randomization follow
up visit.
post-stratification The process of classifying pa
tients into strata after they have been enrolled in
the study— usually for data analysis purposes.
post-treatment follow-up 1. Any patient follow-up
after the first application of the assigned treatment,
especially in clinical trials involving a single appli
cation of the treatment, e.g., as in most surgery
trials. 2. Post-trial follow-up.
post-trial follow-up A term used to refer to any
form of patient follow-up after completion of the
close-out stage of a trial.
post-trial follow-up stage One of the seven stages of
a trial in this book (see Chapter 3 and Appendix
D). Defined as an optional stage that occurs during
or after completion of the termination stage of the
trial and that is designed to provide follow-up data
on mortality or some other outcome measure.
post-trial follow-up visit I. Any follow-up visit that
takes place after the close of the trial, the main
purpose of which is to enable clinic personnel to
collect data on a primary or secondary outcome
measure to assess treatment effects. 2. The same as
I except that the visit may take place any time after
the close-out follow-up visit. See post-close-out fol
low-up visit.
power The probability of rejecting the null hypothe
sis when it is false.
Pre-Market Approval (PMA) application An ap
plication to the Food and Drug Administration for
permission to market a specified medical device
(Food and Drug Administration, 1983). See New
Drug Application for corresponding term for
drugs.
prerandomization examination Any examination
that is part of the evaluation process of a patient
for enrollment into a trial and that is carried out
before the randomization examination.
prerandomization visit Any visit made to the clinic
by a potential study patient for the purpose of
evaluation for enrollment into the trial and that
takes place prior to the randomization visit.
pretreatment examination Any examination done
on a patient before the initiation of treatment and
that is a required part of the procedures for the
trial. Synonymous with prerandomization exami
nation if randomization and initiation of treatment
take place during the same visit.
prevention trial Prophylactic trial.
primary event I. A primary outcome variable that
is binary. 2. The actual occurrence of a primary
outcome.
primary outcome I. The event or condition the
trial is designed to ameliorate, delay, or prevent.
2. The actual occurrence of a primary event in a
study patient.
primary outcome variable The outcome variable
that is designated or regarded as key in the design
or analysis of the results of a trial. Generally, the
variable used for sample size calculations in the
design of the trial or. when no sample size calcula
tion is made, for the main avenue of data analyses.
primary prevention trial A prophylactic trial that
involves patients selected for the absence of a speci
fied disease or condition and a test treatment that
is being used ostensibly to prevent or delay the
onset of that disease or condition.
principal investigator I. The designation used by
the National Institutes of Health to denote the
individual named on a grant application who is
responsible for directing the proposed research.
2. The lead scientist in a research project. (Usage
note: It is best to avoid use of the term to designate
the head of a center in a multicenter trial. It should
be used in such settings only when there is a single
individual, such as the chairman of the study, who
is regarded by everyone in the trial as the principal
investigator. Otherwise some other term, such as
center director, should be used.)
priority score 1. The score assigned to a research
application by an individual member of a review
group that reflects that individual’s judgment re
garding the scientific merit of the proposal. In Na
tional Institutes of Health grant reviews the score
may range from 1 (high scientific merit) to 5 (low
scientific merit). 2. The score assigned to a research
application as computed from the scores assigned
by individual reviewers of the application.
procurement center Procurement and distribution
center.
procurement and distribution center A facility in
the structure of a clinical trial that is responsi
ble for procuring, packaging, and distributing a
needed supply or product (e.g.. drugs, laboratory
supplies, forms) to selected centers in the trial.
program director I. The individual who heads a
research project. 2. Principal investigator. (Usage
note: Term not used in the context of a multicenter
trial in this book. See usage note appearing under
principal investigator for reason.)
program office 1. Project office. 2. An office con
taining a project office for several different but
related studies.
program officer Project officer.
project director I. The individual who heads a re
search project. 2. Principal investigator. (Usage
note: Term not used in the context of a multicenter
trial in this book. See usage note appearing under
principal investigator for reason.)
project office I. The office, located in the sponsor
ing agency and usually staffed with one or more
individuals trained in a medical or research field,
that is responsible for dealing with technical, scien
tific, and programmatic aspects of a grant- or con
tract-funded project. 2. Program office.
project officer I. The individual in the sponsoring
agency who is responsible for dealing with tech
nical, scientific, and programmatic aspects of a
grant or con/ract-funded project. 2. Health scien
tist administrator in National Institutes of Health
grant-funded projects. See also program officer.
prophylactic trial A trial that is designed to assess
the efficacy of a treatment procedure aimed at
preventing the development or progression of a
specific disease or condition.
proposer The party or individual who proposes, nor
mally via submission of a written proposal, to
carry out a designated research project. See also
offeror and applicant.
prospective follow-up study A study in which peo
ple with a specific attribute or characteristic are
identified and then observed for some period of
time thereafter for the occurrence of the outcome
or condition of interest, usually disease or death.
The study may or may not involve a comparison
group. Clinical trials represent a special subset of
prospective follow-up studies.
prospective study Prospective follow-up study.
protocol Study protocol.
protocol development stage The second stage of a
5?
300
clinical trial in this book. Usually undertaken after
the initiation of funding and characterized by work
involving development of the protocol and proce
dures needed to carry out the trial (see Chapter 3
and Appendix D).
pseudo random number A number that has been
generated via a deterministic process, that has,
or appears to have, the properties of a random
number, e.g.. as with some computer-generated
“random" numbers (see Knuth, 1969).
PSRO Professional Standards Review Organiza
tion.
Q
quality assurance Any procedure, method, or phiphi
losophy for collecting, processing, or analyzing
data that is aimed at maintaining or improving the
reliability or validity of the data and the associated
procedures used to generate them.
quality control center One of the possible resource
centers in a trial. Defined as the center with respon
sibility for quality assurance for one or more as
pects of the data collection or analysis processes.
(Usage note: Term not used in this book except
where there is a specified center with designated
quality control functions that are over and above
those normally assumed by the data center, data
coordinating center, or coordinating center.)
quasi random number Pseudo random number.
R
l,h'”
S 301
A. Glossary
R and D Research and development.
RAI Research Award Index (a database of research
funded by the United States Public Health Service;
see Chapter 26).
random (general) I. Having no specific pattern or
objective. Of or designating a chance process in
which the occurrence of previous events is of no
value in predicting future events. From Old
French, random, meaning force, violence, impetu
osity. 2. Sometimes used as a synonym for hap
hazard, but never in this book. Usage in this book
always refers to a formal process meeting, or be
lieved to meet, the conditions specified under the
statistical definition of random.
random 1. (statistics) A term used to refer to a
sequence of observations, activities, assignments,
etc., that is the result of a chance process in which
the probability of any given sequence is known or
can be determined. 2. The term used to refer to a
process that meets the probability conditions out
lined above.
random allocation A method for assigning patients
to treatment using a random process.
random number A number generated or drawn via
some defined random process.
random process Any method or procedure that
yields output that has the defined mathematical
properties of a random variable.
random variable A variable that may assume any
one of a number of different values, where the set
of possible values is determined by a probability
distribution, such as Bernoulli or normal.
randomization I. The process of assigning patients
(treatment units) to treatment using a random pro
cess, such as via use of a table of random numbers.
2. The process of deriving an order or sequence of
items, determinations, specimens, readings, or the
like using a random process.
randomization examination A patient examination
that is done during the randomization visit.
randomization unit Treatment assignment unit.
Usually patient, but may be a larger unit, as in
studies involving families, hospital wards, or the
like.
randomization visit The clinic visit at which the
patient is randomized.
randomized The condition of having been assigned
to a treatment via a random process. Normally
considered to have occurred when the treatment
assignment is revealed to any member of the clinic
staff, e.g., when the envelope containing the treat
ment is opened at the clinic.
randomized control clinical trial Term sometimes
used (e.g., Chalmers et al., 1981) to emphasize the
nature of the randomization process in relation to
the control treatment (i.e., that patients are ran
domly assigned to the control treatment). (Usage
note: Phrase not used in this book. The preferred
term is randomized controlled clinical trial).
randomized control trial Randomized controlled
clinical trial. (See randomized control clinical trial
for usage note.)
randomized controlled clinical trial A clinical trial
(always in man in this book) that involves at least
one test treatment and one control treatment, con
current enrollment and follow-up of the test- and
control-treated groups, and in which the treat
ments to be administered are selected by a random
process, such that neither the patients nor the per
sons responsible for their selection or treatment
can influence the assignments, and where the as
signments remain unknown to the patients and
clinic staff until the patients have been determined
to be eligible for enrollment into the trial (and then
may be revealed to patients and clinic personnel
only by letter or number codes in masked trials).
randomized controlled trial Randomized con
trolled clinical trial.
raw data I. Measurements and observations re
corded on study data forms. 2. Unedited com
puter-generated listings of data from study data
forms, prior to use of reduction and summary
procedures needed for data analysis.
RBO Relative betting odds.
RCT I. Randomized clinical trial. 2. Randomized
controlled trial.
reading center A center that is responsible for inter-
i
,
preting and codifying information from a specified
set of materials, records, or documents (e.g.,
ECGs, fundus photographs, chest X rays, biopsy or
autopsy specimens, death certificates) provided by
the clinical centertf) in the study.
record (n). A paper or electronic document that
contains or is designed to contain a set of facts
related to some occurrence, transaction, or the like.
recruitment goal Patient recruitment goal.
recruitment log A log maintained by a clinic that
lists each patient considered for enrollment into a
study. Usually maintained to provide a description
of the characteristics of the population screened for
enrollment. See screening log.
recruitment quota Patient recruitment quota.
regression to the mean A phenomenon that occurs
when a second determination or measurement is
made only on those individuals with an extreme
initial determination or measurement. On average,
the second determination or measurement tends to
be less extreme than the initial one. Term originally
coined by Sir Francis Gallon (1886) to characterize
the tendency for tall parents to produce shorter
offspring and vice versa.
regular follow-up visit A required follow-up visit,
the main purpose of which is to enable clinic per
sonnel to carry out treatment assessment and data
collection procedures, as specified in the study pro
tocol. Called regular because such visits are normally required at fixed periods over the course of
follow-up. Does not include visits done simply for
treatment application or treatment adjustment.
relative betting odds A method of analysis devel
oped by Cornfield (1966b) involving the ratio of
two likelihood functions computed under the null
hypothesis and a specified alternative.
request for application (RFA) A document pre
pared and distributed by a sponsoring agency to
solicit applications for grant support to perform
work described in the request.
request for proposal (RFP) A document prepared
and distributed by a sponsoring agency to solicit
proposals for execution of specific work. Normally
used in conjunction with contract funding.
required follow-up visit Any follow-up visit that is a
required part of the study protocol and that is to be
done at a specified time after the randomization
visit. Visits include treatment application and ad
justment, regular, close-out, and post-close-out,
and post-trial foltow-up visits.
research group Investigative group.
resource center Any center, other than a clinical
center, identified in the structure of a trial, that is
involved in performing a specified set of support
functions. The term includes data center, data coor
dinating center, treatment coordinating center,
coordinating center, central laboratory, reading
center, quality control center, project office, and
procurement and distribution center.
response variable Outcome variable.
restricted allocation scheme Any allocation scheme
m-iso
in which the treatment allocation schedule is de
signed to satisfy certain preset constraints, as in
blocking in a fixed allocation schedule.
restricted random allocation Restricted allocation
scheme involving use of a random process to make
treatment assignments.
restricted randomization The process of generating
or issuing treatment assignments via a restricted
random allocation treatment schedule.
restricted treatment allocation schedule A treat
ment allocation schedule that is constrained to
yield the expected allocation ratio, as with block
ing in a fixed allocation schedule.
review group A group of individuals, normally re
cruited by the sponsoring agency or its representa
tive, charged with the review of a specific research
proposal or set of research proposals for scientific
merit, study section.
RFA Request for application.
RFP Request for proposal.
risk factor Any environmental exposure, personal
characteristic, or event that affects the probability
of developing a given disease or experiencing a
change in health status (Morgenstern and Bursic,
1982).
risk factor analysis (epidemiology) Any analysis,
usually involving regression or subgroup analyses,
that is aimed at identifying risk factors for a given
disease or condition.
RJE Remote job entry (via a computer terminal).
routine follow-up visit Regular follow-up visit.
S
safety committee Treatment effects monitoring com
mittee.
safety monitoring Treatment effects monitoring.
safety monitoring committee Treatment effects mon
itoring committee.
sample size I. The actual number of patients en
rolled in a study. 2. The anticipated number of
patients to be enrolled in a study, or the patient
recruitment goal.
sample size calculation A mathematical calculation,
usually carried out when a trial is planned, that
indicates the number of patients to be enrolled in
order to provide a specified degree of statistical
precision for a specified type I and type II error
protection (see Chapter 9).
sample size requirement The sample size yielded by
a sample size calculation. See recruitment goal.
SAS Statistical Analysis System (a package of data
analysis programs).
satellite center A center that is subservient to the
parent center and that is organized to perform a
designated set of functions considered to be part of
the workscope of the parent center.
satellite clinic A clinic that is subservient to the
parent clinic and that is organized and operated to
302
S
A. Glossary
screen, identify, enroll, treat, or follow a segment
of the study population that cannot, for matters of
convenience or other reasons, be seen at the parent
clinic.
SAW Show as written.
SC Steering committee.
scheduled follow-up visit Required follow-up visit.
SCI Science Citation Index.
screening log Recruitment log.
secondary event 1. A secondary outcome variable
that is binary. 2. The actual occurrence of a sec
ondary outcome.
secondary outcome I. An event or condition re
lated to the primary outcome but of less clinical or
medical importance than the primary outcome.
2. The actual occurrence of a secondary event in a
study patient.
secondary outcome variable An outcome variable
that is known or believed to be related to the
primary outcome variable and that is used, in addi
tion to the primary outcome variable, for evalua
tion of treatments in the trial (e.g., observation of
patients for the occurrence of nonfatal myocardial
infarctions in a clinical trial using death as the
primary outcome measure). 2. Any other outcome
variable, regardless of its relationship to the pri
mary outcome variable, that is used for treatment
evaluation.
secondary paper Paper dealing with a secondary
objective of the trial and written by study personnel
commissioned by the investigative group or their
representative.
secondary prevention trial A prevention trial in
volving patients with a history of some disease or
condition in which the test treatment is adminis
tered to prevent or delay further development or
progression of that disease or condition. For exam
ple, a drug trial involving use of a daily dose of
aspirin over a period of years for prevention of
myocardial infarction in patients with a prior his
tory of myocardial infarction.
secular trend A trend or pattern that is time related;
temporal trend.
self-checking digit Check digit.
sequential analysis I. The analysis done after en
rollment of a patient, pair of patients, or larger
block of patients, in a sequential trial to determine
whether additional patients should be enrolled.
The decision is made by observing the test-control
difference in observed outcomes. Enrollment of the
next patient, pair of patients, or block of patients is
carried out if the difference does not exceed pre
specified boundary limits. 2. Periodic analyses car
ried out for treatment monitoring in trials with
fixed sample size designs. (Usage note: Use in the
context of definition 2 is avoided in this book. See
interim analysis for preferred term.)
sequential design Any design in which the decision
as to whether to enroll the next patient, pair of
patients, or block of patients is determined by
whether the cumulative treatment difference for all
previous patients is within specified limits. Enroll
ment is continued if the difference does not exceed
the limits. It is terminated if it does.
sequential trial I. A trial involving a sequential de
sign. 2. Term sometimes used (but not in this
book) in conjunction with a fixed sample size de
sign in which decisions concerning the enrollment
of additional patients, or the continued treatment
and observation of patients already enrolled, is
dependent on accumulated data in the trial.
sham Something false presented to be genuine; a
spurious imitation. Derived from the word shame,
meaning trick or fraud.
sham procedure A procedure designed to resemble
the real one and that is performed on a patient for
the purpose of masking the patient or the patient’s
study physician as to whether the patient has re
ceived the real procedure.
side effect A secondary by-product of an action or
procedure. Usually treatment side effect in this
book.
significance level (statistics) I. The permissible type
I error level for a test of the null hypothesis with a
specified test statistic. The null hypothesis is ac
cepted if the test statistic yields a p-value which is
larger than the specified level and is rejected if it is
equal to or less than this value. 2. p-value.
significance test Test of significance.
significance testing (statistics) The act of carrying
out a test of significance.
simple randomization Unrestricted randomization.
simple treatment structure Nonfactorial treatment
structure.
single-blind Single-masked in this book. See blind
for usage comment.
single-blinded Single-masked in this book. See
blind for usage comment.
single-blinded clinical trial Single-masked clinical
trial.
single-center trial 1. A clinical trial involving only
one clinic (with or without satellite clinics) and no
other resource center. 2. A trial involving only one
clinic and a center to receive and process data.
3. A trial involving only one clinic and one or
more resource centers. 4. A trial with no clinical
centers, but one or more resource centers, as in the
Physicians' Health Study sketched in Appendix B.
5. A clinical trial involving two or more clinical
centers, but no center to receive and process study
data. 6. A trial with multiple clinics not having a
common study protocol (see usage note under mul
ticenter clinical trial).
single-mask Single-masked.
single-masked A condition where certain persons
(e.g., study physicians) are informed of some fact
or condition whereas other persons (e.g., patients)
are purposefully denied information regarding that
fact or condition.
single-masked clinical trial I. A clinical trial in
303
change in treatment or terminating follow-up of
which treatments are administered in such a
that patient.
manner that patients in the trial are not informed
stop item An item or response category on a data
of whether they have been assigned to the test or
form that when used or checked indicates the pres
control treatment, but clinic staff are. 2. A clinical
ence of a stop condition (see Chapter 12).
trial in which the patient knows the treatment as
stopping boundary The set of values formed by a
signed, but the treating physician, examiner, or
line or set of lines (or curves), usually specified
observer does not. (Usage note: Term not used in
before or shortly after the start of patient recruit
the context of definition 2 in this book.)
ment, which, if exceeded, indicates the existence of
site visit A visit to a center or prospective center in a
a test-control treatment difference that satisfies cer
trial by personnel from outside that center for the
tain statistical properties (e.g., has zp-value of less
purpose of assessing its performance or perfor
than a certain size). The boundaries will be used as
mance potential in the trial.
a basis for stopping the trial when developed in
soft endpoint Soft outcome. (See usage note for
conjunction with a sequential design, but not nec
endpoint.)
essarily when used in conjunction with a fixed sam
soft outcome Any outcome measure that is subject
ple size design.
to major errors of interpretation or measurement.
stopping rule A rule, usually set before or shortly
Usually, a measurement or assessment that de
after the start of patient recruitment, that specifies
pends on clinical judgment.
a limit for the observed test-control treatment dif
software Computer programs and related manuals
ference for the primary outcome, which, if ex
and documents needed to operate them.
ceeded, automatically leads to termination of the
specified allocation ratio The particular allocation
test or control treatment, depending on the direc
ratio used in constructing the allocation schedule.
tion of the observed difference.
sponsor Sponsoring agency.
strata (pl. of stratum) A series of distinct levels or
sponsoring agency The institution, organization, or
layers. In this book, generally subgroups of pa
foundation that provides fiscal support, and often
tients formed by classification on some variable or
administrative and scientific support as well, for a
set of variables, usually baseline variables.
given project. See funding agency.
stratification I. The process of classifying observa
SPSS Statistical Package for the Social Sciences.
tion units into strata. 2. The process of classifying
SSIE Smithsonian Scientific Information Ex
patients into strata as part of the randomization
change.
process or for purposes of data analysis.
stages of a clinical trial An arbitrary classification
stratification variable A variable used to classify the
to characterize the stages of a trial. The stages used
observational units into strata.
in this book are: Initial design, protocol develop
stratified allocation A method of treatment assign
ment, patient recruitment, treatment and follow
ment in which patients are first classified into de
up, patient close-out, termination, and (optional)
fined subgroups based on one or more baseline
post-trial follow-up (see Chapter 3 and Appendix
variables and then assigned to treatment within the
D).
defined subgroups.
standard treatment The accepted mode of treat
stratified random allocation Random allocation
ment for a given disease or condition. Equivalent
within defined allocation strata.
to the control treatment in clinical trials when
stratified randomization A treatment assignment
chosen to mimic standard medical practice.
process using stratified random allocation.
statistical significance (statistics) p-value.
stratum (sing, of strata) A layer, level, or defined
steering committee (SC) I. A committee responsi
subgroup.
ble for directing the activities of a designated proj
study I. A general term used to refer to any one of
ect. 2. One of the key committees in the organiza
a variety of research activities involving the collec
tional structure of a multicenter clinical trial.
tion, analysis, or interpretation of data. 2. Often
Committee responsible for conduct of the trial and
used in this book as a synonym for clinical trial.
to which all other committees report, except the
3. A project involving multiple types of investiga
treatment effects monitoring committee and advi
tions, only one of which is a clinical trial (e g., as in
sory-review committee or advisory-review and
the Coronary Artery Surgery Study since it in
treatment effects monitoring committee.
cludes both a clinical trial and an uncontrolled
stop condition I. A condition encountered when
prospective follow-up study.
carrying out a procedure (e.g., completing a data
study chairman Chairman of the steering commit
form, performing a patient examination) that re
tee.
quires the person performing the procedure to ter
study clinic A facility with defined responsibilities
minate the procedure until or unless the condition
for recruiting, enrolling, treating, and following
can be removed. 2. A defined condition that, when
patients in a trial.
encountered for a patient enrolled in a trial, re
study database The entire set of data, whether or
quires or permits clinic personnel to take some
not codified and keyed for storage in a computer.
action related to that patient, such as instituting a
9
304
ijin ii
T
A. Glossary
collected on study patients, as contained on official
data forms of the trial. Note: Data contained in
patients’ charts, unless transcribed onto official
study data forms, are not considered part of the
study database.
study group I. Any defined group of patients on
whom specified data are collected. 2. The entire
group of patients included in a study. 3. Often
synonymous with treatment group, as used in this
book. 4. The group of investigators carrying out a
study, especially a multicenter study. (Usage note:
Term not used in the sense of definition 4 in this
book.)
study handbook A book that contains a series of
tables, charts, figures, and specification pages that
detail the main design and operating features of a
study, largely without use of written narrative.
study investigator General term used in this book to
designate any individual who has a key role in the
design, conduct, or analysis of a study.
study manual of operations A document or collec
tion of documents that describes the procedures
used in a center or set of centers in a clinical trial
(e.g., manual of operations for study clinics, coor
dinating center manual of operations, ECG read
ing center manual of operations).
study participant I. A term sometimes used in
place of study patient when there is a desire to
avoid the connotation of illness, as in trials involv
ing well people. 2. Study investigator (but not in
this book).
study patient Term used in this book to characterize
an individual considered for enrollment or actually
enrolled into a trial regardless of whether or not
there is a perceived need for medical care. See
study subject and study participant.
study physician Any physician associated with a
study clinic who is responsible for administering
the study treatments to patients in the trial or who
is responsible for patient care, as dictated by the
study protocol.
study population I. The set of patients enrolled in
a trial. 2. The entire set of patients considered for
enrollment into the trial (not used in this context in
this book).
study protocol A narrative document that describes
the general design and operating features of a trial.
Distinguished from the study manual of operations
by its generality and absence of specific details
needed for day-to-day execution of the trial.
study section I. Any review group of the National
Institutes of Health, especially one that is chartered
to carry out reviews of research applications in a
general area of research and that meets at regular
intervals during the calendar year to perform those
reviews. 2. A group of individuals, normally re
cruited by the sponsoring agency as its representa
tives, charged with the review of a specific research
proposal or set of research proposals to assess scien
tific merit. See review group.
study subject General term used to denote an indi-
vidual enrolled in a study. Not used in this book.
The advantage of the term, as opposed to study
patient, is that it avoids the connotation of ill
ness-useful in cases where well people are being
studied. The disadvantage is that it carries a conno
tation of subjugation—a notion that is at variance
with the concept of informed consent and a pa
tient-investigator partnership.
study treatment General term used throughout this
book to refer to either a test or control treatment.
study vice-chairman The individual elected or de
signed to perform the functions of study chairman
in his absence.
subgroup A subpart of the study population distin
guished by a particular characteristic or set of char
acteristics (e.g., males under age 45 at entry).
subgroup analysis Any data analysis that focuses
on a selected subgroup or patients. Generally in
this book, any analysis that is aimed at elucidating
treatment differences within a defined subgroup of
patients.
subgrouping cut-point The value of a subgrouping
variable used to separate patients (treatment units)
into subgroups. For example, formation of sub
groups of patients less than 35 years of age, 35
through 54 years of age, and 55 years of age or
older requires use of cut-points at 35 and 55 years
of age.
subgrouping variable A variable, such as age, used
to classify patients (treatment units) into sub
groups. Usually a baseline characteristic for most
subgroup analyses in clinical trials.
subject Study subject.
support center Resource center.
surgical trial A trial in which the test treatment is a
surgical procedure.
surrogate outcome An outcome, based on some test
or measurement, that is used instead of a clinical
event in the design or analysis of clinical trial.
surrogate outcome variable A test, measurement,
score, or some other similar variable that is used in
place of a clinical event (e.g., use of blood pressure
change in place of clinical hypertension) in the
design of a trial, or in summarizing results from it.
Used because the variable is believed to be corre
lated with the clinical event of interest and because
of its perceived utility in yielding detectable treat
ment differences.
survival analysis 1. Any method of data analysis
that focuses on the length of survival of the obser
vational units. 2. Lifetable analysis.
T
technical group Investigative group.
TEMC Treatment effects monitoring committee.
terminal Computer terminal.
termination stage The sixth and usually last stage of
a clinical trial as used in this book. Concerned
primarily with analysis of the study results (see
Chapter 3 and Appendix D).
test-control difference Test-control treatment differ
ence.
test-control treatment difference The postulated or
observed difference between the test- and controltreated groups of patients with regard to a specified
outcome measure.
test group A group of patients defined by the study
design—patients assigned to the test treatment in a
clinical trial—who are contrasted with the control
group of patients to reach a conclusion regarding
some factor, condition, or treatment.
test of significance (statistics) I. The evaluation of
observed data by calculating a specified test statis
tic and then deriving the associated p-value. 2. Test
statistic.
test statistic I. The formula or computing algo
rithm used to carry out a test of significance.
2. The numerical value provided by the formula or
computing algorithm for a specified test of signifi
cance using a defined data set.
test-treated group 1. The group of patients as
signed to the test treatment. 2. The group of pa
tients who receive the test treatment. (Usage note:
Use of the term in this book is always from the
point of view of the assignment process, regardless
of the treatment actually administered.)
test treatment The drug, device, or procedure to be
evaluated in a particular trial.
therapeutic trial A trial designed to test the safety
and efficacy of a particular drug, device, or proce
dure that is considered to have therapeutic value.
throwaway medical journal A pejorative term used
to characterize a medical periodical that is distrib
uted by a profit-making firm to a segment of the
medical community free of charge.
time of enrollment The time point at which a pa
tient (treatment unit) is regarded as having offi
cially entered the trial and after which is regarded
as a part of the study population. Operationally,
the time point at which the treatment assignment is
revealed to clinic staff, or when treatment is in
itiated when assignments are known in advance of
enrollment.
time window The permissible time interval for per
forming a specified baseline or follow-up examina
tion.
toxic drug reaction An adverse drug reaction that
results in morbidity or mortality.
toxic side effect An adverse side effect that results
in morbidity or mortality.
treatment I. The act of treating, as in caring for a
patient. 2. The specific regimen, method, or proce
dure being tested in a clinical trial.
treatment adjustment follow-up visit Treatment ap
plication and adjustment follow-up visit.
treatment adjustment visit Treatment application
and adjustment follow-up visit.
treatment adherence The degree to which a patient
follows his assigned treatment regimen. See treat
ment compliance.
treatment allocation I. The process of assigning pa-
305
tients to treatment. 2. The treatment assignment of
a particular patient.
treatment allocation design The plan for assigning
patients to treatment.
treatment allocation ratio The ratio of the number
of patients assigned or to be assigned to the testtreated group to those assigned or to be assigned to
the control-treated group (e.g., an allocation ratio
of 1:2 in a completed trial is one in which twice as
many patients were assigned to the control treat
ment as were assigned to the test treatment).
treatment allocation schedule The schedule used for
issuing treatment assignments.
treatment allocation strata Strata, designated be
fore the start of patient enrollment and defined by
baseline characteristics^) or clinic, that are used to
define subsets of patients who are assigned to treat
ment using allocation schedules constructed for the
individual strata.
treatment application and adjustment follow-up
visit A follow-up visit, the main purpose of which
is to enable clinic staff to apply or adjust treatment,
depending on patient needs and study specifica
tions.
treatment application and adjustment visit TYeatment application and adjustment follow-up visit.
treatment application follow-up visit Treatment ap
plication and adjustment follow-up visit.
treatment application visit Treatment application
and adjustment follow-up visit.
treatment arm Term sometimes used in place of
study treatment, or study group, especially in
cancer trials (but not in this book).
treatment assignment The treatment to be adminis
tered to the assignment unit (usually a patient, but
may be some other larger unit such as all members
of a family or members of a hospital ward) as
indicated in the treatment allocation schedule.
treatment assignment probability The probability
associated with a specified treatment assignment.
The value is fixed over the course of patient enroll
ment in trials with fixed allocation designs. It
changes in trials using adaptive allocation designs.
treatment assignment process The process of as
signing patients to treatment in a clinical trial.
treatment assignment unit The unit used in the treat
ment assignment process, usually patient, but the
unit may be made up of multiple individuals in
special cases such as in a trial involving treatment
of a family unit or an entire hospital ward. Equiva
lent to randomization unit in trials involving ran
dom allocation.
treatment block A block consisting of a prespeci
fied number of patients, all enrolled in reasonably
close proximity to one another and assigned to the
various study treatments in such a way so as to
satisfy a preset allocation ratio. See also treatment
block size.
treatment block size The number of allocations re
quired for a specified treatment block. For exam
ple, a random allocation schedule for a trial involv-
1
306
l:-
I,.
A. Glossary
ing two treatments, constructed using blocks of size
8 and an allocation ratio of |:|, would require
constraints on the assignment process such that the
specified allocation ratio is satisfied after every
eighth assignment.
treatment comparison Any comparison involving
two or more of the study treatment groups for a
designated outcome or follow-up variable.
treatment compliance The degree to which a pa
tient follows his assigned treatment regimen. See
treatment adherence.
treatment coordinating center A center in a clinical
trial that is responsible for coordinating the devel
opment and administration of the treatment proto
col, but that has few or no other responsibilities for
coordinating other aspects of the trial. Usually pres
ent only in multicenter trials involving a compli
cated treatment protocol. See also coordinating
center and data coordinating center.
treatment cross contamination Any instance in
which a patient, who was assigned to receive one
treatment in a trial, is exposed to one of the other
study treatments during the course of treatment or
follow-up.
treatment crossover Any change of treatment for a
patient in a clinical trial involving a switch of study
treatments. The switch may be planned, as in a
crossover trial, or may be unplanned, as in the case
of a noncrossover trial in which a patient assigned
to one treatment is exposed to one of the other
study treatments sometime during the trial. Un
planned crossovers are said to result in treatment
cross contamination.
treatment design The portion of the study design
that specifies the treatments to be evaluated, the
nature of the treatment structure, and the way in
which the treatments are to be administered.
treatment difference I. A difference observed be
tween the test- and control-treated groups of pa
tients for some specified outcome measure. 2. Any
specified or observed difference for a designated
outcome or follow-up variable involving two or
more treatment groups in the trial.
treatment effect I. An effect attributed to the test
treatment. Usually in clinical trials inferred from a
comparison of the test- and control-treated groups
of patients using observed results for a specified
outcome measure. 2. The effect produced or as
sumed to be produced by a treatment in an individ
ual patient. Usually assessed by measurements
made before and after administration of the treat
ment in that individual.
treatment effects monitoring I. Any process of re
viewing accumulated outcome data for groups of
patients in a trial as it proceeds to determine
whether a designated treatment procedure should
be altered or stopped. 2. The process of watching
for treatment effects in an individual patient (term
not ordinarily used in this context in this book).
treatment effects monitoring committee (TEMC)
I. A standing committee responsible for periodi-
V
cally reviewing accumulated data for evidence of
adverse or beneficial treatment effects during the
trial and for initiating recommendations for modifi
cation of a study treatment, including termination
of the treatment when appropriate. 2. One of the
key committees in the organizational structure of a
multicenter trial. Usually composed primarily, if
not exclusively, of individuals not directly involved
in patient care or data collection in the trial.
treatment failure I. Term sometimes used to char
acterize a study patient whose study physician has
found it necessary to alter his assigned treatment
because of the “failure” of the treatment to produce
a desired effect. 2. A patient in a clinical trial who
is no longer maintained on his assigned treatment,
whether or not he continues under follow-up.
(Usage note: Term not used in either context in this
book.)
treatment and follow-up stage The fifth stage of a
clinical trial in this book. Concerned with patient
treatment and follow-up (see Chapter 3 and Ap
pendix D).
treatment group The group of patients assigned to
receive a specified treatment. See study group.
treatment interaction A situation in which the effect
exerted by a treatment is influenced by the level, or
presence or absence, of some other factor or condi
tion not related to treatment (e.g., one would say
there is a treatment-sex interaction if the test-con
trol treatment difference is in one direction for
males and in the other direction, or is of a different
order of magnitude, for females).
treatment interaction effect The observed effect as
sociated with a treatment interaction.
treatment lag The time required, or presumed to be
required, for a treatment to exert its full effect.
treatment mask A condition or procedure that is
imposed to keep someone from knowing the true
identity of the treatment assignment.
treatment masking 1. A process in which treat
ments are administered so as to be single- or
double-masked. 2. Any process that is designed to
withhold information on treatment assignment
from some individual or group of individuals in a
clinical trial.
treatment monitoring Treatment effects monitoring.
treatment procedure The method of applying a par
ticular treatment in a clinical trial.
treatment protocol A document that describes the
treatment procedures used in a clinical trial.
treatment related bias A condition in which the na
ture of a reading, measurement, or classification
recorded on a particular patient is influenced by
the fact that the individual responsible for making
the reading, measurement, or classification has
knowledge of the patient’s treatment assignment.
treatment side effect A by-product of treatment,
either expected or unexpected, desired or unde
sired.
treatment structure The interrelationship of treat-
ments used in a clinical trial, e.g., as characterized
by treatments that are arranged in a factorial struc
ture.
treatment trial A trial in which the test treatment
consists of a procedure used for treatment of a
specific disease or health condition. Therapeutic
trial.
treatment unit The unit to which treatment is ad
ministered in a clinical trial. Usually patient, but
the unit may be composed of multiple individuals,
such as a family unit.
trial I. Clinical trial. 2. Any tentative or experimen
tal action done in order to obtain data for some
judgment or conclusion.
triple-blind Triple-masked in this book. See blind
for usage comment.
triple-blinded I. Triple-masked in this book. See
blind for usage comment. 2. Sometimes used in a
jocular fashion (not in this book) to characterize a
situation in which neither the patient, physician,
nor statistician knows how the trial is designed or
operated.
triple-blinded clinical trial Triple-masked clinical
trial.
triple-mask Triple-masked.
triple-masked Double-masked plus masking for the
individual (or group of individuals) who are re
sponsible for treatment monitoring.
triple-masked clinical trial A double-masked clini
cal trial in which data analyses done for treat
ment monitoring are presented to the individ
ual or group responsible for such monitoring in
a way that conceals the identity of the treatment
groups.
two-armed bandit A method of outcome adaptive
allocation in which the treatment assignment prob
ability for a particular treatment is a function of
the observed treatment difference in outcomes of
patients already enrolled in the trial. The motiva
tion being to minimize the number of patients as
signed to the inferior treatment (Robbins, 1952,
1956; Smith and Pyke, 1965; Zelen, 1969).
two-sided alternative hypothesis Two-tailed alterna
tive hypothesis.
two-sided test (statistics) 7\vo-tailed test.
two-tailed alternative hypothesis An alternative to
the null hypothesis that specifies a range of permis
sible values that are symmetrically distributed
about the null value (e.g., Hq'pi = M2 versus
M2)- See one-tailed alternative hypothe
sis for opposing term.
two-tailed test (statistics) A statistical test of signifi
cance based on the null value of no difference
versus the set of all alternative values (i.e., those
that lie to the right and left of the null value).
type I error (statistics) The probability of rejecting
the null hypothesis when it is true, usually denoted
by the Greek letter a.
type II error (statistics) The probability of accept
ing the null hypothesis when it is false, usually
denoted by the Greek letter /3.
307
U
UGDP University Group Diabetes Program,
uncontrolled Not controlled.
uncontrolled clinical trial A clinical trial that does
not involve a control treatment. In this book, any
study that does not have a control group made up
of patients treated and followed over the same time
period as those in a test-treated group.
uncrossed treatments I. A treatment structure not
involving a crossover design. 2. Nonfactorial treat
ment structure.
uniform treatment allocation A scheme in which
the assignment probability of any one treatment
group is the same as for every other treatment
group in a trial. See equal treatment allocation.
unmask To reveal the treatment assignment of an
individual patient or group of patients to an indi
vidual or group of individuals associated with the
trial (e.g., patients, study physicians, treatment ef
fects monitoring committee) who have heretofore
been denied this information.
unmasked trial Nonmasked trial.
unmasking The process of revealing a previously
masked item of information, such as treatment
assignment, to an individual or group of individu
als in a clinical trial.
unrestricted allocation 1. Any system of treatment
allocation that does not require the imposition of
any restriction on the assignment process, over and
above those implied with the adaptive or fixed
allocation scheme used. 2. Use of allocation sched
ules within clinics in a multicenter trial, or strata
within a clinic, but where there is no blocking
within clinic or strata within clinic.
unrestricted allocation schedule An allocation
schedule constructed using unrestricted allocation.
unrestricted random allocation Any unrestricted al
location scheme that uses a random process for
generating treatment assignments.
unrestricted randomization The process of generat
ing or issuing treatment assignments via an unre
stricted allocation schedule.
unscheduled interim follow-up visit Interim follow
up visit.
USPHS United States Public Health Service.
V
VA Veterans Administration.
VA 43 VACSPNo.43.
Cooperative
VACSP Veterans
Administration
Studies Program.
VACSP No. 43 Veterans Administration Coopera
tive Studies Program Number 43.
variable In this book, any trait, characteristic, test,
measurement, or assessment that is recorded, or
scheduled to be recorded, on patients enrolled, or
to be enrolled, in a clinical trial.'
VDT Video display terminal.
308
A. Glossary
verification A process that is carried out to verify
an item of information.
verify To confirm or substantiate an item of infor
mation recorded in a data file or keyed for entry
into the analysis database.
visit Clinic visit.
w
WHO
World Health Organization.
withdrawal I. A technical term used to refer to the
process of removing a specific individual from a
lifetable analysis because of termination offollow
up, or because of the occurrence of an event that
precludes further follow-up. 2. Dropout (not used
in this context in this book).
writing team A team of study investigators who are
appointed or designated to write a specified manuscript for presentation or publication.
B. Sketches of selected trials
B.l Introduction
B.2 Methods
B.3 Results
Table B-l List of trials sketched
Table B-2 Abstract summaries of trials
sketched
Table B-3 Publication list of sketched trials
Table B-4 Summary tabulations from sketches
Table B-5 Sample sketch for the UGDP
Table B-6 Data coordinating centers for multi
center trials referenced in this book
B.2
METHODS
The initial draft of each sketch was prepared by
the senior author using:
• Published manuscripts produced from the
trial (see Table B-3 for publication list)
• Basic design documents, such as original fund
ing applications or requests for proposals
• Operational documents, such as manuals
of operations, treatment protocols, data
forms, etc.
• Personal communications with study person
nel
Each sketch consisted of:
B.l
i:
•i.:
INTRODUCTION
Table B-l provides a list of the trials sketched in
this Appendix. They are all multicenter trials,
except the Physicians’ Health Study (Sketch 1),
and they all involve periods of follow-up of a
year or longer. They represent seven different
disease areas. The majority of the trials involve
cardiovascular disease. The National Institutes
of Health (NIH) serve as the sole source of sup
port for 11 of the 14 trials, and they share fund
ing responsibilities with a European foundation
in the case of the International Reflux Study in
Children. One of the other two trials was funded
by the Veterans Administration and the other
was funded by a pharmaceutical firm.
Eight of the trials involved tests of drugs, four
involved surgical procedures, and one, the Hy
pertension Prevention Trial (HPT), involved test
ingdiet modifications as preventive measures for
hypertension. The remaining trial, the Multiple
Risk Factor Intervention Trial (MRF1T), in
volved testing several different forms of inter
vention, all aimed at reducing known risk fac
tors for heart disease.
The sketches are designed to:
• Acquaint readers with the design and operat
ing features of some typical long-term trials
• Supplement information contained in the
body of this book on some of the more
frequently cited trials
• Provide a data resource for tabulations pre
sented in chapters throughout the book
• An abstract summap' of the trial
• A list of study publications
• Enumeration of the operating features of the
trial
A copy of the draft sketch was sent to the chair
man or vice-chairman of the study, director of
the coordinating center or data coordinating cen
ter, or project officer for review. The date the
review was completed (see item 33, Table B-4)
was used as the cutoff date for information con
tained in the sketch and is considered the com
pletion date for the sketch. Committee listing
and membership information (items 27, 28, and
29, Table B-4) are as of this date for trials that
had not yet entered the patient close-out stage.
The committee structure is characterized as of
the start of the close-out stage for trials that
were already in this or in a later stage when the
sketch was reviewed. Information on the steering
committee and the committee responsible for
treatment monitoring, as represented in items 28
and 29 of Table B-4, is based on data in the
sketch (see Table B-5 for sample) and was col
lected on a supplementary form that was com
pleted by the individual chosen to review the
sketch.
B.3
RESULTS
Table B-2 contains the abstract summary of
each trial sketched. Table B-3 lists official pa
pers of the 14 study groups. Only papers appear-
309
310
B. Sketches of selected trials
ing in peer review journals or periodicals are
listed. The list does not include:
• Papers published after the date in item 33,
Table B^l
• Papers in preparation, submitted for publica
tion but still under review, or accepted for
publication but not yet published as of the
date in item 33, Table B-4
• Abstracts or editorials concerning the study
• Papers published as part of a book, mono
graph, or the like, except where papers so
published are part of a periodical indexed
by the National Library of Medicine
• Reports published by the federal government
• Reports and documents placed on deposit at
the National Technical Information Service
or other similar repositories
The full list of publications is much more exten
sive than is shown in Table B-3 for some of the
trials sketched, such as NCGS and CASS.
The author citations in Table B-3 are repro
duced as supplied from the study, via the indi
vidual who reviewed the sketch, except for the
exclusions listed above and minor editing. A
comparison of citations in the table with those
appearing in the body of the book will reveal
differences. For example, named authors appear
in the author field of citation 5.2 in Table B-3.
B.3 Results
The author designation in the body of the book
for this citation is simply the National Coopera
tive Gallstone Study Group.
There are several reasons for the differences in
the citations. First, it is not always clear how a
paper should be listed from the arrangement of
information in the masthead of the paper. Sec
ond, preference was given to corporate author
designations when there was a choice between a
conventional or corporate format in the body of
the book. Use of corporate designations yielded
shorter citations and made for a more logical
grouping of related publications appearing in
the combined bibliography (Appendix 1) at the
end of this book. Third, the listings in Table B-3
as supplied by the individuals selected to review
the sketches were used to answer item 31.b in
Table B-4. This was considered to provide the
best basis for making the counts needed for that
item.
Table B-4 contains summary tabulations de
rived from the sketches (See Table B-5 for sam
ple sketch). The notes below relate to those tabu
lations. Table B-6 contains the name of the
director of the data coordinating center or coor
dinating center and address of the center for all
trials sketched, as well as the other multicenter
trials referenced in this book.
Item number in
Table B 4
I l.c
12, 13, 17
I9.a
I9.b
19.c
I9.d
20.a
20.b
20.d
20.c
20.f
Item number in
Table B-4
Comment
I
• See Glossary for definition of type of trial.
2
3
• See Chapter 9 or Glossary for definition of fixed sample size design. All trials were
of this type. None involved a sequential design.
• See Chapter 3 or Glossary for definitions of stages. The category Completed was
checked if all funding for the trial had terminated by the date recorded in item
33.
5
• See Glossary for definitions.
6
• See Chapter 21.
• See Glossary for definitions of direct, indirect, and consortium awards.
• Start of funding taken as date of first award to any center in the trial. Awards
issued simply for planning purposes were not counted in fixing the date. The
ending date is the projected date for termination of all financial support for the
trial. It is the termination date for completed trials.
• Number of clinics as of sketch completion for trials in the treatment and follow-up
stage or earlier stage. Number of clinics entering the patient close-out stage for
trials in that stage or beyond as of the sketch completion date given in item 33.
• See Glossary for definitions.
• Degree of coordinating center or data coordinating center director. Indicated as
Bio (biostatistics). Epi (epidemiology), or Med (medicine).
7
8
9
10
II.b
20.g
21
22.a
22.b
24
26. a
Comment
• Answered No if the coordinating center or data coordinating center was financially
and administratively independent of all other centers in the trial. Answered Yes
if the center was funded through a clinical center in the trial or if it was under the
administrative control of a clinic center director.
• See Glossary for definitions.
• See Chapter 14. Direct checked when potential study patients were identified and
approached by study personnel, such as when patients are recruited from a
primary care facility under the control of clinic investigators, or when patients
are identified through special screening or direct mailings initiated by clinic
personnel. Indirect checked if the initial contact is through some other agent or
party outside the clinic, such as a referring physician, through review of records
held by nonstudy physicians or at nonstudy hospitals, or via mass advertising
compaigns initiated from the study and aimed at the general public.
• Month and year first patient was randomized.
• Month and year last patient was randomized. Projected date for trials still in the
patient recruitment stage.
• Total number of patients enrolled (all treatment groups combined). Count at or
before the date given in item 33.
• All 14 trials involved formal methods of randomization, as opposed to informal,
nonrandom, or quasi-random methods, such as discussed in Chapter 8. All trials
used patients as the randomization unit except one, the Macular Photocoaguiation Study in which eyes served as the randomization unit.
• See Glossary for defninitions.
• The total number of allocation strata. Given by the product of the number of
subgroups formed with each stratification variable. For example, a trial involv
ing stratification by clinic (10 clinics), age (three levels), and sex would have
10X3X2 = 60 allocation strata.
• See Chapter 10 and Glossary for definitions. Characterized as uniform if the
allocation scheme was designed to yield equal numbers of assignments to the
study treatments within a strata. Classified as nonuniform (denoted as Nonuni
in the table) if this condition does not apply.
• Answered Yes if the allocations are blocked (see Chapter 10) to force the treatment
assignments to satisfy a specified allocation ratio at various points during the
patient enrollment process.
• Locus of control for the randomization classified as Central if release of individual
assignments was triggered by written or telephone contact of clinic staff with
staff of the coordinating center or data coordinating center (or staff of some
other control center), or release was controlled via an on-site computer under
the control of the coordinating center or data coordinating center. Control
considered Local if clinic staff could obtain an allocation without use of an on
site computer controlled by the coordinating center or data coordinating center
or without any contact with a coordinating center of other control facility.
• See Chapter 12 and Glossary for definitions. Regular follow-up visits do not
include visits done simply for treatment application or adjustment.
• Recorded as the average length of follow-up or as a range. Anticipated values for
trials that had not yet entered the patient close-out stage. Ranges recorded for
trials in the close-out stage or beyond based on actual times of enrollment of the
first and last patients entered into the study.
• Indicated as NA (not applicable) for trials still in the patient close-out stage or an
earlier stage. Indicated as None done if trial is completed and no post-trial
follow-up was done. The average length of follow-up provided or anticipated if
post-trial follow-up was done or is under way.
• See Chapters 8 and 15.
• Original recruitment goal: That set when the trial was planned. Value recorded is
as stated in the original design documents of the trial (e.g., original grant
application, RFP, or RFA), or as reported in a study publication.
311
312
B. Sketches of selected trials
hem number in
Table B-4
26.b
Comment
• The category Sample size calculation was checked if the goal was based on a
sample size calculation with a specified level of type I and II error protection.
The category Pragmatic was checked if a recruitment goal was set, but was based
on practical considerations rather than on a formal sample size calculation.
26.c
• The number of patients recruited (all treatments combined). Listed as NA (not
applicable) if recruitment not yet completed as of date in item 33.
26.d
• Answer based on information in published reports of results from the listed trials
(see Table B-3 for list). The category Not applicable checked for trials that have
yet to publish any results. The category None required checked for trials that
produced a significant effect before patient recruitment was completed. The
category None stated checked if the treatment effect observed was not consid
ered to be significant by the authors of the report and the report contains no
discussion of the rationale for the achieved sample size, or of the power pro
vided to detect a specified treatment difference with the observed sample
iple size
and control event rate. The category Sample size calculation checked if the
original recruitment goal was achieved and if the goal was the result of a sample
size calculation made during or prior to the close of the patient recruitment
stage. The category Pragmatic checked if recruitment was completed and the
report contains a statement indicating the achieved sample was the result of
practical considerations (i.e., was not the result of a formal sample size calcula
tion). A check in both categories Power calculation and Sample size calculation
indicates the report satisfied the requirements for both categories.
27
• See Chapter 23 and Glossary for definitions.
28.a, ii
• Patient care responsibilities? Answered Yes if chairman was responsible for treat
ment or care of any patients in the trial.
28.a. iii
• Recorded as Self-appt (self-appointed) in investigator-initiated trials where the
chairman was designated on the initial application. Recorded as £ (elected) if
chairman was chosen by the investigative group after the trial was funded, either
by acclamation or through a formal election. Recorded as Appt (appointed) if
chairman was selected by the sponsor or the advisory-review committee of the
trial.
28.a, iv
• Recorded as WT (without term) if the chairman, regardless of whether selfappointed, elected, or appointed, serves without term. Recorded as Term if
chairman selected or appointed for a specified term less than the expected
duration of the trial.
28.b, ii
through iv
• See comments for 28.a, ii through iv.
28.c, i
• The study centers referred to in this item and in item 28.c, ii include clinical
centers, as well as all resource centers. The number of members and their voting
status is based on information supplied in study publications and as supplied
ipplied
from the study on a special form completed by the individual selected to review
the sketch, as described in Section B.2.
28.d
28.e
• See Glossary for definitions of the positions listed. The positions denoted by items
i through vi that were represented on the steering committee (SC) are marked
Yes. No indicates that the position exists in the study, but that it is not
represented on the SC. Positions that do not exist in the study are marked NA
(not applicable). The positions represented by items vii and viii were marked Yes
if individuals of the type indicated were on the SC, and were marked No if not.
• This item indicates the number of individuals elected to membership on the SC by
some body of the study—generally the entire investigative body. The letter T
following the number indicates election for a specified term. The letters WT
indicate election without term.
Table B-l List of trials sketched 313
hem number in
Table R 4
Comment
30
• See Chapter 23 for distinguishing features of communication models. Classifica
tions made by author.
3l.a
• Number from Table B-3.
31.b
• A paper was counted in the first category (corporate format) if the author field, as
displayed in Table B-3. only contained the study name. It was counted in the
second category (conventional format) if the author field only contained names
of individual authors. It was counted in the third category (both formats) if the
author field contained both the study name and the name of one or more
authors.
32
• Item used to indicate the degree of involvement of the senior author of this book in
the particular trials sketched.
33
• Taken as the date of review, as discussed in Section B.2.
Table B-l
List of trials sketched
Sketch
number
Study name
Acronym
Disease
Sponsor
1
Physicians’ Health Study
PHS
Cancer,
cardiovascular
NIH
2
University Group Diabetes Program
UGDP
Diabetes
NIH
3
VA Cooperative Studies Program
Number 43
VA 43
Diabetes
VA
4
Macular Photocoagulation Study
MPS
Eye
NIH
5
National Cooperative Gallstone Study
NCOS
Gallbladder
NIH
6
Coronary Drug Project
CDP
Cardiovascular
NIH
7
Aspirin Myocardial Infarction Study
AMIS
Cardiovascular
NIH
8
Persantine Aspirin Reinfarction Study
PARIS
Cardiovascular
Industry
9
Hypertension Detection and Follow-Up
Program
HDFP
Cardiovascular
NIH
NIH
10
Multiple Risk Factor Intervention Trial
MRF1T
Cardiovascular
II
Coronary Artery Surgery Study
CASS
Cardiovascular
NIH
POSCH
Cardiovascular
NIH
29.a
• The committee scheduled to perform the treatment effects monitoring function for
trials where monitoring is not yet under way. The actual committee performing
that function for all other trials. See Glossary and Chapter 23.
12
Program on the Surgical Control of Hy
perlipidemia
13
Hypertension Prevention Trial
HPT
Cardiovascular
NIH
29.b through f
• See comments for 28.a through e.
14
International Reflux Study in Children
IRSC
Kidney
29.g
• The actual or planned number of meetings per year of the committee listed in item
29.a.
NIH and
foundation
314
Table B-2 Abstract summaries of trials sketched 315
B. Sketches of selected trials
Table B-2 Abstract summaries of trials sketched (continued)
Table B-2 Abstract summaries of trials sketched
1. Physicians' Health Study (PHS)
The PHS is a randomized, double-masked, placebo-controlled clinical trial designed to test the value of regular
use of aspirin on all cause and cardiovascular mortality after 4.5 years of follow-up, as well as beta-carotene on
total cancer incidence in the last 2.5 years of the trial. Patients are randomly assigned to one of the four
treatments listed below.
Treatment
Dosage
• Aspirin + beta-carotene: ASA + /3
• One (325 mg) aspirin tablet every other day. One (30 mg)
beta-carotene capsule on alternate days.
• One aspirin tablet every other day. One beta-carotene
placebo capsule on alternate days.
• One aspirin placebo tablet every other day. One beta-caro
tene capsule on alternate days.
• One aspirin placebo tablet every other day. One beta-caro
tene placebo capsule on alternate days.
• Aspirin + beta-carotene placebo:
ASA +
• Aspirin placebo + beta-carotene:
ASA + p
• Aspirin and beta-carotene placebos:
ASA + p
Over 21,500 male physicians, 40 to 84 years of age at entry, are to be enrolled. Physicians volunteering to
participate will receive their assigned medication via mail and will be asked to complete a short questionnaire
first at 6-month and later at I-year intervals after enrollment. The questionnaires will be collected by mail and
will be used to assemble information on treatment adherence, treatment side effects, and morbidity.
Participants are not required to make any clinic visits.
2. University Group Diabetes Program (UGDP)
The UGDP was a randomized, controlled, multicenter clinical trial designed to evaluate the effectiveness of
long-term hypoglycemic drug therapy in preventing or delaying the vascular complications of diabetes. Only
newly diagnosed, noninsulin dependent, adult-onset diabetics were eligible for enrollment. The study started
patient enrollment in early 1961. All patient follow-up terminated in 1975. A total of 1,027 patients were
enrolled and randomly assigned to one of the treatments listed below.
b
Treatment
Dosage
• Insulin variable: IVAR
• As much insulin (U-80 Lente Iletin or other insulins) per
day as required to maintain “normal” blood glucose lev
els. Administered via subcutaneous injections.
• 10, 12, 14, or 16 units of insulin (U-80 Lente Iletin) per day,
depending on patient body surface. Administered via sub
cutaneous injections.
• 3 tablets per day, each containing 0.5 gms of tolbutamide.
• 1 capsule per day during first week of treatment, thereafter 2
capsules per day; 50 mg of phenformin per capsule.
• Number of tablets or capsules similar to those used for
tolbutamide or phenformin treatments.
p I r i ii ■
• Insulin standard: ISTD
• Tolbutamide: TOLB
• Phenformin: PHEN
• Lactose placebo: PLBO
I-
i
The tolbutamide and phenformin treatments were terminated in 1969 and 1971, respectively, because of lack
of efficacy. The two insulin treatments were continued to the end of planned patient follow-up (1975), but were
not judged to be any more effective than placebo medications in prolonging life or in delaying the onset and
development of vascular complications.
3. VA Cooperative Studies Program Number 43 (VA 43)
The study is a long-term, randomized, double-masked, placebo-controlled clinical trial of aspirin and dipyrida
mole in diabetics with advanced vascular disease. The test treatment is a combination of aspirin and dipyrida-
mole (one 325 mg tablet of aspirin and one 75 mg tablet of dipyridamole, three times per day). Patients assigned
to placebo treatment received a prescription for a tablet schedule identical to that of the test-treated group.
Patients in the trial had to have gangrene of the feet at enrollment or had to have had an amputation on one or
both of their feet for gangrene in the last 12 months prior to enrollment. The study enrolled 231 patients.
Recruitment was completed in May 1980. The study investigators plan to announce the results of the trial
sometime in 1984.
4. Macular Photocoagulation Study (MPS)
The MPS is a multicenter study designed to assess the value of photocoagulation in eyes with choroidal
neovascularization. The study consists of two sets of trials. The first was started in 1979 and focuses on the
assessment of argon laser photocoagulation in eyes with leaking choroidal neovascular membranes that are
between 200 and 2,500 microns from the center of the foveal avascular zone (FAZ). The second set of trials,
started in 1982, involves use of krypton laser photocoagulation. This mode of treatment is restricted to eyes that
were judged ineligible for argon laser photocoagulation because the choroidal neovascular membranes to be
treated fell within 200 microns of the FAZ. Both studies involve three different types of eye diseases, as outlined
below.
Type of eye disease
Eligibility criteria
• Senile macular degeneration
• Neovascularization; visible drusen bodies as large or larger
than those defined by standard MPS fundus photo
graphs; age > 50 at entry.
• Neovascularization; at least one atrophic scar (histo spot) in
either eye; age > 18 at entry.
• Neovascularization; no evidence of SMD or any other cause
for neovascularization; no drusen bodies greater than
those defined by standard MPS fundus photographs; no
histo spots in either eye.
(SMD)
• Presumed ocular histoplasmosis
(H1STO)
• Idiopathic neovascular membrane
(1NVM)
Eligible eyes in both sets of trials are randomly assigned to receive photocoagulation or no treatment. All
patients are followed for changes in vision. The only results available from the trial through the date listed in
item 33, Table B-4, relate to argon-treated SMD patients. Patient recruitment for that trial was terminated
because of the apparent superiority of argon treatment. Of the SMD untreated eyes, 60% had reduced visual
capacity by the eighteenth month of follow-up, compared with only 25% of the argon-treated SMD eyes.
5. National Cooperative Gallstone Study (NCGS)
The NCGS was a double-masked, randomized, controlled, trial designed to assess the efficacy and safety of
chenodiol (chenodeoxycholic acid) for dissolution of radiolucent gallstones. The treatments are outlined below.
Treatment
Dosage
• High dose chenodiol: H
• Low dose chenodiol: L
• Placebo: P
• 6 capsules per day, each containing 125 mg of chenodiol.
• 6 capsules per day, each containing 62.5 mg of chenodiol.
• 6 capsules per day, each containing 3 mg of sodium cholate.
Nine hundred sixteen patients (not counting the 128 patients enrolled in a preliminary biopsy study) were
enrolled, treated, and followed by the ten clinical centers participating in the trial. The percentages of patients
with complete gallstone dissolution, after two years of treatment, as determined by radiographic metrology,
were 13.5 for H, 5.2 for L, and 0.8 for P. Partial (over 50% dissolution) or complete dissolution occurred in
40.8% of H-treated patients, 23.6% of the L-treated patients, and 11.0% of the P-treated patients. Clinically
significant hepatotoxicity requiring termination of the assigned treatment occurred in 3% of the H-treated
patients and in 0.4% of the L-treated and P-treated patients.
316
Table B-2 Abstract summaries of trials sketched
B. Sketches of selected trials
Table B-2 Abstract summaries of trials sketched (continued)
Table B-2 Abstract summaries of trials sketched (continued)
6. Coronary Drug Project (CDP)
The CDP was a double-masked, randomized, controlled clinical trial designed to evaluate the efficacy of several
different Iipid-influencing drugs in prolonging the lives of men (aged 30 through 64 at entry) with a prior history
of myocardial infarction. The treatments investigated are listed below.
respectively. However, subgroup analyses of the data suggested the combination of Persantine® and aspirin may
be beneficial in prolonging life, if used within a few months following an ML This finding led to initiation of
PARIS, Part II (not sketched).
Treatment
Dosage per day*
• Low dose estrogen: ESGI
• High dose estrogen: ESG2
• Clofibrate: CPIB
• 2.5 mg of mixed conjugated equine estrogen (Premarin®)
• 5.0 mg of mixed conjugated equine estrogen (Premarin®)
• 1.8g of ethyl alpha parachlorophenoxy-isobutyrate
(Atromid-S®)
• 6.0 mg of dextrothyroxine (Choloxin®)
• Dextrothyroxine: DT4
• Nicotinic acid: NICA
• Placebo: PLBO
• 3.0 mg of nicotinic acid
• 3.8g of lactose (placebo)
•Patients were required to take 9 capsules per day (3 capsules 3 times a day) to receive the specified dosage. They were
started on 3 capsules per day. They were stepped to 6 capsules per day I month later and then to 9 capsules per day I month
thereafter. They were then maintained on 9 capsules per day. except where contraindicated.
The study involved 55 clinics, a coordinating center, project office, central laboratory, ECG reading center,
and drug procurement and distribution center. A total of 8,341 patients were enrolled. All patients were
followed for a minimum of 5 years.
The two estrogen and dextrothyroxine treatments were discontinued during the course of the trial because of
lack of efficacy. In addition, the clofibrate and nicotinic acid treatments, while continued to the end of the trial,
did not show any evidence of efficacy. The 5-year mortality rates were 20.0, 21.2, and 20.9 per 100 population
for CPIB. NICA, and PLBO, respectively.
•Iri n ■ 1 1
317
7. Aspirin Myocardial Infarction Study (AMIS)
AMIS was designed to test the efficacy of aspirin in prolonging life in patients with a prior history of myocardial
infarction. A total of 4,524 patients were enrolled and followed through the efforts of 30 clinical centers, a
coordinating center, project office, central laboratory, ECG reading center, and drug procurement and distribu
tion center. Patients assigned to aspirin treatment (ASA) received 1.0g of aspirin per day (two capsules per day,
each containing 0.5g of aspirin). Patients assigned to the placebo treatment (PLBO) received a capsule schedule
similar to that for aspirin-treated patients. Patients were followed for a minimum of 3 years.
The study failed to show any benefit for ASA treatment. In fact, the 3-year mortality rate for the ASA
treatment group was higher than that for the PLBO treatment group (9.6 versus 8.8 per 100 population).
8. Persantine Aspirin Reinfarction Study (PARIS)
PARIS was an industry-sponsored randomized, controlled, clinical trial designed to test the efficacy of
Persantine® (dipyridamole) and aspirin in prolonging lives of patients with an ECG-documented history of
myocardial infarction (Ml). The trial involved 2,026 patients. Clinics from both the United States and the
United Kingdom participated. Patients were treated and followed for a minimum of 3 years. The treatments
were as outlined below.
Treatment
Dosage
• Persantine® + Aspirin: PR + ASA
• 2 tablets, 3 times per day. 1 containing 75 mg of Persantine®
and the other containing 324 mg of aspirin.
• 2 tablets, 3 times per day. I containing 324 mg of aspirin
and the other containing placebo medication.
• 2 tablets, 3 times per day. Both containing placebo medica
tion (starch, calcium phosphate, and microcrystalline cel
lulose).
• Persantine® placebo + Aspirin:
PR + ASA
• Persantine® placebo + Aspirin
placebo:
+ ASA
There was no significant difference among the treatment groups in mortality. The percentages dead at the end
of the study were 10.7, 10.5. and 12.8 for the PR + ASA, PR + ASA, and PR" + ASA treatment groups.
9. Hypertension Detection and Follow-Up Program (HDFP)
The HDFP was a randomized, controlled, clinical trial of stepped care versus referred care for patients with
hypertension. The study involved 14 clinics, a coordinating center, project office, central laboratory, ECG
reading center, and a drug procurement and distribution center. A total of 10,940 men and women with a
qualifying diastolic blood pressure (DBP) of 90 mmHg or higher were enrolled. Patients assigned to stepped
care were treated at the study clinics by clinic personnel using a treatment protocol calling for stepped increases
in the dosage of a prescribed medication or in the number of antihypertensive agents used in order to achieve
desired BP reductions. Patients assigned to the referred care group were referred to their usual source of medical
care for treatment.
Five-year mortality (all cause) was 17% lower for the stepped care group than for the referred care group (6.4
for reterred
referred care). The
concluded
per 100 population for stepped care versus 7.7 per 100 population tor
1 ne investigators conciuoea
that the findings “indicate that the systematic effective management of hypertension has a great potential for
reducing mortality for the large number of people with high BP in the population, including those with ‘mild’
hypertension.”
10. Multiple Risk Factor Intervention Trial (MRFIT)
MR FIT was a randomized, controlled, clinical trial designed to assess the value of a multifactor intervention
program aimed at reducing known risk factors for coronary heart disease (CHD). The three risk factors were
elevated serum cholesterol, high blood pressure, and cigarette smoking. Only men aged 35 through 57 at entry,
with no overt evidence of CHD, were eligible for enrollment. The 12,866 men enrolled were randomly assigned
to either special intervention (SI) or usual care (UC). Those assigned to SI were placed on specific treatments for
the risk factors present. A dietary approach was used for cholesterol reduction, antihypertensive drugs (plus
weight reduction where appropriate) were used for blood pressure reduction, and a behavioral approach was
used to achieve cessation or reduction of cigarette smoking. Participants assigned to UC were not given any care
via study clinics for elevated blood pressure or advice on how to reduce cholesterol or cigarette consumption.
However, hypertensives diagnosed via the study were referred to their usual source of care for treatment.
The trial completed participant recruitment in early 1976. Participants assigned to SI continued to be exposed
to the required interventions until termination of follow-up in early 1982. The first report of findings appeared
in late 1982. The report showed the interventions practiced on the Si-treated group to be effective in lowering
blood pressure, cholesterol, and cigarette consumption. However, these reductions had virtually no effect on
mortality. There was a slight but nonsignificant reduction in deaths from coronary heart disease (17.9 versus
19.3 per 100 population for the Si-treated versus the UC-treated groups). However, all cause mortality was
slightly higher in the Si-treated group than in the UC-treated group (4.1 versus 4.0 per 100 population).
Subgroup analyses presented in the 1982 publication raise the possibility that Si-treated men with hypertension
and resting ECG abnormalities at entry may have fared worse than the corresponding UC-treated group of men.
11. Coronary Artery Surgery Study (CASS)
CASS is a multicenter study consisting of two components: A trial designed to assess the efficacy of coronary
artery surgery in patients with proven coronary artery disease and a registry of consecutive patients undergoing
coronary arteriography. The trial component involved 780 patients assigned to coronary bypass surgery or
conventional medical therapy. The registry is made up of 24.959 patients, including randomized patients.
Patients in both components are followed for mortality, as well as for various nonfatal cardiovascular events.
The study involves 15 clinical centers, a coordinating center, project office, and ECG reading center. Results
comparing surgical and medical treatment in the randomized portion of the study had not been published, as of
the completion date for this sketch.
12. Program on the Surgical Control of Hyperlipidemia (POSCH)
POSCH is a randomized clinical trial designed to determine whether reducing cholesterol levels via partial ileal
bypass, in patients with high cholesterol levels and a prior history of myocardial infarction (Ml), is useful in
prolonging life and mitigating atherosclerosis. Patients in the trial are randomly assigned to bypass surgery or
regular medical care. Patient recruitment is scheduled to continue through May 1983 with the goal being to
enroll 800+ patients. Follow-up is expected to continue for a minimum of 5 years after completion of
recruitment. There are no publications containing treatment results, as of the date in item 33, Table B 4.
318
Table B-3 Publication list of sketched trials
B. Sketches of selected trials
Table B-2 Abstract summaries of trials sketched {continued)
Table B-3
13. Hypertension Prevention Trial (HPT)
The HPT is a randomized, controlled, multicenter trial designed to assess the efficacy of different forms of
dietary intervention in preventing the development of hypertension. Current funding is for the first stage of a
possible two-stage effort. The first stage is designed to develop and test methods and procedures needed for the
second stage and will involve 800 participants randomly assigned to the treatment groups indicated below. The
second stage, if warranted by results from the first stage, may involve as many as 6,000 participants and is
expected to start in 1985 or 1986.
1. Physicians’ Health Study (PHS)
Treatment
High weight strata
• Sodium restriction: Na
• Sodium restriction and potassium supple
mentation: NaK
• Sodium restriction plus caloric restriction
for weight reduction: NaCal
• Caloric restriction for weight reduction:
Cal
• Control: Ct
Dietary goal
• Reduce sodium intake to < 70 mEq per day.
• Reduce sodium intake to < 70 mEq per day and
increase potassium intake to > 100 mEq per day.
• Reduce sodium intake to < 70 mEq per day and
restrict calorie intake to bring body weight within
normal limits.
• Reduce calorie intake to bring body weight within
normal limits.
• None.
Normal weight strata
• Sodium restriction: Na
• Sodium restriction and potassium supple
mentation: NaK
• Control: Ct
• Reduce sodium intake to < 70 mEq per day.
• Reduce sodium intake to < 70 mEq per day and
increase potassium intake to > 100 mEq per day.
• None.
Only nonhypertensive individuals with diastolic blood pressures > 78 mm Hg but < 90 mm Hg are eligible for
enrollment in the first stage. Individuals who fall in the high weight strata, as determined by Quetelet’s Index,
are assigned to any one of the five treatments listed above. Individuals who are not considered to be overweight
by this index are assigned either to the control treatment or to one of the two dietary treatments not involving
caloric restriction.
The dietary goals stated above are pursued via a series of group counseling sessions in which individuals
are shown how to shop, cook, and eat to achieve the desired goals. The counseling process will be maintained
over the course of the trial. All participants will be followed for a period of 2 to 3 years for blood pressure
changes.
14. International Reflux Study in Children (1RSC)
The IRSC is a randomized, controlled, clinical trial of surgical versus conventional medical treatment of
vesicoureteral reflux (VUR) in children under the age of ten at entry. The study involves a multinational set of
clinics directed by a steering committee with representatives from Europe and the United States. Data collection
in the United States is supervised by a data center based in New York. Data collection in Europe is supervised by
a data center based in Essen. The German data center will serve as the analysis center for the combined United
States-European data set.
Grade III (European clinics only) and IV reflux patients are being enrolled and followed for evidence of renal
scarring and measurement of renal growth. The trial has been under way for 3 years and is scheduled to continue
for several more years. No results have been published as of the date in item 33, Table B -4.
319
Publication list of sketched trials
None
2. University Group Diabetes Program (UGDP)
2.1
University Group Diabetes Program: A study of the effects of hypoglycemic agents on vascular
complications in patients with adult-onset diabetes. 1: Design, methods and baseline characteris
tics. Diabetes I9(suppl 2): 747-783, 1970.
2.2
University Group Diabetes Program: A study of the effects of hypoglycemic agents on vascular
complications in patients with adult-onset diabetes. II: Mortality results. Diabetes I9(suppl 2):
785-830, 1970.
2.3
University Group Diabetes Program: Effects of hypoglycemic agents on vascular complications in
patients with adult-onset diabetes. Ill: Clinical implications of UGDP results. JAMA 218:14001410, 1971.
2.4
University Group Diabetes Program: Effects of hypoglycemic agents on vascular complications in
patients with adult-onset diabetes. IV: A preliminary report on phenformin results. JAMA
217:777-784, 1971.
2.5
2.6
2.7
2.8
2.9
2.10
Prout TE, Knatterud GL, Meinert CL, Klimt CR: The University Group Diabetes Program: The
UGDP Controversy: Clinical trials versus clinical impressions. Diabetes 21:1035-1040, 1972.
University Group Diabetes Program: A study of the effects of hypoglycemic agents on vascular
complications in patients with adult-onset diabetes. V: Evaluation of phenformin therapy. Dia
betes 24(suppl I ):65— 184, 1975.
University Group Diabetes Program: A study of the effects of hypoglycemic agents on vascular
complications in patients with adult-onset diabetes. VI: Supplementary report on nonfatal events
in patients treated with tolbutamide. Diabetes 25:1129-1153, 1976.
University Group Diabetes Program: Effects of hypoglycemic agents on vascular complications in
patients with adult-onset diabetes. VII: Mortality and selected nonfatal events with insulin
treatment. JAMA 240:37-42, 1978.
Prout T, Knatterud G, Meinert C: The University Group Diabetes Program: Diabetes drugs:
Clinical trial (letter). Science 204:362-363, 1979.
University Group Diabetes Program: Effects of hypoglycemic agents on vascular complications in
patients with adult-onset diabetes. VIII: Evaluation of insulin therapy: Final report. Diabetes
31 (suppl 5): 1-78, 1982.
3. VA Cooperative Studies Program Number 43 (VA 43)
None
4. Macular Photocoagulation Study (MPS)
4.1
Macular Photocoagulation Study Group: Argon laser photocoagulation for senile macular degener
ation. Arch Ophthalmol 100:912-918, 1982.
5. National Cooperative Gallstone Study (NCGS)
5.1
Croke G: Recruitment for the National Cooperative Gallstone Study. Clin Pharmacol and Ther
25:691-694, 1979.
5.2
Lachin JM, Marks JW, Schoenfield LI; the NCGS Protocol Committee (Malcolm P. Tyor,
Chairman, Peter H. Bennett, Scott M. Grundy, William G. M. Hardison, Lawrence W. Shaw,
Johnson L. Thistle, Z. R. Vlahcevic) and the National Cooperative Gallstone Study Group:
Design & methodological considerations in the National Cooperative Gallstone Study: A multi
center clinical trial. Controlled Clin Trials 2:177-229, 1981.
Lasser EC, Amberg JR, Baily NA, Varady P, Lachin J, Okun R, Schoenfield L: Validation of a
5.3
computer-assisted method for estimating the number and volume of gallstones visualized by
cholecystography. Invest Radiol 16:342-347, 1981.
5.4
Schoenfield U, Lachin JM, the Steering Committee (Baum RA, Habig RL, Hanson RF, Hersh T,
Hightower NC, Jr., Hofmann AF, Lasser EC, Marks JW, Mekhjian H, Okun R, Schaefer RA,
Shaw L, Soloway RD, Thistle JL, Thomas FB, Tyor MP), and the National Cooperative
Table B-3 Publication list of sketched trials
320
Table B-3 Publication list of sketched trials (continued)
Gallstone Study Group: Chenodial (chenodeoxycholic acid) for dissolution of gallstones: The
National Cooperative Gallstone Study. A controlled trial of efficacy and safety. Ann Intern Med
95:257-282,1981.
Albers JJ, Grundy SM, Cleary PA, Small DM, Lachin JM, Schoenfield LJ, and the National
5.5
Cooperative Gallstone Study Group: National Cooperative Gallstone Study: The effect of
chenodeoxycholic acid on lipoproteins and apolipoproteins. Gastroenterology 82:638-646, 1982.
Fisher RL, Anderson DW, Boyer JL, Ishak K, Klatskin G, Lachin JM, Phillips MJ, and the
5.6
Steering Committee for the National Cooperative Gallstone Study Group: A prospective mor
phologic evaluation of hepatic toxicity of chenodeoxycholic acid in patients with cholelithiasis:
The National Cooperative Gallstone Study (NCGS). Hepatology 2:187-201, 1982.
Hofmann AF, Grundy SM, Lachin JM, Lan SP, Baum RA, Hanson RF, Hersh T, Hightower NC,
5.7
Jr., Marks JW, Mekhjian H, Schaefer RA, Soloway RD, Thistle JL, Thomas FB, Tyor MP, and
the National Cooperative Gallstone Study Group: Pretreatment biliary lipid composition in white
patients with radiolucent gallstones in the National Cooperative Gallstone Study (NCGS).
Gastroenterology 83:738-752, 1982.
Habig RL, Thomas P, Lippel K, Anderson D, Lachin J: Central laboratory quality control in the
5.8
National Cooperative Gallstone Study. Controlled Clin Dials 4:101-123, 1983.
Lachin JM, Schoenfield LJ, and the National Cooperative Gallstone Study Group: Effects of dose
5.9
relative to body weight in the National Cooperative Gallstone Study: A fixed-dose trial. Con
trolled Clin Trials 4:125-131, 1983.
5.10 Phillips MJ, Fisher RL, Anderson DW, Lan SP, Lachin JM, Boyer JL and the Steering Committee
for the National Cooperative Gallstone Study Group. Ultrastructural evidence of intrahepatic
cholestasis before and after chenodeoxycholic acid (CDCA) therapy in patients with cholelithia
sis: The National Cooperative Gallstone Study (NCGS). Hepatology 3:209-220, 1983.
5.11 Schoenfield U, Grundy SM, Hofmann AF, Lachin JM, Thistle JL, Tyor MP, for the National
Cooperative Gallstone Study: The National Cooperative Gallstone Study viewed by its investiga
tors. Gastroenterology 84:644-648, 1983.
h
321
B. Sketches of selected trials
6. Coronary Drug Project (CDP)
Coronary Drug Project Research Group: The Coronary Drug Project: Initial findings leading to
6.1
modifications of its research protocol. JAMA 214:1303-1313, 1970.
6.2 Coronary Drug Project Research Group: The Coronary Drug Project: Findings leading to further
modifications of its protocol with respect to dextrothyroxine. JAMA 220:996-1008, 1972.
Coronary Drug Project Research Group: The prognostic importance of the electrocardiogram after
6.3
myocardial infarction: Experience in the Coronary Drug Project. Ann Intern Med 77:677-689,
1972.
Coronary Drug Project Research Group: The Coronary Drug Project: Design, methods and
6.4
baseline results. Circulation 47(suppl I): 1-1 —1-50 (plus appendixes), 1973.
Coronary Drug Project Research Group: The Coronary Drug Project: Findings leading to disconti
6.5
nuation of the 2.5-mg/day estrogen group. JAMA 226:652-657, 1973.
Coronary Drug Project Research Group: Prognostic importance of premature beats following
6.6
myocardial infarction: Experience in the Coronary Drug Project. JAMA 223:1116-1124, 1973.
Coronary Drug Project Research Group: Factors influencing long-term prognosis after recovery
6.7
from myocardial infarction: Three-year findings of the Coronary Drug Project. J Chronic Dis
27:267-285, 1974.
Coronary Drug Project Research Group: Left ventricular hypertrophy patterns and prognosis:
6.8
Experience post-infarction in the Coronary Drug Project. Circulation 49:862-869, 1974.
Coronary Drug Project Research Group: The prognostic importance of premature ventricular
6.9
complexes in the late post-infarction period: Experience in the Coronary Drug Project. Acta
Cardiol (suppl l8):33-53, 1974.
6.10 Coronary Drug Project Research Group: The Coronary Drug Project. Clofibrate and niacin in
coronary heart disease. JAMA 231:360-381, 1975.
6.11 Coronary Drug Project Research Group: Reply to letter from D. J. Gans. JA MA 234:22-23, 1975.
Table B-3 Publication list of sketched trials (continued)____________________ _________
6.12
Coronary Drug Project Research Group: Aspirin in coronary heart disease. J Chronic Dis 29:625-
M2, 1976.
Coronary Drug Project Research Group: Serum uric acid: Its association with other risk factors and
with mortality in coronary heart disease. J Chronic Dis 29:557-569, 1976.
6.14 Coronary Drug Project Research Group: The prognostic importance of plasma glucose levels and of
the use of oral hypoglycemic drugs after myocardial infarction in men. Diabetes 26:453-465,
1977.
6.15 Coronary Drug Project Research Group: Gallbladder disease as a side-effect of drugs influencing
lipid metabolism: Experience in the Coronary Drug Project. N Eng! J Med 296:1185-1190, 1977.
6.16 Coronary Drug Project Research Group: The Coronary Drug Project: Implications for clinical care.
Primary Care 4:247-253, 1977.
6.17 Coronary Drug Project Research Group: The Coronary Drug Project Aspirin Study: Implications
for clinical care. Primary Care 5:91-95, 1978.
Coronary Drug Project Research Group: The natural history of myocardial infarction in the
6.18
Coronary Drug Project: Long-term prognostic importance of serum lipid level. Am J Cardiol
42:489-498, 1978.
Coronary Drug Project Research Group: Clofibrate and gallbladder disease (letter). N Engl J Med
6.19
298:461, 1978.
Coronary Drug Project Research Group: Estrogens and cancer (letter). Z4M4 239:2758-2759,
6.20
1978.
Coronary Drug Project Research Group: Reply to editorial by L. Carlson (letter). Atherosclerosis
6.21
30:239-241, 1978.
6.22 Coronary Drug Project Research Group: Reply to letter from C. A. Caceres and K. Enslein. JAMA
240:1483-1484, 1978.
risk factor in men with a prior
6.23 Coronary Drug Project Research Group: Cigarette smoking as a
history of myocardial infarction. J Chronic Dis 32:415-425, 1979.
6 24 Coronary Drug Project Research Group: Influence of adherence to treatment and' response of
cholesterol on mortality in the Coronary Drug Project. N Engl J Med 303:1038-1041, 1980.
Coronary Drug Project Research Group:
risk factors-hypercholesterolemia,
smoking,
Group. Treatable
Treatable^risk
P™i™
t dafa
6.25
and hypertension—after myocardial infarction: Implications of the Coronary Drug Project data
for clinical management. Primary Care 7:175-179, 1980.
. Coronary Drug Project Research Group: Practical aspects of decision making m clinical trials: The
6.26
Coronary Drug Project as a case study. Controlled Clin Dials 1:363-376, 1981.
' Coronary Drug Project Research Group: Implications of findings in the Coronary Drug Project for
6.27
secondary prevention trials in coronary heart disease. Circulation 63:1342-1350, 1981.
I Coronary Drug Project Research Group: The natural history of coronary heart disease. Prognostic
6.28
factors after recovery from myocardial infarction m 2789 men. The 5-year findings of the
Coronary Drug Project. Circulation 66: 401-414. 1982.
Coronary Drug Project Research Group: High-density lipoprotein cholesterol and prognosis after
6.29
myocardial infarction. Circulation 66:1176-1178, 1982.
6.13
7. Aspirin Myocardial Infarction Study (AMIS)
Aspirin Myocardial Infarction Study Research Group: A randomized, controlled trial of aspirin in
7.1
persons recovered from myocardial infarction.
243:661-669, 1980.
Howard J, Whittemore AS. Hoover JJ, Panos M. and the Aspirin Myocardial Infarction Study
7.2
Research Group: Commentary: How blind was the patient blind in AMIS. Clin Pharmacol Ther
7.3
32:543-553, 1982.
Wasserman AG, Bren GB. Ross AM, Richardson DW Hutchinson RG. RiosJC: Pr°gnost*c
implications of diagnostic Q waves after myocardial infarction. Circulation 65.1451-1455. 1982.
8. Persantine Aspirin Rcinfarction Study (PARIS)
Persantine Aspirin Reinfarction Study Research Group: Persantine Aspirin Reinfarction Study:
8.1
Design, methods, and baseline results. Circulation 62(suppl II).Il 1 II 42, 198V.
322
Table B-3
8.2
i..
Table B-3 Publication list of sketched trials
B. Sketches of selected trials
Publication list of sketched trials (continued)
Persantine Aspirin Reinfarction Study Research Group: Persantine and aspirin in coronary heart
disease. Circulation 62:449-461, 1980.
9. Hypertension Detection and Follow-Up Program (HDFP)
9.1
Labarthe DR. Hawkins CM, Remington RD: Evaluation of performance of selected devices for
measuring blood pressure. Am J Cardiol 32:546-553, 1973.
9.2
Remington RD: The Hypertension Detection and Follow-Up Program. Institut National de la
Same et de la Recherche Medicale (1NSERM) 21:185-194, 1973.
9.3
Hypertension Detection and Follow-Up Program Cooperative Group (Borhani NO, Kass EH,
Langford HG, Payne GH, Remington RD, Stamler J): The Hypertension Detection and FollowUp Program. Prev Med 5:207-215, 1976.
9.4
Hypertension Detection and Follow-Up Program Cooperative Group: The Hypertension Detection
and Follow-Up Program: A progress report. Circ Res 40(suppl I):1-IO6—I 109, 1977.
9.5
Hypertension Detection and Follow-Up Program Cooperative Group (Castle CH, Daugherty S,
Detels R, Hawkins CM, Krishan I, Oberman A, Wassertheil-Smoller S): Blood pressure studies in
14 communities: A two-stage screen for hypertension. JAMA 237:2385-2391, 1977.
Hypertension Detection and Follow-Up Program Cooperative Group (Heymsfield S, Kraus J, Lee
9.6
ES, McDill M, Stamler R, Watson R): Race, education and prevalence of hypertension. Am J
Epidemiol 106:351-361, 1977.
9.7
Hypertension Detection and Follow-Up Program Cooperative Group (Apostolides A, Schnaper H,
Stamler R, Taylor J, Tyler M, Wassertheil-Smoller S): Patient participation in a hypertension
control program.
239:1507-1514. 1978.
Hypertension Detection and Follow-Up Program Cooperative Group (Apostolides A, Blaufox
9.8
MD, Borhani NO, Cutter G, Daughtery S, Lewin AJ, Polk BF): Mild hypertensives in the
Hypertension Detection and Follow-Up Program. Ann NY Acad Sci 304:254-266, 1978.
9.9
Hypertension Detection and Follow-Up Program Cooperative Group (Cutter G, Hebei JR, La
barthe D, Oberman A, Prineas R, Varady P): Variability of blood pressure and the results of
screening in the Hypertension Detection and Follow-Up Program. J Chronic Dis 31:651-667,
1978.
9.10 Hypertension Detection and Follow-Up Program Cooperative Group (Blaufox MD, Curb D,
Kralios A, Polk BF, Tyler M): Therapeutic control of blood pressure in the Hypertension
Detection and Follow-Up Program. Prev Med 8:2-13, 1979.
9.11 Hypertension Detection and Follow-Up Program Cooperative Group: Five-year findings of the
Hypertension Detection and Follow-Up Program: I. Reduction in mortality of persons with high
blood pressure, including mild hypertension.
242:2562-2571, 1979.
9.12 Hypertension Detection and Follow-Up Program Cooperative Group: Five-year findings of the
Hypertension Detection and Follow-Up Program: II. Mortality by race-sex, and age. JAMA
242:2572-2577, 1979.
9.13 Wassertheil-Smoller S, Apostolides A, Miller M, Oberman A, Thom T (on behalf of the Hyperten
sion Detection and Follow-Up Program): Recent status of detection, treatment, and control of
hypertension in the community. J Community Health 5:82-93, 1979.
9.14 Cowan L, Detels R, Farber M, Lee ES, McCray G, O’Flynn S, Parnell MJ (on behalf of the
Hypertension Detection and Follow-Up Program): Residential mobility and long-term treatment
of hypertension. J Community Health 5:159-166, 1980.
9.15 Cutter G, Heyden S, Kasteler J, Kraus JF, Lee ES, Shipley T, Stromer M (on behalf of the
Hypertension Detection and Follow-Up Program): Mortality surveillance in collaborative trials.
Am J Public Health 70:394 400, 1980.
9.16 Apostolides A, Cutter G, Daugherty SA, Detels R, Kraus J, Wassertheil-Smoller S, Ware J: Threeyear incidence of hypertension in thirteen U.S. communities. Prev Med 11:487-499, 1982.
9.17
9.18
Curb JD, Hardy RJ, Labarthe DR, Borhani NO, Taylor JO: Reserpine and breast cancer in the
Hypertension Detection and Follow-Up Program. Hypertension 4:307-311, 1982.
Hypertension Detection and Follow-Up Program Cooperative Group: Five-year findings of the
Hypertension Detection and Follow-Up Program: HI. Reduction in stroke incidence among
persons with high blood pressure. JAMA 247:633-638, 1982.
Table B-3
9.19
9.20
9.21
9.22
9.23
9.24
323
Publication list of sketched trials (continued)
Hypertension Detection and Follow-Up Program Cooperative Group: The effect of treatment on
mortality in “mild” hypertension—Results of the Hypertension Detection and Follow-Up Pro
gram. N Engl J Med 307:976-980, 1982.
Kraus JF, Conley A, Hardy R, Sexton M, Sweezey Z: Relationship of demographic characteristics
of interviewers to blood pressure measurements. J Community Health 8:3-12, 1982.
Shulman N, Cutter G, Daugherty R, Sexton M, Pauk G, Taylor MJ, Tyler M: Correlates of
attendance and compliance in the Hypertension Detection and Follow-Up Program. Controlled
Clin Trials 3:13-27, 1982.
Smith EO, Curb JD, Hardy RJ, Hawkins CM, Tyroler HA: Clinic attendance in the Hypertension
Detection and Follow-UP Program: Hypertension 4:710-715, 1982.
Hardy RJ, Hawkins CM: The impact of selected indices of antihypertensive treatment on all-cause
mortality. Am J Epidemiol 117:566-574, 1983.
Hypertension Detection and Follow-Up Program Cooperative Group (SA Daugherty, G Entwisle,
JD Curb, BF Polk, JO Taylor, editors): Hypertension Detection and Follow-Up Program:
Baseline characteristics of the enumerated, screened, and hypertensive participants. Hypertension
5(suppl IV):IV-1— 1V-205, 1983.
10. Multiple Risk Factor Intervention Trial (MRFIT)
Multiple Risk Factor Intervention Trial Group: Statistical design considerations in the NHLBI
10.1
Multiple Risk Factor Intervention Trial (MRFIT). J Chronic Dis 30:261-275, 1977.
Zukel WJ, Paul O, Schnaper HW: The Multiple Risk Factor Intervention Trial (MRFIT): I.
10.2
Historical perspective. Prev Med 10:387-401, 1981.
Sherwin R, Kaelber CT, Kezdi P, Kjelsberg MO, Thomas HE, Jr.: The Multiple Risk Factor
10.3
Intervention Trial (MRFIT): II. The development of the protocol. Prev Med 10:402-425, 1981.
Benfari RC: The Multiple Risk Factor Intervention Trial (MRFIT): III. The model for intervention.
Prev Med 10:426-442, 1981.
Caggiula AW, Christakis G, Farrand M, Hulley SB, Johnson R, Lasser NL, Stamler J, Widdowson
10.5
G: The Multiple Risk Factor Intervention Trial (MRFIT): IV. Intervention on blood lipids. Prev
Med 10:443-475, 1981.
Hughes GH, Hymowitz N, Ockene JK, Simon N, Vogt TM: The Multiple Risk Factor Intervention
10.6
Trial (MRFIT): V. Intervention on smoking. Prev Med 10:476-500, 1981.
Cohen JD, Grimm RH, Jr., Smith WM: Multiple Risk Factor Intervention Trial (MRFIT): VI.
10.7
Intervention on blood pressure. Prev Med 10:501-518, 1981.
Neaton JD, Broste S, Cohen L. Fishman EL, Kjelsberg MO, Schoenberger J: The Multiple Risk
10.8
Factor Intervention Trial (MRFIT): VII. A comparison of risk factor changes between the two
study groups. Prev Med 10:519-543, 1981.
Benfari RC, Sherwin R: The Multiple Risk Factor Intervention Trial after 4 years: A summing-up.
10.9
Prev Med 10:544-546, 1981.
10.10 Multiple Risk Factor Intervention Trial Research Group: Multiple Risk Factor Intervention Trial:
Risk factor changes and mortality results. J/4AL4 248:1465-1477, 1982.
10.4
11. Coronary Artery Surgery Study (CASS)
Kronmal RA, Davis K, Fisher LD, Jones RA, Gillespie MJ: Data management for a large
11.1
collaborative clinical trial (CASS: Coronary Artery Surgery Study). Comput Biomed Res
11:553-566, 1978.
Davis K, Kennedy JW, Kemp HG, Jr., Judkins MP, Gosselin AJ, Killip T: Complications of
11.2
coronary arteriography from the Collaborative Study of Coronary Artery Surgery (CASS).
Circulation 59:1105-1112, 1979.
11.3
Weiner DA, Ryan TJ, McCabe CH, Kennedy JW, Schloss M, Tristani F, Chaitman BR, Fisher LD:
Exercise stress testing: Correlations among history of angina, ST-segment response and preva
lence of coronary artery disease in the Coronary Artery Surgery Study (CASS). N Engl J Med
301:230-235, 1979.
324
Table B-3 Publication list of sketched trials (coniinued)
Chaitman BR. Rogers WJ, Davis K. Tyras DH. Berger R, Bourassa MG, Fisher L, Hertzberg VS,
Judkins MP. Mock MB, KillipT: Operative risk factors inpatients with left main coronary artery
disease. N Engl J Med 303:953-957, 1980.
Kennedy JW, Kaiser GC, Fisher LD. Maynard C, Fritz JK, Myers W, Mudd JG, Ryan TJ, Coggin
11.5
J. Multivariate discriminant analysis of the clinical and angiographic predictors of operative
mortality from the Collaborative Study in Coronary Artery Surgery (CASS). J Thorac Cardiovasc Surg 80:876-887, 1980.
Vlietstra RE, Frye RL, Kronmal RA. Sim DA, Tristani FE. Killip T 111, and participants in the
11.6
Coronary Artery Surgery Study: Risk factors and angiographic coronary artery disease: A report
from the Coronary Artery Surgery Study (CASS). Circulation 62:254-261, 1980.
Chaitman BR, Bourassa MG, Davis K, Rogers WJ, Tyras DH, Berger R, Kennedy JW, Fisher LD,
11.7
Judkins MP. Mock MB, KillipT: Angiographic prevalence of high-risk coronary artery disease m
patient subsets (CASS). Circulation 64:360-367, 1981.
Chaitman BR, Fisher L, Bourassa MG, Davis K, Rogers WJ, Maynard C, Tyras DH, Berger RL,
11.8
Judkins MP. Ringqvist 1, Mock MB, Killip T: Effect of coronary bypass surgery on survival
patterns in subsets of patients with left main coronary artery disease. Report of the Collaborative
Study in Coronary Artery Surgery (CASS). Am J Cardiol 48:765-777, 1981.
Fisher LD. Kennedy JW, Chaitman BR, Ryan TJ, McCabe C, Weiner D. Tristani F, Schloss M,
11.9
Warner HR: Diagnostic quantification of CASS (Coronary Artery Surgery Study) clinical and
exercise test results in determining presence and extent of coronary artery disease. Circulation
63:987-1000. 1981.
11.10 Berger RL, Davis KB, Kaiser GC, Foster ED, Hammond GL, Tong TGL, Kennedy JW. Scheffield
T. Ringqvist I, Wiens RD, Chaitman BR, Mock M: Preservation of the myocardium during
coronary artery bypass grafting. Circulation 64(suppl 1I):II—61—11-66, 1981.
11.11 Kennedy JW, Kaiser GC, Fisher LD, Fritz JB, Myers W, Mudd JG, Ryan TJ: Clinical and
angiographic predictors of operative mortality from the Collaborative Study in Coronary Artery
Surgery (CASS). Circulation 63:793-802, 1981.
11.12 Principal Investigators of CASS and their associates, T Killip (editor), LD Fisher, MB Mock
(associate editors): The National Heart. Lung, and Blood Institute Coronary Artery Surgery
Study (CASS). Circulation 63(suppl 1):I-I—1-81, 1981.
11.13 Aiderman EL, Fisher L, Maynard C, Mock MB. Ringqvist I, Bourassa MG, Kaiser GC, Gillespie
MJ: Determinants of coronary surgery in a consecutive patient series from geographically
dispersed medical centers: The Coronary Artery Surgery Study.Circulation 66(suppl I):
I-6-I-15, 1982.
11.14 Faxon DP, Ryan TJ, Davis KB. McCabe CH, Myers W, Lesperance J, Shaw R, Tong TGL:
Prognostic significance of angiographically documented left ventricular aneurysm from the
Coronary Artery Surgery Study (CASS). Am J Cardiol 50:157-164, 1982.
11.15 Fisher LD, Judkins MP, Lesperance J, Cameron A, Swaye P, Ryan T, Maynard C, Bourassa M,
Kennedy JW, Gosselin A, Kemp H, Faxon D, Wexler L, Davis KB. Reproducibility of coronary
arteriographic reading in the Coronary Artery Surgery Study (CASS). Cathet Cardiovasc Diagn
8:565-575. 1982.
11.16 Fisher LD. Kennedy JW, Davis KB, Maynard C, Fritz JK, Kaiser G, Myers WO: The association of
sex, physical size, and operative mortality after coronary artery bypass in the Coronary Artery
Surgery Study (CASS). J Thorac Cardiovasc Surg 84:334-341, 1982.
11.17 Kemp H, Davis K, Judkins MP, Gosselin A, Kennedy JW, Cameron A, Swaye PS, Maynard C,
Fisher LD: Intrareader variability in the interpretation of coronary arteriograms from the
Coronary Artery Surgery Study (CASS). Ischemic heart disease: Second USA-USSR Sympo
sium, Seattle, Washington, March 20, 1981. Kardiologiia 22:37-42, 1982.*
11.18 Kennedy JW, Killip T, Fisher LD, Aiderman EL, Gillespie MJ. Mock MB: The clinical spectrum of
coronary artery disease and its surgical and medical management, 1974-1979: The Coronary
Artery Surgery Study. Circulation 66(suppl II1):IILI6-111-23, 1982.
11.19 Killip T: Indications for coronary arteriography. Ischemic heart disease: Second USA-USSR
Symposium, Seattle, Washington, March 20, 1981. Kardiologiia 22:33-37, 1982.*
11.4
I'
I
I
I
•
Table B-3 Publication list of sketched trials
B. Sketches of selected trials
Table B-3 Publication list of sketched trials (continued)
325
__________ __________________
Lundberg ED, McBride R, Rawson TE, Mauritsen R. Ormond TH, Fisher LD, Kronmal RA,
11.20
Gillespie MJ: C2: A data base management system developed for the Coronary Artery Surgery
Study (CASS) and other clinical studies. J Med Systems 6:501-518. 1982.
Mock MB, Ringqvist I, Fisher LD, Davis KB, Chaitman BO, Kouchoukos NT, Kaiser GC,
11.21
Aiderman E, Ryan TJ, Russell RO Jr, Mullin S, Fray D, Killip T HI, Participants in the
Coronary Artery Surgery Study: Survival of medically treated patients in the Coronary Artery
Surgery Study (CASS) Registry. Circulation 66:562-568, 1982.
Rogers WJ. Chaitman BR, Fisher LD, Bourassa MG, Davis K, Maynard CL, Tyras DH, Berger
11.22
RL, Judkins MP, Ringqvist I, Mock MB, Killip T: Comparison of the cumulative survival of
medically and surgically treated patients with left main coronary artery disease. The CASS
experience. Ischemic heart disease: Second USA-USSR Symposium. Seattle, Washington. March
20, 1981. Kardiologiia 22:53-57, 1982.*
Ryan TJ, Fisher LD, Weiner DA. McCabe CH, Chaitman B. Kennedy JW, Ferguson J, Tristani F:
11.23
Experience with electrocardiographic exercise testing in the Coronary Artery Surgery Study
(CASS). Ischemic heart disease. Second USA-USSR Symposium, Seattle. Washington, March 20,
11.24
11.25
11.26
11.27
11.28
11.29
)
11.30
11.31
1981. Kardiologiia 22:22-26, 1982.*
Vlietstra RE. Kronmal RA, Seth A. Frye RL: Correlation of risk factors for coronary artery disease
with coronary angiographic findings. Ischemic heart disease: Second USA-USSR Symposium,
Seattle, Washington, March 20, 1981. Kardiologiia 22:67-72, 1982.*
Vlietstra RE, Kronmal RA, Frye RL, Seth AK, Tristani FE, Killip T III: Factors affecting the
extent and severity of coronary artery disease in patients enrolled in the Coronary Artery Surgery
Study. Arteriosclerosis 2:208-215, 1982.
Weiner DA McCabe CH, Ryan TJ, Chaitman BR, Sheffield LT, Ferguson J, Fisher LD: Assess
ment of the negative exercise test in 4373 patients from the Coronary Artery Surgery Study
(CASS). J Cardiac Rehab 2:562-568, 1982.
Wexler LF, Lesperance J, Ryan TJ, Bourassa MG, Fisher LD, Maynard C, Kemp HG, Cameron A,
Gosselin AJ, Judkins MP: Interobserver variability in interpreting contrast left ventriculograms
(CASS). Cathet Cardiovasc Diagn 8:341-355, 1982.
Zimmern SH, Rogers WJ. Bream PR, Chaitman BR. Bourassa MG Davis KA, Tyras DH. Berger
R. Fisher L, Judkins MP. Mock MB. Killip TA: Total occlusion of the kfla^ery
The Coronary Artery Surgery Study (CASS) experience. Am J Cardio! 49:2003-2010. 1982.
Chaitman BR. Aiderman FL. Sheffield LT. TongT, Fisher LD. Mock MB. Wiens RD Kaiser GC.
Roitman D, Berger R, Gersh B, Schaff H. Bourassa MG, Killip T: Use of survival analysis to
determine the clinical significance of new Q waves after coronary bypass surgery. Circulation
67:302-309, 1983.
Gersh BJ, Kronmal RA. Frye RL. Schaff HV. Ryan TJ. Gosselin AJ, Kaiser GC Killip T:
Coronary arteriography and coronary artery bypass surgery: Morbidity and morality' Pat,™ts
65 or older. A report from the Coronary Artery Surgery Study. Circulation 67.483-491. 1983.
Swaye PS, Fisher LD, Litwin P. Vignola PA. Judkins MP, Kemp HG, Mudd JG, Gosselin AJ:
Aneurysmal coronary artery disease. Circulation 67:134-138, 1983.
... on
_.i the Surgical Control of Hyperlipidemia (POSCH)
12. Program
Buchwald H, Moore RB, Varco RL: Maximum lipid reduction by partial ileal bypass: A test of the
12.1
lipid-atherosclerosis hypothesis. Lipids 12:53-58, 1977.
Moore RB, Long JM, Matts JP, Amplatz K, Varco RL, Buchwald H. and the POSCH Group:
12.2
Plasma lipoproteins and coronary arteriography in subjects in the Program on the Surgical
Control of the Hyperlipidemias. Atherosclerosis 32:101-119. 1979.
Long JM, Brashear JR. Matts JP. Bearman JE, and the POSCH Group: The POSCF> information
12.3
management system: Experience with alternative approaches. J Med Syst 4.355-366. 1980.
Moore RB. Buchwald H, Varco RL. and the Participants in the Program on the Surgical Control of
12.4
the Hyperlipidemias:
Hvnerlinidemias. The
The effect
effect of
of partial
partial ileal
ileal bypass
bypass on plasma lipoproteins. Circulation
the
12.5
62:469-476, 1980.
Buchwald H, Fitch L, Moore RB: Overview of randomized clinical trials of lipid intervention for
atherosclerotic cardiovascular disease. Controlled Clin Trials 3:271-283, 1982.
2
?
5
5‘
i
? 2
5
£
“3r g 1
* * 2
. 2 . 2 S‘
2 5 2. 3 5
CO
QB
co
I
'Ifln?
I
lI
I if 1 5 I *
l
gr 5|.’ s|
5- ”
> is
r
2,
■Q
? n2 >*? I
y, S'
5
I
-
=»
OO 2
I
I
Pf
2 3
g S
s
73
n
•
3 «r=2 ? =
§
55‘
-9 o3-*
£.2 1*2 =
?
s-n?
~ x S- - S B
2>" ’r"
£■ 5 □ -< 3
J L
!7)
s. 2,2
2
9-. 5 S..
2 '■£ 2_
f
as isa
I
n’ 9- o’
I
3 ?3
non
2
2
^3-3 2-o±:
r;t i--2
o J n
£ s-g
2 QS
2,
Sr
hii...
8:2-3 : srn
riMH-
"
=;
r»
M
2
o
■
=r
o
tu
3
NJ o
LZl —i
1 Bl
<n '_ n<
oo n
C
“ C
3 S
F O.s
3 ^3
o
Q. n 7 M
00 2-^ ^in =
I I -I
I
i I
—>
£ ? si”
i? =
o3
3 75 ■ 3 >
o
2,P
3 "
Table B-4 Summary tabulations from sketches
PHS UGDP VA43
Item
1
MPS NCOS CDP
2
3
4
5
n/
y/
y/
7
6
Acronym and sketch no.
AMIS PARIS HDFP MRFIT CASS POSQI HPT
IRSC
13
14
Freq.
v/
11
7
8
9
10
11
12
s/
y/
1. Type of trial
Therapeutic
y/
y/
3
y/
y/
Prophylactic
2. Type of design
y/
'_aJ
y/
y/
y/
y/
y/
Fixed sample
>/
y/
V
y/
V
y/
V
14
3. Stage
0
Initial design
0
Protocol development
Patient recruitment
3
y/
V
y/
Treatment and follow-up.
7
y/
3
0
Patient close-out
y/
y/
3
Termination
V
y/
Post-trial follow-up
y/
Completed
y/
y/
a
o 73 “• c/i 3
"
3
O
n
LU
NJ
O'
2
3
i
*
Table B-4 Summary tabulations from sketches (continued)
PUS
hem
UGDP VA43
I
MPS
4
3
A crony m and sketch no.
NCGS CDP ~ AMIS PARIS HD! P MRHT CASS POSQI HPT
5
6
7
8
9
10
x/
x/
II
12
13
IRSC
14
Freq.
4. Funding source
x/
NIH
x/
VA
x/
Industry
bj
oo
12
x/
I
Private foundation.
x/
1
x/
8
5. Funding type
Grant.
x/
x/
x/
x/
x/
x/
Contract.
x/
x/
x/
x/
5
v/*
Neither.
I
6. Mode of initiation
Investigator.
x/
x/
x/
x/
x/
X/
Sponsor.
x/
Sponsor-investigator.
x/
x/
x/
7
x/
5
x/
2
Table B-4 Summary tabulations from sketches (continued)
PHS UGDP VA43
hem
'jj
bJ
7. Mode of fund dispersal
Direct to individual
centers...................
Indirect via a consort
ium award............
Direct to some centers
indirect to others.
8. Funding date
Start
End
1
2
NA
MPS NCOS CDP
3
4
x/
x/
5
Acronym and sketch no.
AMIS PARIS HDFP MRFIT CASS POSCH HPT
6
7
x/
yj
9
10
11
y/
x/
12
IRSC
14
x/
No. outside U.S.
Total no.
4
1
x/
Sept
1981
Dec
1986
Sept
1960
April
1982
Oct
1976
Oct
1983
Jan
1979
Dec
1986
June
1973
Oct
1983
Mar
1965
April
1984
Oct
1974
Nov
1979
April
1974
Sept
1980
June
1971
June
1985
June
1972
June
1985
May
1973
May
1988
June
1973
Not
set
Sept
1981
Aug
1986
July
1979
Not
set
NA
12
II
12
10
53
30
16
14
22
13
4
4
18
NA
0
0
0
0
0
0
4
0
0
1
0
0
8
NA
12
II
12
10
53
30
20
14
22
14*
4
4
26
9. Clinics
No. in U.S. (incl P R.)
Freq.
7
yJ
NA
13
yj
x/
x/
NA
8
Table B-4 Summary tabulations from sketches (continued)
PHS UGDP VA43
Hem
MPS NCGS CDP
Acronym and sketch no.
AMIS PARIS HDFP MRHT CASS posai HPT
IRSC
1
2
3
4
5
6
7
8
9
10
II
12
13
14
Freq.
NA
NA
1
NA
1
NA
NA
NA
NA
NA
NA
NA
I
NA
3
NA
NA
NA
NA
NA
NA
I
NA
3
10. Resource centers
Data coord, center.
Trt. coord, center.
Coord, center.
Project office.
I
NA
1
NA
NA
NA
1
NA
1
NA
1
1
1
1
1
1
I
NA
!•
II
1
I
I
I
I
I
0
1
1
1
1
1
1*
13
1
2
2
0
2
1
1
I
1
1
0
I
1
0
11
0
1
0
I
3
1
1
1
1
2
1
2
1
2*
12
0
0
0
0
0
0
0
1
0
1
0
0
0
0
2
0
0
1
0
0
1
1
1
1
1
0
0
0
0
6
3
4
6
3
8
5
5
5
5
7
3
5
5
4*
14
CU
O
Central laboratories.
Reading centers.
Quality control center.
Procurement and
distribution center...
Total no.
/ /. Coord, center or data
coord, center
a. Location
School of public health...
School of medicine.
4
s/
x/
x/
x/
7
x/
Other teaching inst.
1
x/
Nonteaching inst.
x/
2
I
Table B-4 Summary tabulations from sketches (continued)
PUS UGDP FA43
hem
11. CC or data CC (cant’d)
b. Primary degree of
director.....................
c. Affiliation with other
study centers
UJ
Yes.
Acronym and sketch no.
MPS XCGS CDP ~ AMIS PARIS HDFP MRFIT CASS POSC7I HPT
IRSC
14
4
5
1
2
3
MD
(Epi)
PhD
(Bio)
PhD MSc ScD
(Bio) (Bio) (Bio)
NA
7
8
9
10
H
12
13
PhD
(Bio)
PhD
(Bio)
MD
(Epi)
ScD
(Bio)
PhD
(Bio)
PhD
(Bio)
EdD
(Bio)
MD
PhD
(Bio) |(Med)
11
Bio
V*
x/*
x/*
5
x/'
x/*
NA
y/
y/
y/
>/
x/
X/
x/
x/
x/
Freq.
6
x/
x/
V
x/
x/
V
x/
V
x/
x/
x/
V
x/
8
No.
12. Type of treatment design
Noncrossover.......
13. Type of treatment
structure
x/
x/
Simple.
x/
Partial factorial
Full factorial.
x/
x/
x/
x/
14
x/
11
x/
I
-
Summary tabulations from sketches (continued)
Table B-4
Acronym and sketch no.
hem
PUS
UGDP
VA43
MPS
MCGS
I
2
3
4
5
3
3
1
2
I
2
1
4
5
2
y/
y/
y/
GDP
AMIS PARIS UDI P MRFIT CASS POSCH
HPT
IRSC
9
K)
11
12
13
14
2
1
1
1
1
4
1
1
1
1
1
1
I
I
2
3
2
2
2
2
5
2
yJ
y/
6
7
8
2
5
I
I
1
I
3
3
6
Freq.
14. Study treatments
No. of test treatments,
No. of control treatments...
GJ
GJ
NJ
Total no. of study trts.
75. Type of test treatment
Drug..
y/
9
y/
y/
Surgery.
y/
Behavior change.
4
y/
2
16. Type of control treatment
y/
y/
Placebo pills.
y/
y/
y/
y/
y/
7
y/
Standard med. trt.
y/
V
6
y/
y/
No treatment.
Table B-4
3
Summary tabulations from sketches (continued)
Acronym and sketch no.
Item
PHS
UGDP
VA43
MPS
NCGS
CDP
1
2
3
4
5
6
~ AMIS PARIS HDFP MRFIT CASS
7
8
9
10
y/
y/
11
posai HPT
IRSC
12
13
14
Freq.
y/
y/
V
8
17. Level of trt. masking
y/
y/
None.
0
Single-masked......
y/
y/
y/
y/
y/
y/
7
y/
y/
v/
y/
y/
y/
v/
y/
y/
y/
y/
y/
y/
y/
V
Double-masked.
18. Patients studied
y/
V
y/
y/
14
y/
y/
y/
10
y/
1
Males.
y/
V
Females.
Children.
y/
y/
xZ
xZ
y/
y/
y/
y/
y/
yJ
y/
y/
y/
y/
\Z
xZ
y/
x/
13
y/
Adults.
19. Patient recruitment
a. Primary mode of contact
y/
yJ
10
Direct.
\Z
Indirect.
y/
xZ
y/
y/
y/
y/
7
’ 5
Table B-4 Summary tabulations from sketches (continued')
MPS NCOS CDP
1
2
3
4
5
6
7
8
9
10
Mid
1982
Mid
1983
Feb
1961
Feb
1966
Mar
1977
May
1980
Mar
1979
Not
set
Sept
1976
June
1978
Mar
1966
Oct
1969
May
April
1975
Sept
1976
Feb
1973
May
1974
Nov
1973
Feb
1976
17,350 1,027
231
756
916
V
x/
Item
19. Patient recruitment (cant’dI
b. Start of enrollment.
c. End of enrollment.
d. Total no. enrolled.
LU
LU
Acronym and sketch no.
AMIS PARIS HDFP MRFTT CASS POSCH HPT
PHS UGDP VA43
IRSC
II
12
13
14
Aug
1975
May
1979
Sept
1975
June
Sept
1982
Oct
1983
1983
Jan
1980
Not
set
8,341 4,524 2,026 10,940 12,866 780
838
235
260*
\/
V
x/
V
1975
Aug
1976
Freq.
20. Method of randomization
a. Type
Fixed allocation ratio.
x/
\/
V
13
x/
Number adaptive..
1
b. Stratification variables
Clinic
Disease state or type
Demographic
characteristics..
Other
Na
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
No
Yes
Yes
No
Yes
No
No
Yes
No
Yes
Yes
No
Yes
Age
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
Yes*
Yes
1
2
2
1
2
1
1
2
1
3
4
2
MPS
NCGS
CDP
c. Total no. of variables
1
controlled...............
Wt
13
Yes
7
Yes
Yes
2
Age,sex Yes
2
No
Yes
0
4*
Table B-4 Summary tabulations from sketches (continued)
PHS UGDP VA43
Item
20. Randomization (cont'd)
d. Total no. of allocation
strata. ..........................
e. Allocation ratio.
Acronym and sketch no.
AMIS PARIS HDFP MRF1T CASS POSCH HPT
IRSC
1
2
3
4
5
6
7
8
9
10
11
12
13
14
7
12
20
36
10
106
30
20
42
22
66
48
8
216*
Uni
form
Non
uni
Uni
form
Uni
form
Uni
form
Non
uni
Uni
form
Non
uni
Uni
form
Uni
form
Uni
form
Uni
form
Uni
form
Uni
form
II
Uni
x/
x/
x/
x/
x/
x/
x/
x/
x/
x/
x/
x/
x/
13
Freq.
f. Blocking in strata
LU
Yes.
1
x/
No.
g. Locus of control
Central.
x/
x/
x/
x/
x/
x/
x/
x/
Local.
h. Method of release
From control center to
clinic via phone
From control center to cli
nic via sealed schedule.....
From clinic via self
administered schedule....
From clinic via on-site
micro-computer..............
From control center direct
to patient.........................
x/
x/
x/
x/
x/
x/
x/
x/
x/
x/*
x/
7
5
x/
I
x/
x/
x/
13
1
x/
1
1
Table B-4 Summary tabulations from sketches (coniinued')
PHS UGDP VA43
2
3
NA*
2
a. Baseline clinic visits......
b. Regular follow-up clinic
4
NA*
visits/year.....................
2
1
Item
MPS NCOS CDP
4
Acronym and sketch no.
AMIS PARIS HDFP MRFIT CASS posai HPT
IRSC
5
6
7
8
9
10
11
12
13
14
2
3
3
2
2
3
1
2
3
1-2
3
0-3*
I*
2
1
2
1
Freq.
21. Data collection schedule
LU
LU
O'
4
2
3-6*
3
3
22. Length of follow-up (yrs.)
4.5
9.514.5
3-6
2+
2
5-8
3-4+
3-4+ 5-6.5
6-8
4-8
>5
>2
Not
set
NA
2
NA
NA
NA
6+
None
done
None
done
2-4
NA
NA
NA
NA
a. During trial.
b. Post-dial.
3
23. Primary outcome
\/
V
Deaths all causes.........
Deaths from specified
cause....................
8
2
4
Other.
Table B-4 Summary tabulations from sketches (continued)
PHS UGDP VA43
Item
1
2
3
Acronym and sketch no.
MPS NCOS CDP~ AMIS |P4/?/S|/7DFP|MR/7T| CASS
4
5
6
7
8
9
10
y/
y/
11
12
HPT
IRSC
13
14
Freq.
24. Method of follow-up close-out
yj
Common calendar date.
Common period of
follow-up............
10
y/
1
y/
y/
V
3
y/
12
Not yet specified.....
LU
25. Data entry
a. Primary entry site
y/
y/
y/
y/
y/
y/
y/
y/
y/
y/
y/
CC or data CC.
2
y/
y/
Clinic.
b. Primary mode of entry
Direct from forms...........
Indirect from code sheets
prepared from forms.
yj
y/
y/
y/
x/
v'
y/
V
y/
y/
y/
y/
0
26. Sample size specification
a. Original recruitment goal 21,500 1,000
(all trts. combined).
14
456
522*
900
8,379 4,250 2,000 10,500 11,000 1,500 1,000
800
250*
z
Table B-4
Summary tabulations from sketches (continued)
PHS UGDP VA43
Item
3
1
MPS NCGS CDP
4
5
6
Acronym and sketch no.
AMIS PAR1S\HDFP MRHT CASS POSGI HPT
7
8
9
10
11
12
y/
y^
y/
y/
13
IRSC
14
Freq.
y/
11
26. Sample size (cont'd)
b. Rationale for original recruitment goal
Sample size calculation.
x/
Pragmatic.
LU
LU
QO
y/
3
c. Achieved sample size
All trts. combined
NA
1,027
231
236*
916
8,341
4,524
2,026 10,940 12,866
780
NA
NA
NA
yJ
y/
y/
y/
d. Published rationale for achieved
sample size
Not applicable.
y/
1
None required.
0
None stated.
yJ
y/
y/
y/
4
y/
4
Power calculation.
y/
y/
Sample size calculation.
y/
Pragmatic.
Table B-4
6
1
Summary tabulations from sketches (continued)
Acronym and sketch no.
PHS
Item
UGDP VA43
MPS
NCGS
CDP ' AMIS PARIS HDFP MRFIT CASS
posai HPT
1RSC
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
Yes
No
No
No
No
No
No
Yes
Yes
No
Yes
No
Yes
No
NA
Yes
Yes
Yes
NA
Yes
Yes
NA
NA
Yes
Yes
Yes
No
Yes
NA
No
Yes
Yes
NA
Yes
Yes
NA
NA
Yes
Yes
No
NA
NA
Yes
NA
NA
NA
Yes
NA
NA
Yes
Yes
NA
NA
NA
2
10
2
1
7
11
8
1
8
8
4
3
3
0
4
13
5
3
10
14
10
4
11
11
7
6
6
2
MD
MD
MD
MD
MD
MD
MD
MD
MD
MD
MD
MD
MD
MD
Yes
No
Yes
No
No
No
No
Yes
No
No
Yes
Yes
Yes
E
E
WT
WT
27. Committees represented
Steering committee
Executive committee......
Treatment effects moni
toring committee....
Advisory review committee.....
Advisory review & treatment
effects monitoring comm. .
No. of other comrnittees.
Freq.
14
Yes
5
Yes
9
Yes
7
Yes
4
Yes
Total no. of committees.
28. Steering committee
a. Chairman
i. Primary degree.............
ii. Patient care responsi
No
bilities...........................
Selfappt
iii. Elected or appointed
WT
iv. Term of office
E
WT
E
WT
Self
appt,
Selfapp^
Appt
Appt
Appt
E
Appt
Appt
Selfappt
WT
WT
WT
WT
WT
WT
WT
WT
WT
14
MD
6
Yes
5
E
TF
WT
Table B-4 Summary tabulations from sketches (continued)
PHS UGDP VA43
Item
MPS NCGS CDP
Acronym and sketch no.
AMIS PARIS] HDFP MRF1T CASS POSCH HPT
IRSC
I
2
3
4
5
6
7
8
9
10
II
12
13
14
MPH
NA
NA
NA
NA
MD
MD
MD
MD
MD
NA
NA
MD
MD
No
NA
NA
NA
NA
Yes
Yes
Yes
No
No
NA
NA
Yes
Yes
Appt
NA
NA
NA
NA
Appt
Appt
E
E
Appt
NA
NA
E
E
WT
NA
NA
NA
NA
WT
WT
WT
WT
WT
NA
NA
WT
WT
7
26
14
12
14
14
11
9
19
39
17
6
10
10
4
1
0
0
1
0
7
0
0
0
2
0
2
0
11
27
14
12
15
14
18
9
19
39
19
6
12
10
Freq.
28. Steering committee (cont'd)
b. Vice-chairman
i. Primary degree...........
ii. Patient care responsibilities
iii. Elected or appointed
o
iv. Term of office.
7
MD
6
Yes
4
E__
8
WT
c. Membership
i. From study centers
Voting.
Nonvoting.
Total.
ii. From outside study centers
Voting.
Nonvoting.
Total.
5
0
0
I
2
I
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
I
0
5
0
0
1
2
1
0
3
0
0
0
0
I
0
Table B-4 Summary tabulations from sketches (continued)
PHS UGDP VA43
Acronym and sketch no.
MPS NCGS CDP ~ AMIS PARIS HDFP MRF1T CASS POSCH HPT
IRSC
4
5
6
7
8
9
10
11
12
13
14
14
12
26
12
Voting.................................... ..
0
1
4
I
16
15
11
11
19
39
17
6
10
10
1
0
7
1
0
0
2
0
3
0
Item
1
2
3
28. Steering committee (cont'd)...
c. Membership (cont'd)
iii. Total
Nonvoting
16
27
14
13
17
15
18
12
19
39
19
6
13
10
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
NA
NA
NA
NA
Yes
Yes
Yes
Yes
Yes
NA
NA
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
NA
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes*
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
NA
Yes
Yes
Yes
No
Yes
No
NA
No
No
Yes
No
No
Yes
No
No
Yes
No
Yes
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
Freq.
Total
d. Positions represented
on committee
i. Study chairman.
ii. Vice-chairman
iii. CC or data CC director.
iv. Clinic directors
Project officer
vi. Clinic coordinators
vii. Nonhealth professional
viii. Lay representative
14
Yes
8
Yes
14
Yes
13
Yes
10
Yes
4
Yes
0
Yes
0
Yes
Table B-4 Summary tabulations from sketches (continued)
PHS UGDP VA43
MPS NCGS CDP
Item_____________________
1
2
3
4
28. Steering committee (cont'd)
e. No. of members elected
0
0
0
0
A crony m and sketch no.
AMIS PARIS HDFP MRHT CASS POSCH HPT
IRSC
5
6
7
8
9
10
11
12
13
14
0
4.T
5.T
0
0
4.WT
0
0
0
0
y/
y/
y/
y/
7
Freq.
29. Treatment effects monitoring committee
a. Responsible group
Trt. effects monitoring
comm.................................
Advisory review & trt.
effects monitoring comm...
y/
y/
yj
9
4
1
Steering committee.
b. Chairman
MD
(Epi)
i. Primary degree..........
ii. Patient care responsi
No
bilities..................
Appt
iii. Elected or appointed
iv. Term of office.
WT
MD
MD
PhD
MD PhD*
MD
MD*
(Med) (Med) (Bio) (Med) (Bio) (Med) (Epi)
MD
(Epi)
MD
MD
MD
(Med) (Med) (Epi)
PhD
(Bio)
PhD
(Bio)
Yes
No
No
No
No
No
No
No
No
No
E
E
Appt
Appt Appt* Appt Appt*
E
Appt
Appt
Appt
E
Appt
WT
WT
WT
WT
WT
WT
WT
WT
WT
WT
No*
No
WT*
WT
No*
WT*
10
MD
1
Yes
10
Appt
14
WT
Table B-4 Summary tabulations from sketches (continued)
PHS UGDP VA43
Item
1
2
3
29. Treatment effects monitoring committee (cont'd)
c. Vice-chairman
NA
NA
NA
Primary degree...
Patient care
NA
NA
NA
responsibilities
NA
NA
NA
Elected or appointed
NA
NA
NA
Term of office.....
d. Membership
i. From study centers
0
26
1
Voting................
4
1
1
Nonvoting.
27
4
2
Total.
ii. From outside study
Voting................
Nonvoting.
Total.
MPS NCGS CDP
Acronym and sketch no.
AMIS PARIS HDFP MRFTT CASS posai HPT
IRSC
4
5
6
7
8
9
10
11
12
13
14
NA
NA
MD*
(Epi)
NA
MD*
(Epi)
NA
NA
NA
NA
NA
NA
NA
NA
No*
NA
No*
NA
NA
NA
NA
NA
NA
NA
NA
Appt*
NA
Appt*
NA
NA
NA
NA
NA
NA
WT*
NA
NA
NA
NA
NA
NA
5
5
0
2
0
7
0
0
4
NA
NA
WT*
NA
0
0
II
0
6
7
0
I
0
2
2
0
2
6
7
11
1
5
7
2
2
2
7
4
3
0
4
5
5
5
7
3
6
7
11
8
0
4
0
0
0
0
0
0
0
1
0
0
0
1
0
0
3
0
4
5
5
5
7
4
6
7
11
9
0
4
Freq.
2
MD
2
No
2
Appt
2
WT
I
Table B-4 Summary tabulations from sketches (continued)
PHS UGDP VA43
2
1
Item
MPS NCOS CDP
3
29. Treatment effects monitoring committee (cont’d)
d. Membership (cont’d)
iii. Total
4
26
4
Voting.
4
1
1
Nonvoting.
5
27
8
Total.
Acronym and sketch no.
AMIS PARIS HDFP MRJ-1T CASS posai HPT
IRSC
4
5
6
9
10
11
12
13
14
5
5
--------- ------------------7
16
8
11
7
13
8
7
4
6
7
0
1
1
2
2
0
3
0
.4
11
12
16
8
9
13
9
13
11
7
8
7
8
Freq.
e. Positions represented
on committee
i. Study chairman
ii. Vice-chairman.
iii. CC or data CC director.
iv. Clinic directors
v. Project officer.
vi. Clinic coordinator.
vii. Nonhealth professional..
viii. Lay representative
No
Yes
Yes
Yes
Yes
Yes
Yes*
Yes
Yes
Yes*
Yes
Yes
Yes
Yes
No
NA
NA
NA
NA
Yes
No
No
Yes
Yes*
NA
NA
Yes
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes*
Yes
Yes
Yes*
Yes
Yes
Yes
Yes
NA
Yes
No
Yes*
No
Yes*
No
No
Yes
No
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
NA
Yes
Yes
Yes*
Yes
Yes
Yes
NA
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
Yes
No
No
No
No
No
No
Yes
No
No
No
No
No
No
NA
No
No
No
No
Yes
No
No
No
No
No
Acronym and Sketch no.
AMIS PARIS HDFP MRFIT CASS POSOI HPT
IRSC
13
14
13
Yes
5
Yes
13
Yes
6
Yes
13
Yes
0
Yes
2
Yes
1
Yes
Table B-4 Summary tabulations from sketches (continued)
PHS UGDP VA43
Item
2
1
3
MPS NCOS CDP
4
5
6
7
8
9
10
11
12
Freq.
29. Treatment effects monitoring committee (cant'd)
f No. of members elected.
g. Meetings per year.
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
2
2
2
2
2
2
2
2
2
2
3
2
2
n
30. Type of communications model
v'
\/
V
y/
Sponsor nondirective.
4
\/
y/
Sponsor directive.
10
31. Study publications
I
10
0
1
11
29
3
2
24
10
31
7
0
2
0
8
0
1
0
29
1
2
6
2
0
0
0
2
51
0
2
0
0
3
0
1
0
8
8
29
3
0
0
55
0
0
0
0
8
0
1
0
10
0
2
4
0
0
24
a. No. of publications.
b. Method of authorship
Papers with corporate
format alone
Papers with conventional
format alone
Papers with both formats
I
Summary tabulations from sketches (continued)
Table B-4
Acronym and sketch no.
PHS
Item
Lu
UGDP VA43
1
2
CDP~ AMIS PARIS HDFP MRF1T CASS posai HPT
MPS NCOS
4
3
7
6
5
8
10
9
11
12
IRSC
13
14
Freq.
32. Author participation in trial
4
None.
v'
Member of CC..........................
Member of trt. monitoring
or advisory review committee
33.
3
y/
Mar
1983
Date sketch completed.
April
1983
May
April
1983
Mar
1983
1983
Mar
1983
April
1983
Mar
1983
May
May
1983
1983
V
V
Mar
1983
May
1983
7
Mar
Mar
1983
1983
I
• *S
• S'
’S'
•
^3=3
^5
—
2
z3 ^3
72
-,2.s =
8 Qte - “
00 c>
?
n
= 3 < « S o---- 2 5- —
2; a o
•
n GO
<
ge: it
b t
r ’nj
“ ° z w ST o’ <
«
S' 3
V)
S S8
S’
S k
CX TO
S' 01
M-
e
ii4
3
o’p o g
2 => 3 ■“» *
3 © - - n
c/>
S
g.q
S'
TO
8 « M
■S
S' o
I
S
2 a.*
© £
S'
i
2
o
5
s.
sv>
3
P rX
3
ex w
q
?'to
g
■
S’. TO
s»
tx —
’
0’3
S
©
= n
o '»
§3
5=8
p 2 « <T>
-
?' o 2
w o.
P' 3
fl
h
g.s
2. ex
"□ «
o
3
3
ex
fl
^g
S.
o
S £
2*
•©
8
rt TO
*
3 er
o’ £.
-1
Vi
>©
8
n
©•
—
S
c
3
3
??
s
2
3
1.
B
Io c
Vi
-1
<aE- i2
2 w
? 8.
w
? 2
o
hs
s a^g. =5
- I ..- qs-:§ a.- sQ
O
TO
£>
3
p
o
S’
■O_
o
n
o’
22 s-« »<
2. F*
& o,
8
?
2JX a
n
— • vi
S’ g c
O.
TO
•3
o
o
§ 2.
ex _
li
N
O 3
-*> fX
o ■o.
C
bi
II
<2 *
lH
2|§:
■o
W CL
—
re
a
«2.^f
9-. §
o’
8
p
3
£ 2. Hi
»<z>
w
=s
n
q.
1.0i’is
Q.
1
3
.O
ft
o’
M TO
r* ft
p —
9LU C
T?
*
c
8
•g
3
s
O
ft
ft
8 3
X
C
s’
8
•8
s
K
5
Q.
o’
O
Q.
<■»
I
I
o
t»
o’
•
•£
-0
. Z a
•g
r
s
1
S’
rt"
3
a
X
g
q
s.
i
TO
n
i
M,
czi
O
iS'
n
A, _
8*§
a
g 2
S’ §. w
S’ S' ?
I'
n"
n
Q.
O
O
o
2
3
q
q G> 3
j? 2 r>2.
t
I’ «
2H.
^1'
£ 3w 83
o “
.
8
'Z)
15
M -1
2 a f
^3’1
i.
2.
n
<z>
5T^
o’
'-0
□
Vi
?8 n
g;
P s
2-ia-
3
o
9
w
o.
M 3
C. TO
H- : H.
O
»g
3q
3
S’
O-
q
O
Q.
.S'
g
s'-
8
a
I’
2.
I.
« 3 ?
_ 2.S h
s s’: ?
M
l»
3
O
2-^ |: §
« 8 q- *< c 8
2
«TO
TO
3
g
3.
q
o.
f
O
a"
*
<«
-Q 5
8_
8
3 s. og 12
8’ « 8
2 nn2
” o TJ
5 3* O
O
_ »
8-1 n
o _
8 “ 2
2 “
3 8
M
S
9s
o- o 3 «
3 52 = ft
03
f 2 O a.
s’
S'
0
i i- 8 n §
rt TO
Sgrt v> o'
«X s
ft
•S'8 I
go I
5-2
« q S on « 3.
„ e- q c «
8
3
;n
s
3 8 ?■ §
2 2
vi
re
•
TO
3
3
I
TO
g
o’
o
aq
“1 % s
3- ex
o
s 5 a..
TO
* S'
C; 2
si3:«
O
Q.
I3
•
T5
a£a^g2P'S|2 y? H -
O
” ° c
M
s B,
3 o
8 «•
S =
S
8
3
s-
8 S’
^■
"• o
2
n
<z> 2
w
=-. K»
3 >
o' O
3 S
ft
o’
c
o *
=
_ 2 3 o o. w
2.
• s'
* 8
n2
3>
2,g
g’o1
3
1
o’
TO S
M
81
2. a
o.
i-s
3.
o
if
-■1
si
£
■g 2
s8.
3
s 3
<»
8
TO.
S’
4
o
3
i
■Q
B’
w
r»
g
a.
o,
3
cq
S'
cn
o'
J
»
S 2 rt00 2
8
a
s?= H
o
m
o-* •< q §
00
n
o
J
a 2.1 a
2
S
K_
3
c
—*> X- 3
Z
£3
3
a
Q
I*
5o
y>
n
N
5
&
B
2
348
• (HPT) Blood pressure change.
• (IRSC) Renal growth and scarring.
hem 26.a
• (MPS) Stated recruitment goals: 522 in SMD argon trial, 736 in HISTO argon trial, and
2I2 in SMD krypton trial. No goal set for 1NVM in the argon trial or for INVM or
HISTO in the krypton trial.
• (IRSC) Goal for reflux grade IV patients. No goal set for grade III.
hem 26. c
• (MPS) The only trial with published results at the time the sketch was completed.
hem 28.d.iv
• (POSCH) The chairman of the study is the only clinic director represented.
hem 29.a
• (MRFIT) Originally the study had separate advisory-review and treatment effects monitor
ing committees. The latter committee was disbanded in I977. Its functions were assumed
by the advisory-review committee at that time.
hem 29. b and 29.c
• (CDP) The treatment effects monitoring committee was headed by co-chairmen. See items
29.b and 29.c for information on the two individuals.
• (PARIS) The treatment effects monitoring committee was headed by co-chairmen. See
items 29.b and 29.c for information on the two individuals.
hem 29.e
• (MPS) The clinic director represented on the committee was also chairman of the study.
• (CDP) The only clinic director represented on the committee was also the vice-chairman of
the study.
• (AMIS) The study chairman and director of the coordinating center attended meetings of
the committee, but were not official members of the committee.
• (MRFIT) The chairman and vice-chairman of the study and the director of the coordinat
ing center were present at meetings of the committee, but were not official members of
the committee.
• (CASS) The project officer was present at meetings of the committee, but was not an
official member of the committee.
r...
1;
Table B-5 Sample sketch for the UGDP 349
B. Sketches of selected trials
Table B-5 Sample sketch for the UGDP
1. General
a. Official name: University Group Diabetes Program
b. Official acronym: UGDP
c. Sketch number: 2
d. Type of trial: Therapeutic
e. Type of design: Fixed sample size design
f. Stage: Completed
2. Funding
Source: Public: National Institute of Arthritis, Metabolism and Digestive Diseases (now the National Institute of
Arthritis, Diabetes, Digestive and Kidney Diseases), Bethesda, Maryland
b. Type: Grant
c. Mode of initiation: Investigator-initiated
d. Mode offund dispersal to centers: Direct
e. Start offunding: September I960
f. End offunding: April 1982
3. Clinics: 12 (including one in Puerto Rico)
4. Resource centers
a. T\>pes of centers represented
• Coordinating center
• Reading center
• Central laboratories (2)
b. Coordinating center
• Study name: Coordinating Center (Baltimore)
• Director: Genell Knatterud, Ph.D. (Christian R. Klimt, M.D., Dr. P.H., through 1977)
• Affiliations with other centers: None
• Address: University of Maryland, School of Medicine, Baltimore, Maryland
c. Reading center
• Study name: ECG Reading Center
• Director: Henry Blackbum, M.D. (Cardiology)
• Affiliations with other centers: Lipid Laboratory and ECG Reading Center both under the direction of Henry
Blackburn. There was a clinic at Minnesota as well, but it was organizationally and administratively independ
ent of these two resource centers.
• Address: University of Minnesota, School of Public Health, Minneapolis, Minnesota
d. Central Laboratory
• Study name: Lipid Laboratory (Minnesota)
• Director: Henry Blackburn, M.D. (Cardiology)
• Affiliations with other centers: Lipid Laboratory and ECG Reading Center both under the direction of Henry
Blackburn. There was a clinic at Minnesota as well, but it was organizationally and administratively independ
ent of the two resource centers.
• Address: University of Minnesota, School of Public Health, Minneapolis, Minnesota
e. Central laboratory
• Study name: Lipid Laboratory (Morgantown)
• Director: Margaret J. Albrink, M.D. (Medicine)
• Affiliations with other centers: None
• Address: West Virginia University Medical Center, Morgantown, West Virginia
5. Project office
• Study name: Liaison Office
• Project Officer: Keatha K. Krueger, Ph.D.
wo*
350
B. Sketches of selected trials
Table B-5 Sample sketch for the UGDP (continued)
• Affiliations with other centers: None
• Address: National Institute of Arthritis, Metabolism, and Digestive Diseases (now the National Institute of
Arthritis, Diabetes, Digestive, and Kidney Diseases), Bethesda, Maryland
6. Treatments
a. Type of treatment design: Noncrossover
b. Type of treatment structure: Simple
c. Test treatments
i. Number: 3
ii. Mode of intervention: Drug
iii. Treatment description: See abstract summary
d. Control treatments
i. Number: 2
ii. Type of treatment administered: Placebo pills or capsules, or standard dose of insulin via injections
iii. Treatment description: See abstract summary
e. Level of masking: Oral hypoglycemic agents administered double-masked. Two insulin treatments administered in
unmasked fashion.
7. Patient characteristics
a. Eligihility criteria: Adult-onset diabetes diagnosed within 12 months prior to enrollment.
b. Demographic characteristics: 20-79 years of age at entry. Mean age: 52.7, 71% female, 54% white.
8. Patient recruitment
■ I.
hr....
a. Mode of initial patient contact: Direct from primary care clinics
b. Start of patient enrollment: February 1961
c. End of patient enrollment: February 1966
d. Total number of patients randomized: 1,027
9. Method of randomization
a. Type: Fixed allocation ratio
b. Stratification variable: Clinic
c. Total number of allocation strata: 12(1 per clinic)
d. Allocation ratio: I:0:l:l:l for TOLB, PHEN, ISTD, IVAR, and PLBO, respectively, for 6 clinics not administering
phenformin and for first 32 patients in the Boston clinic. Ratio of 1:3:1:1:1 was used in the 5 clinics using
phenformin and after enrollment of the 32nd patient in the Boston clinic.
d. Blocking constraints: After every 16th allocation for the 1:0:1:1:1:1 allocation ratio and after every 14th allocation
for the l:3:l:l:l allocation ratio.
f. Locus of control: Central
g. Method of release: Coordinating center, usually via telephone. By letter if time permitted.
19. Data collection schedule
a. Baseline: 2 examinations about 1 month apart
b. Follow-up: Examinations at 3-month intervals after enrollment
c. Post-trial follow-up: Some by individual clinics (see reference citation 2.10, Table B-3, for details)
11. Length of patient follow-up
a. During the trial: Minimum: 9.5 years. Maximum: 14.5 years.
b. Post-trial: 2 years. See comment for item lO.c.
12. Outcome
a. Primary: Death
b. Secondary: Nonfatal vascular complications, especially those affecting the eyes, heart, kidney, or peripheral vascular
system.
13. Treatment effects monitoring
a. Frequency: Twice a year in conjunction with semiannual investigator meetings.
Table B-5 Sample sketch for the UGDP 351
Table B-5 Sample sketch for the UGDP (continued)
b. Approach: Data reports prepared by the coordinating center; reviewed by investigative group.
14. Method of close-out:
Common close-out date. Close-out examinations performed over a 3-month period, with separation completed by
August 1975.
15. Dita entry
a. Site of entry: At the coordinating center
b. Primary mode of entry: Direct from the data forms
16. Sample size specification
a. Original recruitment goal: 200 per treatment group
b. Rationale for the original recruitment goal: Pragmatic
c. Achieved sample size: 200+ per treatment group. Total of 1,027 patients assigned to the 5 treatment groups.
d. Published rationale for achieved sample size: Power argument as stated in reference citation 2.10, Table B-3.
17. Organizational structure
a. Committees
i. Key committees
• Steering Committee
• Executive Committee
• Advisory-Review Committee
ii. Standing subcommittees of the Steering Committee
• Analysis Coordination Committee
• Eye Committee
• Heart Committee
• Kidney Committee
• Medical Technology and Quality Control Committee
• Mortality Committee
• Peripheral Vascular and Neurological Committee
• Statistical Committee
• Clinic Review Committee
• Editorial Review Committee
b. Steering Committee
• Name: Investigative Group
• Chairman: Max Miller, M.D., Case-Western Reserve University, Cleveland, Ohio
• Affiliations with other centers: Director of one of the clinics in the trial.
• Number of members: 26
• Membership representation: 2 voting members from each of the 12 clinics and the coordinating center.
c. Executive Committee
• Name: Executive Committee
• Chairman: Max Miller, M.D., Case-Western Reserve University. Cleveland, Ohio
• Affiliations with other centers: Director of one of the clinics in the trial.
• Number of members: 9
• Membership representation: Study chairman, director of the coordinating center, plus 2 other coordinating center
representatives, plus 3 elected members from the study clinics (3-year terms), the project officer, and the
chairman of the advisory-review committee.
d. Advisory-Review Committee
• Name: Advisory-Review Board (appointed in 1971)
• Chairman: Thomas Chalmers, M.D., Mount Sinai School of Medicine, New York, New York
• Affiliations with other centers: None
• Number of members: 9
• Membership representation: Members appointed by the National Institute of Arthritis, Metabolism, and Diges
tive Diseases without term. Members included study chairman and director of the coordinating center. No other
member had any affiliation with the trial.
1 CTO
V l+ISO
352
B. Sketches of selected trials
Table B-5 Sample sketch for the UGDP (continued)
Table B-6 Data coordinating centers for multicenter trials referenced in this book
Table B-6 Data coordinating centers for multicenter trials referenced in this book
Data coordinating
center director
18. Study publications
a. Number of papers published: 10 (See Table B—3)
b. General method of authorship: Corporate, with writing committee indicated.
I. Anturane Reinfarction Trial (ART)
Sidney H. Kane, M.D.
19. Information sources used for completion of sketch
• Published papers
• UGDP manual of operations
2. Aspirin Myocardial Infarction Study
(AMIS)
William F. Krol, Ph D.
3. Beta Blocker Heart Attack Trial
(BHAT)
C. Morton Hawkins, Sc.D.
4. Coronary Artery Surgery Study (CASS)
Lloyd D. Fisher, Ph D.
5. Coronary Drug Project (CDP)
Paul L. Canner, Ph.D.
6. Diabetes Control and Complications
Trial (DCCT)
John M. Lachin, Sc.D.
7. Diabetic Retinopathy Study (DRS)
Genell L. Knatterud, Ph.D.
8. Early Treatment of Diabetic
Retinopathy Study (ETDRS)
Genell L. Knatterud, Ph.D.
9. Eastern Cooperative Oncology Group
(ECOG)
Marvin Zelen, Ph.D.
10. Hypertension Detection and Follow-Up
Program (HDFP)
C. Morton Hawkins, Sc.D.
II. Hypertension Prevention Trial (HPT)
Curtis L. Meinert, Ph.D.
12. International Mexilitene Placebo
Antiarrhythmic Coronary Trial
(IMPACT)
Jean-Pierre Boissel, M.D.
20. Author’s involvement in trial
• Deputy director of coordinating center from start of study to mid-1979
21. Person reviewing sketch
• Name: Genell L. Knatterud, Ph.D.
• Position in study: Director of Coordinating Center
Study name and acronym
22. Date sketch completed
April 12, 1983
h«...
!.
I
I Kt
i? i
Christian R. Klimt, M.D.
Address
Ciba-Geigy Corporation
Pharmaceutical Division
Summit, New Jersey 07901
Maryland Medical Research
Institute
600 Wyndhurst Avenue
Baltimore, Maryland 21210
School of Public Health
University of Texas
1100 Holcombe Blvd.
Houston, Texas 77025
School of Public Health
University of Washington
1107 NE 45th Street
Seattle, Washington 98105
School of Medicine
University of Maryland
600 Wyndhurst Avenue
Baltimore, Maryland 21210
Department of Statistics
The George Washington
University
7979 Old Georgetown Road
Bethesda, Maryland 20814
School of Medicine
University of Maryland
600 Wyndhurst Avenue
Baltimore, Maryland 21210
School of Medicine
University of Maryland
600 Wyndhurst Avenue
Baltimore, Maryland 21210
School of Public Health
Harvard University
44 Binney Street
Boston, Massachusetts 02115
School of Public Health
University of Texas
1100 Holcombe Blvd.
Houston, Texas 77025
School of Hygiene and
Public Health
The Johns Hopkins
University
615 North Wolfe Street
Baltimore, Maryland 21205
Dept. Unite de
Pharmacologic Clinique
Hopital Cardiologique
B.P. Lyon Montchat
69394 Lyon Cedex 3
France
Maryland Medical Research
Institute
600 Wyndhurst Avenue
Baltimore, Maryland 21210
353
&
354
B. Sketches of selected trials
Table B-6 Data coordinating centers for multicenter trials referenced in this book (continued)
Study name and acronym
13. International Reflux Study in Children
(IRSC)
Data coordinating
center director
Tytti Tamminen, M.D.
Robert Weiss, M.D.
I
I r rn
I
L
I I..-'!
14. Lipid Research Clinics Coronary
Primary Prevention Trial (LRCCPPT)
O. Dale Williams, Ph.D.
15. Macular Photocoagulation Study
(MPS)
Barbara S. Hawkins, M.Sc.
16. Multicenter Investigation for Limiting
Infarction Size (MILIS)
W. Kenneth Poole, Ph.D.
17. Multiple Risk Factor Intervention Trial
(MRFIT)
Marcus O. Kjelsberg, Ph.D.
18. National Cooperative Dialysis Study
(NCOS)
Edmund G. Lowrie, M.D.
19. National Cooperative Gallstone Study
(NCOS)
John M. Lachin, Sc.D.
20. Persantine Aspirin Reinfarction Study
(PARIS)
Christian R. Klimt, M.D.,
Dr. PH.
21. Physicians’ Health Study (PHS)
Charles Hennekins, M.D.
22. Program on the Surgical Control of
Hyperlipidemia (POSCH)
John M. Long, Ed.D.
23. University Group Diabetes Program
(UGDP)
Genell L. Knatterud, Ph.D.
24. Veterans Administration Cooperative
Studies Program No. 43 (VA 43)
Stephen F. Bingham, Ph.D.
C. Year 1980 clinical trial publications’
Address
University Children’s
Hospital
Hufelandstrabe 55
Essen, West Germany
D-4300
Albert Einstein
College of Medicine
1825 Eastchester Road
New York, New York 10461
School of Public Health
University of North Carolina
Chapel Hill, North
Carolina 27514
The Wilmer Ophthalmology
Institute
The Johns Hopkins
University
550 North Broadway
Baltimore, Maryland 21205
Research Triangle Institute
Research Triangle Park,
North Carolina 27709
School of Public Health
University of Minnesota
Minneapolis,
Minnesota 55455
School of Public Health
Harvard University
721 Huntington Avenue
Boston, Massachusetts 02115
Department of Statistics
The George Washington
University
7979 Old Georgetown Road
Bethesda, Maryland 20814
Maryland Medical Research
Institute
600 Wynd hurst Avenue
Baltimore, Md 21210
Department of Medicine
Harvard Medical School
55 Pond Avenue
Boston, Massachusetts 02146
School of Medicine
University of Minnesota
Minneapolis,
Minnesota 55455
School of Medicine
University of Maryland
600 Wyndhurst Avenue
Baltimore, Maryland 21210
Veterans Administration
Medical Center
Perry Point, Maryland 20801
C.l
PAPERS REVIEWED (113)
I. Aktulga E, Altac M, Muftuoglu A, Ozyazgan Y, Pazarli H, Tuzun Y, Yalcin B, Yazici H, Yurdakul S: A
double blind study of colchicine in Behcet’s disease. Haematologica 65:399-402, 1980.
2. Alacron-Segovia D: Long-term treatment of symptomatic osteoarthritis with benoxaprofen. Double-blind
comparison with aspirin and ibuprofen. 7 Rheumatol 7(suppl 6):89-99, 1980.
3. Antarkar DS, Vaidya AB, Doshi JC, Athavale AV, Vinchoo KS, Natekar MR, Tathed PS, Ramesh V,
Kale N: A double-blind clinical trial of Arogya-wardhani—an Ayurvedic drug—in acute viral hepatitis
Indian J Med Res 72:588-593, 1980.
4. Australian Therapeutic Trial in Mild Hypertension Management Committee: Australian Therapeutic Trial
in Mild Hypertension. Lancet 1:1261-1267, 1980.
5. Bain J, Rachlis V, Robert E, Khait Z: The combined use of oral medroxyprogesterone acetate and
methyltestosterone in a male contraceptive trial programme. Contraception 21:365-379, 1980.
6. Banham SW, Moran F: A clinical trial of oxatomide in asthma. Br J Clin Pract 34:323-326, 1980.
7. Barlow JJ, Piver MS, Lele SB: High-dose methotrexate with “rescue” plus cyclophosphamide as initial
chemotherapy in ovarian adenocarcinoma. A randomized trial with observations on the influence of C
parvum immunotherapy. Cancer 46:1333-1338, 1980.
8. Blechman WJ: Crossover comparison of benoxaprofen and naproxen in osteoarthritis. 7 Rheumatol
7(suppl 6):116-124, 1980.
9. Bone GE, Pomajzl MJ: Prospective comparison of polytetrafluorethylene and bovine grafts for dialysis J
Surg Res 29:223-227, 1980.
10. Brewer C: Prevention of infection after abortion with a supervised single dose of oral doxycycline. Br Med
7 281:780-781, 1980.
11. Brooks PM, Hill W, Geddes R: Diclofenac and ibuprofen in rheumatoid arthritis and osteoarthritis. Med
J A ust 1:29-30, 1980.
12. Cardozo LD, Stanton SL, Robinson H, Hole D: Evaluation of flurbiprofen in detrusor instability Br Med
7 280:281-282, 1980.
13. Carmichael J, Bloomfield P, Crompton GK: Comparison of fenoterol and terbutaline administered by
intermittent positive pressure breathing. Br J Dis Chest 74:268-272, 1980.
14. Cohen MM, Cheung G, Lyster DM: Prevention of aspirin-induced fecal blood loss in men with prostaglandin E2. Adv Prostaglandin Thromboxane Res 8:1525-1528, 1980.
15. Collum LM, Benedict-Smith A, Hillary IB: Randomised double-blind trial of acyclovir and idoxuridine in
dendritic corneal ulceration. Br J Ophthalmol 64:766-769, 1980.
16. Coope J: Blood pressure in the elderly. 7 R Coll Gen Pract (Occas Pap) 12:35-37, 1980.
17. Cooper SA, Precheur H, Rauch D, Rosenheck A, Ladov M, Engel J: Evaluation of oxycodone and
acetaminophen in treatment of postoperative dental pain. Oral Surg 50:496-501, 1980.
18. Cooper SA, Reynolds DC, Kruger GO, Gottlieb S: An analgesic relative potency assay comparing
zomepirac sodium and aspirin. 7 Clin Pharmacol 20:98-106, 1980.
19. Cross FS, Long MW, Banner AS, Snider DE, Jr: Rifampin-isoniazid therapy of alcoholic and nonalco
holic tuberculosis patients in a U.S. Public Health Service cooperative therapy trial. Am Rev Respir Dis
122:349-353, 1980.
1. Reprinted with permission of Elsevier Science Publishing Co.. Inc., New York (from reference 321 in Appendix 1). See Chapter 2.
J
355
356
C. Year 1980 clinical trial publications
20. D’Angelo U, Sokol RJ: Short- versus long-course prophylactic antibiotic treatment in cesarean section
I > e ii....
r.
..
patients. Obstet Gynecol 55:583-586, 1980.
21. Davidson ED. Hersh T, Perkel MS, Moore C, Fajman WA: The effects of coherin on patients with
idiopathic delayed gastric emptying. Am J Gastroenterology 74:419-422, 1980.
22. Davies J, Dixon AS, Steele CE: Tolmetin sodium and indomethacin in the treatment of osteoarthrosis of
the hip: A double-blind crossover study. Curr Med Res Opin 7:115-120, 1980.
23. DeAndrade JR, Honig S, Ciccone WJ, Leffall L: Clinical comparison of zomepirac with pentazocine in
the treatment of postoperative pain. J Clin Pharmacol 20:292-297, 1980.
24. Depew W, Boyer T, Omata M, Redeker A, Reynolds T: Double-blind controlled trial of prednisolone
Gastroenter-
therapy in patients with severe acute alcoholic hepatitis and spontaneous encephalopathy. Gastroenter
ology 78:524-529, 1980.
25. Dyson AJ, Mackay AD: Two oral beta-adrenergic stimulant drugs, pirbuterol and salbutamol, in reversi
ble airway obstruction. Br J Dis Chest 74:70-74, 1980.
26. Fairfax AJ, McNabb WR, Davies HJ, Spiro SG: Slow-release oral salbutamol and aminophylline in
nocturnal asthma: Relation of overnight changes in lung function and plasma drug levels. Thorax
35:526-530, 1980.
27. Fiasse R, Hanin C, Lepot A, Descamps C, Lamy F, Dive C: Controlled trial of cimetidine in reflux
esophagitis. Dig Dis Sci 25:750-755, 1980.
28. Forbes J A, White RW, White EH, Hughes MK. An evaluation of the analgesic efficacy of proquazone and
aspirin in postoperative dental pain. J Clin Pharmacol 7:465-474, 1980.
29. Foster CS, Duncan J: Randomized clinical trial of topically administered cromolyn sodium for vernal
keratoconjunctivitis. Am J Ophthalmol 90:175-181, 1980.
30. Gairola RL, Gupta PK, Pandley K: Antagonists of morphine-induced respiratory depression: A study in
postoperative patients. Anaesthesia 35:17-21, 1980.
31. Groggins RC, Lenney W, Milner AD, Stokes GM: Efficacy of orally administered salbutamol and
theophylline in pre-school children with asthma. Arch Dis Child 55:204-206, 1980.
32. Guinan ME, MacCalman J, Kern ER, Overall JC, Jr, Spruance SL: Topical ether and herpes simplex
labialis. JAMA 243:1059-1061, 1980.
33. Henkin RE, Woodruff A, Chang W, Green AM: The effect of radiopharmaceutical incubation time on
bone scan quality. Radiology 135:463-466, 1980.
34. Hill JF: Clinical comparison of the (polymacon) spin-cast hydrogel contact lens to the (polymacon) lathe
cut hydrogel lenses. Am J Optom Physiol Opt 57:523-527, 1980.
35. Hillas JL, Somerfield SD, Wilson JD, Aman MG: Azatadine maleate in perennial allergic rhinitis: Effects
on clinical symptoms and choice reaction time. Br J Clin Pharmacol 10:573-577, 1980.
36. Hillson RM, Boyd E, Cunningham J: Prophylactic disopyramide: Its clinical effects related to plasma
concentration in myocardial infarction. J Int Med Res 8:314-320, 1980.
37. Hortobagyi GN, Yap HY, Wiseman CL, Blumenschein GR, Buzdar AU, Legha SS, Gutterman JU, Hersh
EM, Bodey GP: Chemoimmunotherapy for metastatic breast cancer with 5-fluorouracil, adriamycin,
cyclophosphamide, methotrexate, L-asparaginase, corynebacterium parvum and pseudomonas vaccine.
Cancer Treat Rep 64:157-159, 1980.
38. Hubay CA, Pearson OH, Marshall JS, Rhodes RS, DeBanne SM, Rosenblatt J, Mansour EG, Hermann
RE Jones JC, Flynn WJ, Eckert C, McGuire WL: Adjuvant chemotherapy, antiestrogen therapy and
immunotherapy for stage II breast cancer: 45-month follow-up of a prospective, randomized clinical
trial. Cancer 46(suppl 12):2805—2808, 1980.
39. Iles JD: Relief of postoperative pain by ibuprofen: A report of two studies. Can J Surg 23:288-290, 1980.
40. Jacobi GH, Altwein JE, Kurth KH, Basting R, Hohenfellner R: Treatment of advanced prostatic cancer
u>i*k
with narAntArai
parenteral rvnmtprnnp
cyproterone acetate: A nhase
phase III randomized trial. Br J Urol 52:208-215,
52:208 215, 1980.
41. Joos C, Kewitz H, Reinhold-Kourniati D: Effects of diuretics on plasma lipoproteins in healthy men. Eur
J Clin Pharmacol 17:251-257, 1980.
42. Kayasseh L, Gyr K, Keller U, Stalder GA, Wall M: Somotostatin and cimetidine in peptic-ulcer hemor
rhage: A randomised controlled trial. Lancet 1:844-846, 1980.
43. Khan AKA, Akhtar S, Mahtab H: Treatment of diabetes mellitus with Coccinia indica. Br Med J 28:1044,
1980.
C.l Papers Reviewed 357
44. Kim KK, Sirman A, Trainor FS, Lee BY: Anxiolytic efficacy and safety of ketazolam compared with
diazepam and placebo. Clin Ther 3:9-14, 1980.
45. Kobayashi K, Nakaoka K, Tsuji H, Shohmori T: Effects of thyrotropin-releasing hormone in chronic
schizophrenic patients. Acta Med Okayama 34:263-273, 1980.
46. Koyama H, Wada T, Takahashi Y, Nishizawa Y, Iwanaga T, Aoki Y, Terasawa T, Kosaki G, Kajita A,
Wada A: Surgical adjuvant chemotherapy with mitomycin C and cyclophosphamide in Japanese
patients with breast cancer. Cancer 46:2373-2379, 1980.
47. Lambert WG, Mullinger BM: Single-dose cefuroxime in the prophylaxis of abdominal wound sepsis. Curr
Med Res Opin 6:404-406, 1980.
48. Lassus A: Systemic treatment of psoriasis with an oral retinoic acid derivative (Ro 10-9359). Br J Dermatol
102:195-202, 1980.
49. Lawrence CM, Millac P, Stout GS, Ward JW: The use of choline chloride in ataxic disorders. J Neurol
Neurosurg Psychiatry 43:452-454, 1980.
50. Levitt NS, Vinik Al, Sive AA, Laff U, Phillips C: Synthetic luteinizing hormone-releasing hormone in
impotent male diabetics: A double-blind cross-over trial. S Afr Med J 57:701-704, 1980.
51. Lutterodt A, Nattel S, McLeod PJ: Duration of antihypertensive effect of a single daily dose of hydrochlo
rothiazide. Clin Pharmacol Ther 27:324-327, 1980.
52. MacGregor AJ, Addy A: Value of penicillin in the prevention of pain, swelling and trismus following the
removal of ectopic mandibular third molars. Int J Oral Surg 9:166-172, 1980.
53. Mackay AD: Amoxycillin versus ampicillin in treatment of exacerbations of chronic bronchitis. Br J Dis
Chest 74:379-384, 1980.
54. Manyam NV, Hare TA, Katz L: Effect of isoniazid on cerebrospinal fluid and plasma gaba levels in
Huntington’s disease. Life Sci 26:1303-1308, 1980.
55. Marcial VA, Hanley JA, Chang C, Davis LW, Moscol JA: Split-course radiation therapy of carcinoma of
the nasopharnyx: Results of a national collaborative clinical trial of the Radiation Therapy Oncology
Group. Int J Radial Oncol Biol Phys 6:409-414, 1980.
56. Marks IN, Wright JP, Denyer M, Garisch JA, Lucke W: Comparison of sucralfate with cimetidine in the
short-term treatment of chronic peptic ulcers. S Afr Med J 57:567-573, 1980.
57 Mendlewicz J, Linkowski P, Rees J A: A double-blind comparison of dothiepin and amitriPly**n® I"
patients with primary affective disorder: Serum levels and clinical response. Br J Psychiatry 136:154160, 1980.
58. Merkatz 1R, Peter JB, Barden TP: Ritodrine hydrochloride: A betamimetic agent for use in preterm labor.
II. Evidence of efficacy. Obstet Gynecol 56:7-12, 1980.
59. Meyhoff HH, Hess J, Olesen KP: Pulmonary atelectasis following upper urinary tract surgery on patients
in the 25° and 45° “jack-knife” position: A sequential analysis. Scand J Urol Nephrol 14:107-109, 1980.
60. Milman N, Scheibel J, Jessen O: Lysine prophylaxis in recurrent herpes simplex labialis: A double-blind,
controlled crossover study. Acta Derm Venereol 60:85-87, 1980.
61. Mishra PC, Agarwal VK, Rahman H: Therapeutic trial of amitryptiline in the treatment of nocturnal
enuresis—a controlled study. Indian Pediatr 17:279-285, 1980.
62. Moertel CG, Hanley JA, Johnson LA: Streptozocin alone compared with streptozocin plus fluorouracil in
the treatment of advanced islet-cell carcinoma. N Engl J Med 303:1189-1194, 1980.
63. Mogg GA, Arabi Y, Youngs D, Johnson M, Bentley S, Burdon DW, Keighley MR: Therapeutic trials of
antibiotic associated colitis. Scand J Infec Dis Suppl 22:41-45, 1980.
64. Moncloa F, Hwang IK, Muccilli AC: Multiclinic evaluation of the antihypertensive effect of a methyl
dopa, hydrochlorothiazine, and amiloride combination. Clin Ther 3:168-175, 1980.
65. Morisky DE, Levine DM, Green LW, Russell RP, Smith C, Benson P, Finlay J: The relative impact of
health education for low- and high-risk patients with hypertension. Prev Med 9:550-558, 1980.
66. Muller-Lissner SA, Sonnenberg A, Eichenberger P, Blum AL: Ranitidine inhibits gastric acid and pepsin
secretion following sham feeding. Hepatogastroenterology 27:377-380, 1980.
1?OA
67. Munzenberg J, Tachibana S: Preliminary double-blind evaluation of a new, non-steroidal anti-inflammatory drug: proctacine. Pharmatherapeutica 2:279-284, 1980.
68. Mussche RA, Kluyskens P: Prognosis of primarily treated localized laryngeal carcinoma ameliorated
through levamisole treatment: A randomized pilot study. Oncology 37:329-335, 1980.
358
| I r' n 4 n <
; J?
C.2 Papers Excluded
C. Year 1980 clinical trial publications
69. Nanda RN, Arthur GP, Johnson RH, Lambie DG: Cimetidine in the prophylaxis of migraine. Acta
Neurol Scand 62:90-95, 1980.
70. Norberg A, Norberg B, Parkhede U, Gippert H, Lundbeck K: Randomized double-blind study of
prophylactic methenamine hippurate treatment of patients with indwelling catheters. Eur J Clin Phar
macol 18:497-500, 1980.
71. North JB, Penhall RK, Hanieh A, Hann CS, Challen RG, Frewin DB: Postoperative epilepsy: A double
blind trial of phenytoin after craniotomy. Lancet 1:384-386, 1980.
72. Paavonen J, Kousa M, Saikku P, Vartiainen E, Kanerva L, Lassus A: Treatment of nongonococcal
urethritis with trimethoprim-sulphadiazine and with placebo. A double-blind partner-controlled study.
Br J Vener Dis 56:101-104, 1980.
73. Pack AR, Thomson ME: Effects of topical and systemic folic acid supplementation on gingivitis in
pregnancy. J Clin Periodontol 7:402-414, 1980.
74. Pegram V, Hyde P, Linton P: Chronic use of triazolam: The effects on the sleep patterns of insomniacs. J
Int Med Res 8:224-231, 1980.
75. Petrie WM, McEvoy JP, Wilson WH, Ban TA, Guy W: Viloxazine in the treatment of depressive neurosis:
A placebo and standard (imipramine) controlled clinical study. Int Pharmacopsychiatry 15:193-196,
1980.
76. Pettersson RF, Hellstrom PE, Penttinen K, Pyhala R, Tokola O, Vartio T, Visakorpi R: Evaluation of
amantadine in the prophylaxis of influenza A(H1N1) virus infection: A controlled field trial among
young adults and high-risk patients. J Infect Dis 142:377-383, 1980.
77. Prins D, Mandelkom T, Cerf FA: Principal and differential effects of haloperidol and placebo treatments
upon speech disfluencies in stutterers. J Speech Hear Res 23:614-629, 1980.
78. Reiling RB, Reiling WA Jr., Bernie WA, Huffer AB, Perkins NC, Elliott DW: Prospective controlled study
of gastrointestinal stapled anastomoses. Am J Surg 139:147-152, 1980.
79. Ritchie JA, Truelove SC: Comparison of various treatments for irritable bowel syndrome. Br Med J
281:1317-1319, 1980.
80. Rivkin L, Rapaport M: Clinical evaluation of a new erythromycin solution for acne vulgaris. Cutis 25:552555, 1980.
81. Rofman BA, Kulaga SF, Gabriel MA, Thiyagarajan B, Nancarrow JF, Abrams WB: Multiclinic evalua
tion of timolol in the treatment of mild-to-moderate essential hypertension. Hypertension 2:643-648,
1980.
82. Rosenthal AL: Clocortolone pivalate: A paired comparison clinical trial of a new topical steroid in
eczema/atopic dermatitis. Cutis 25:96-98, 1980.
83. Rune SJ, Zachariassen A: Acute relief of epigastric pain by antacid in duodenal ulcer patients. Scand J
Gastroenterol Suppl 15:41-45, 1980.
84. Ruoff GE, Andelman SY, Cannella JJ: Long-term safety zomepirac: A double-blind comparison with
aspirin in patients with osteoarthritis. J Clin Pharmacol 20:377-384, 1980.
85. Santini A: The clinical assessment of the pulpotomy technique on teeth of various post-eruptive age
groups: A four year assessment using standardized clinical methods (11). Quintessence Int 12:77—80,
1980.
86. Schroeder JS, Rosenthal S, Ginsburg R, Lamb I: Medical therapy of Prinzmetal’s variant angina. Chest
78:231-233, 1980.
87. Scott DH, Arthur GR, Scott DB: Haemodynamic changes following buprenorphine and morphine.
Anaesthesia 35(suppl 1): 957-961, 1980.
88. Seedat YK: Trial of atenolol and chlorthalidone for hypertension in black South Africans. Br Med J
281:1241-1243, 1980.
89. Simpson RJ, Tiplady B, Skegg DC: Event recording in a clinical trial of a new medicine. Br Med J
280:1133-1134,1980.
90. Sleijfer DT, Mulder NH, DeVries-Hospers HG, Fidler V, Nieweg HO, Van Der Waaij D, Van Saene HK:
Infection prevention in granulocytopenic patients by selective decontamination of the digestive tract.
Eur J Cancer 16:859-869, 1980.
91. Smith JA, Skidmore AG, Forward AD, Clarke AM, Sutherland E: Prospective randomized, double-blind
comparison of metronidazole and tobramycin with clindamycin and tobramycin in the treatment of
intra-abdominal sepsis. Ann Surg 192:213-220, 1980.
359
92. Souka AR. Osman M, Sibaie F, Einen MA: Therapeutic value of indomethacin in threatened abortion.
Prostaglandins 19:457-460, 1980.
93. Stuart RK, Braine HG, Lietman PS, Saral R, Fuller DJ: Carbenicillin-trimethoprim/sulfamethoxazole
versus carbenicillin-gentamicin as empiric therapy of infection in granulocytopenic patients: A prospec
tive, randomized, double-blind study. Am J Med 68:876-885, 1980.
94. Stunkard AJ, Craighead LW, O’Brien R: Controlled trial of behaviour therapy, pharmacotherapy, and
their combination in the treatment of obesity. Lancet 2:1045-1047. 1980.
95. Svennevig JL, Bugge-Asperheim B, Bjorgo S, Kleppe H, Birkeland S: Methyl-prednisolone in the
treatment of lung contusion following blunt chest trauma. Scan J Thorac Cardiovasc Surg 14:301-305,
1980.
96. Svensson G, Hegardt B, Lofkvist T: Effects of topical use of beta-adrenoceptor stimulants on nasal
mucosa: Rhinomanometric evaluations in experiments with terbutaline and KWD2131. Acta Otolaryn
gol 90:297-303, 1980.
97. Szmuness W, Stevens CE, Harley EJ, Zang EA, Oleszko WR, William DC, Sadovsky R, Morrison JM,
Kellner A: Hepatitis B vaccine: Demonstration of efficacy in a controlled clinical trial in a high-risk
population in the United States. N Engl J Med 303:833-841, 1980.
98. Theroux P, Waters DD, Debaisieux JC, Szlachcic J, Mizgala HF, Bourassa MG: Hemodynamic effects of
calcium ion antagonists after acute myocardial infarction. Clin Invest Med 3:81-85, 1980.
99. Turkington RW: Depression masquerading as diabetic neuropathy. JAMA 243:1147-1150, 1980.
100. Vedin A, Wikstrand J, Wilhelmsson C, Wallentin 1: Left ventricular function and beta-blockade in chronic
ischaemic heart failure: Double-blind cross-over study of propranolol and penbutolol using noninvasive techniques. Br Heart J 44:101 -107, 1980.
101. Velasco M, Guevara J, Morillo J, Ramirez A, Urbina-Quintana A, Hemandez-Pieretti O: Antihyperten
sive effect of atenolol alone or combined with chlorthalidone in patients with essential hypertension. Br
J Clin Pharmacol 9:499-504, 1980.
102. Von Knorring L: A double-blind trial: Vivalan against placebo in depressed elderly patients. J Int Med Res
8:18-21, 1980.
103. Walinder J, Carlsson A, Persson R, Wallin L: Potentiation of the effect of antidepressant drugs by
tryptophan. Acta Psychiatr Scand Suppl 61:243-249, 1980.
104. Washton AM, Resnick RB: Clonidine versus methadone for opiate detoxification: Double-blind out
patient trials. Natl Inst Drug Abuse Res Monogr Ser 34:89-94, 1980.
105. Weibel RE, Villarejos VM, Klein EB, Buynak EB, McLean AA, Hilleman MR: Clinical and laboratory
studies of live attenuated RA 27/3 and HPV77-DE rubella virus vaccines (40931). Proc Soc Exp Biol
Med \bSM-A9, 1980.
106. White PF, Ham J, Way WL, Trevor AJ: Pharmacology of ketamine isomers in surgical patients.
Anesthesiology 52:231-239, 1980.
107. Whittington J, Raftery EB: A controlled comparison of oxyfedrine, isosorbide dinitrate and placebo in the
treatment of patients suffering attacks of angina pectoris. Br J Clin Pharmacol 10:211-215, 1980.
108. Williams KA, Ting A, French ME, Oliver D: Peroperative blood transfusions improve cadaveric renalallograft survival in non-transfused recipients. Lancet 1:1104-1106, 1980.
109. Wilson EA: Diazepam and local anaesthetic spray versus general anaesthesia for gastroduodenoscopy.
S Afr Med 7 57:111-113, 1980.
110. Wilson WR, Byl FM, Laird N: The efficacy of steroids in the treatment of idiopathic sudden hearing loss:
A double-blind clinical study. Arch Otolaryngol 106:772-776, 1980.
111. Wordsworth BP, Ebringer RW, Coggins E, Smith S: A double-blind crossover trial of fenoprofen and
phenylbutazone in ankylosing spondylitis. Rheumatol Rehabil 19:260-263, 1980.
112. Zykov MP, Rudenko LG, Zoshchenkova NY, Rafael’SkayaTI: Subunit influenza vaccine and its testing in
clinical immunology. J Hyg Epidemiol Microbiol Immunol 24:212-218, 1980.
113. (No author listed): The treatment of dyspepsia in general practice: A multicentre trial. Practitioner
224:105-107, 1980.
C.2 PAPERS EXCLUDED (67)
114. Adams FG, Horton PW. Selim SM: Clinical comparison of three liver scanning agents. Eur J Nucl Med
5:237-239, 1980.
360
C. Year 1980 clinical trial publications
115. Andreasen G: A clinical trial of alignment of teeth using a 0.019 inch thermal nitinol wire with a transition
temperature range between 31° C. and 45° C. Am J Orthod 78:528-537, 1980.
116. Anturan: New use for an old drug. Can Med Assoc J 120:714-716, 1980.
117. Baker HW, Pepperell RJ: Lack of effect of bromocriptine on semen quality in men with normal or slightly
elevated prolactin levels. Aust NZ J Obstet Gynaecol 20:158-161, 1980.
118. Banta HD. Thacker SB: More fetal monitoring debate (letter). Pediatrics 65:366-368, 1980.
119. Bedikian AY, Valdivieso M, Heilbrun LK, Withers RH. Bodey GP, Freireich EJ: Glycerol: An alternative
to dexamethasone for patients receiving brain irradiation for metastatic disease. South Med J 73:12101214, 1980.
120. Beutler E, Dale GL, Kuhl W: Replacement therapy in Gaucher disease. Birth Defects 16:369-381, 1980.
121. Richer HI, Sandhu TS, Hetzel FW: Hyperthermia and radiation in combination: A clinical fractionation
regime. Int J Radial Oncol Biol Phys 6:867-870, 1980.
122. Bird HA, Wright V, Galloway D: Clinical metrology—a future career grade? Lancet 2:138-140, 1980.
123. Borden EC, Hawkins MJ: Interferons for human neoplastic and viral diseases. Compr Ther 6:6-15, 1980.
124. Bruni J, Wilder BJ, Bauman AW, Willmore LJ: Clinical efficacy and long-term effects of valproic acid
therapy on spike-and-wave discharges. Neurology 30:42-46, 1980.
125. Caine RY: Immunosuppression for organ grafting. Transplant Proc 12:239-243, 1980.
I
nU
I
126. Cantell K: Interferon: The state of the art. Triangle 19:47-51, 1980.
127. Carter SK: Clinical chemotherapy: Its correlation with experimental models. Recent Results Cancer Res
75:31-36,1980.
128. Chastang C, Auquier A, Gremy F: Prognostic knowledge: Interest and methods. Bull Cancer 67:430-436,
1980.
129. Chew CY, Brown BG. Singh BN, Hecht HS, Schnugg SJ, Wong M, Shah PM, Dodge HT: Mechanism of
action of verapamil in ischemic heart disease: Observations on changes in systemic and coronary
hemodynamics and coronary vasomobility. Clin Invest Med 3:151-158, 1980.
130. Conti CR: Management of stable and unstable angina with observations on lessons learned from the
prospective controlled studies. Adv Cardiol 27:191-198. 1980.
131. Crowe JP, Phelan JJ, Cleere WF, Fottrell PF, McNicholl B, McCarthy CF: The lymphocyte transforma
tion test in coeliac disease: Effect of gliadin and detoxified gliadin. Digestion 20:95-99, 1980.
132. Cunliffe WJ: Techniques in the investigation of acne. Acta Derm Venereal Suppl 89:39 46, 1980.
133. Davis TG, Pickett DL, Schlosser JH: Evaluation of a worldwide spontaneous reporting system with
cimetidine. JAMA 243:1912-1914, 1980.
134. Donald JF. Rimmer DM: An open evaluation of a 3-day course of pivmecillinam (ten 200 mg tablets) in
women with acute uncomplicated cystitis. J /nt Med Res 8:112-117, 1980.
135. Emanueli A, Sacchetti G: An algorithm for the classification of untoward events in large scale clinical
trials. Agents Actions Suppl 7:318-322, 1980.
136. Epstein M: Sounding boards. The LeVeen shunt for ascites and hepatorenal syndrome. N Engl J Med
302:628-630, 1980.
137. Fink M, Irwin P: EEG and behavioral profile of flutroline (CP-36, 584): A novel antipsychotic drug.
Psychopharmacology 72:67-71, 1980.
138. Freeman H, Soni SD: Oxypertine for tardive dyskinesia (letter). Br J Psychiatry 136:522-523, 1980.
139. Friedman LM: Summary of design features: Clinical trials of platelet-active drugs in cerebrovascular
disease. Circulation 62 (suppl V): V-88—V-89, 1980.
140. Furberg CD: Principles of clinical trials: U.S. viewpoint. Triangle 19:99-102, 1980.
141. George SL: Sequential clinical trials in cancer research. Cancer Treat Rep 64:393-397, 1980.
142. Gonzalez ER: Study to begin on whether aspirin can delay senile cataract formation. JAMA 244:25932594, 1980.
143. Gross F: Medical, ethical and legal aspects of clinical trials in pediatrics. Summary of a forum discussion
held at the International Workshop on Perinatal and Pediatric Aspects of Clinical Pharmacology.
Heidelberg, Federal Republic of Germany, February 27-29, 1980. Eur J Clin Pharmacol 18:121-127,
1980.
144. Hansen M, Hansen HH, Dombernowsky P: Long-term survival in small cell carcinoma of the lung. JAMA
244:247-250, 1980.
C.2 Papers Excluded
361
145. Herting RL, Lane AZ, Lorber RR, Wright JJ: Netilmicin: Chemical development and overview of clinical
research. Scand J Infec Dis Suppl 23:20-29, 1980.
146. Hildebrand H, Berg NO, Hoevels J, Ursing B: Treatment of Crohn’s Disease with metronidazole in
childhood and adolescence: Evaluation of a six months trial. Gastroenterol Clin Biol 4:19-25, 1980.
147. Holt JA: Microwave adjuvant to radiotherapy and chemotherapy for advanced lymphoma (letter). Med J
Aust 2:159-160, 1980.
148. Jerie P: Clinical experience with guanfacine in long-term treatment of hypertension. Br J Clin Pharmacol
Suppl IO:37S^7S, 1980.
149. Lazzara R, Scherlag B: Treatment of arrhythmias by blocking slow current. Ann Inter Med 93:919-921,
1980.
150. Lewison EF: Changing concepts in breast cancer. Cancer 46(suppl 4):859-864, 1980.
151. Lorenz W, Fischer M, Rohde H, Triodl H, Reimann HJ, Ohmann C: Histamine and stress ulcer: New
components in organizing a sequential trial on cimetidine prophylaxis in seriously ill patients and
definition of a special group at risk (severe polytrauma). Klin Wochenschr 58:653-665, 1980.
152. Marcus P. Clinical trial or promotional exercise? (letter). Lancet 1:827, 1980.
153. McConkey B, Amos RS, Billingham ME, Constable TJ, Crockson RA, Crockson AP, Forster PJ:
Rheumatoid arthritis: Effects of a new agent (ICI 55 897) on serum acute phase proteins and the
erythrocyte sedimentation rate. Ann Rheum Dis 39:18-21, 1980.
154. Measham AR, Khan AR, Huber DH: Dizziness associated with discontinuation of oral contraceptives in
Bangladesh, /nt J Gynaecol Obstet 18:109-112, 1980.
155. Menken MM: Evaluation of surgical therapy (letter). N Engl J Med 303: 399, 1980.
156. Meyer BH. Hugo JM: Etomidate as anaesthetic induction agent in open-heart surgery. S Afr Med J
58:759-761, 1980.
157. Muser RK: Design of clinical trials for new drugs in animals. J Am Vet Med Assoc 176:1148-1150, 1980.
158. Norton A: Psychosurgery—why not ban it? J R Soc Med 73:526-528, 1980.
159. Oswald I: Sleep studies in clinical pharmacology. Br J Clin Pharmacol 10:317-326, 1980.
160. Owen M, Hills U: How safe is dextropropoxyphene? (letter). Med J Aust 1:617-618, 1980.
161. Pavelka K, Vojtisek O, Kankova D: The occurrence of adverse side-effects during controlled clinical trials
of nonsteroidal antirheumatic agents. Rheumatol Rehabil 19:109-112, 1980.
162. Pethybridge RJ: Data Analysis: 1. Some statistical tests for small related samples. J R Nav Med Serv
66:135-139, 1980.
163. Piver MS, Barlow J J, Dunbar J: Doxorubicin, cyclophosphamide, and 5-fluorouracil in patients with
carcinoma of the cervix or vagina. Cancer Treat Rep 64:549-551, 1980.
164. Poster DS, Penta J, Marsoni S, Bruno S, Macdonald JS: Bis-diketopiperazine derivatives in clinical
oncology: 1CRF-159. Cancer Clin Trials 3:315-320, 1980.
165. Redaksie VD: Sulphinpyrazone in the prevention of cardiac death after myocardial infarction (editorial).
S Afr Med J 57:301-302, 1980.
166. Reiman AS: Sulfinpyrazone after myocardial infarction: No decision yet (editorial). N Engl J Med
303:1476-1477, 1980.
167. Sauerbruch T, Kaess H: The Anturane Reinfarction Trial (letter). N Engl J Med 303:49-50, 1980.
168. Schultz RM: Less empirical interferon trials in cancer (letter). Lancet 2:1362, 1980.
169. Sebille A, Hugelin A: Muscle reinnervation enhanced by isaxonine in man. Br J Clin Pharmacol 9:275276, 1980.
170. Sklaroff RB, Yagoda A: Methotrexate in the treatment of penile carcinoma. Cancer 45:214-216, 1980.
171. Spiers AS, Kasimis BS, Janis MG: High-dose intravenous infusions of 5-fluorouracil for refractory solid
tumours—the HI-FU regimen. Clin Oncol 6:63-69, 1980.
172. Svedmyr N: General aspects on evaluation of drug effects on cough and expectoration. Eur J Respir Dis
1 IO(suppl 61):8I -92. 1980.
173. Van Durme JP: Evaluation of the antiarrhythmic efficacy of mexiletine in patients with chronic ventricular
arrhythmias. A eta Cardiol Suppl 25:121-125, 1980.
174. Verstraete M: Registry of prospective clinical trials: Fourth report. Thromb Haemost 43:176-181, 1980.
175. Vietti TJ: Evaluation of toxicity: Clinical issues. Cancer Treat Rep 64:457-461, 1980.
362
C. Year 1980 clinical trial publications
176. (no author listed): Ethical and legal considerations associated with clinical field trials. Discussion. J Dent
Res 59(special Issue C): 1267-1270, 1980.
177. (no author listed): Experimental evaluation of antitumor drugs in the USA and USSR and clinical
correlations: Natl Cancer Inst Monogr 55:1-179, 1980.
178. (no author listed): Radial keratotomy—an operation for myopia. Med Lett Drugs Ther 22:97-98, 1980.
179. (no author listed): Small cell carcinoma of the bronchus—real progress is hard to come by (editorial).
Lancet 1:77-78, 1980.
180. (no author listed): Trifluridine (Viroptic) for herpetic keratitis. Med Lett Drugs Ther 22:46-48, 1980.
D. Activities by stage of trial
11. Population(s) from which patients
are to be selected
12. Stratification varibles to be used in
randomization
13. Patient safeguard procedures
14. Projected timetable for the trial
15. Approach to patient close-out, i.e.,
fixed closing date versus fixed
follow-up interval
This appendix lists the activities and functions
required for each stage of a trial. The stages are
(see Chapter 3):
I. Initial design stage
II. Protocol development stage
III. Patient recruitment stage
IV. Treatment and follow-up stage
V. Patient close-out stage
VI. Termination stage
VII. Post-trial follow-up stage (optional)
The time at which certain activities are started
will vary from trial to trial. This should be recog
nized if this outline is used as a management tool
in planning activities for a specific trial. Activi
ties listed in one stage in this schedule may not
begin until the following stage in an actual trial.
B. Organizational structure
I. INITIAL DESIGN STAGE
A. Design specifications
•r. ... ■
I !
t;
Specify the initial design features con
cerning:
1. Purpose and rationale of trial
2. Number and type of test treatments
3. Number and type of control treat
ments
4. Level and method of masking
5. Primary and secondary outcomes
6. Type and frequency of observations
7. Required number of patients based
on a sample size calculation for a
specified outcome, type 1 and II
error, length of follow-up, and
projected treatment differences
to be detected
8. Estimate of number of clinical cen
ters required based on the sam
ple size calculation
9. Number and type of other centers,
e.g., coordinating center(s), cen
tral laboratory, and reading cen
ters)
10. Patient eligibility and exclusion cri
teria
tml
363
1. Develop general guidelines concerning:
a. Desired qualities of the study
chairman and members of the
steering, advisory-review, and
treatment effects monitoring
committees
b. Terms of office of study chair
man and members of key
committees
c. Method of selection of study
chairman and members of
key committees
d. Voting rules for key committees
2. Outline responsibilities of the steer
ing, advisory-review, and treat
ment effects monitoring commit
tees
3. Establish key elements of overall
organizational structure of the
trial
4. Outline meeting schedule for key
committees
5. Specify functions/duties of the co
ordinating centerfs)
6. Outline general plans for paper
writing and authorship proce
dures
C. Patient recruitment, treatment, and
data collection procedures
1. Develop methods for patient re
cruitment and procedures for
randomization
II. Protocol Development Stage 365
364
D. Activities by stage of trial
2. Outline data collection procedures
3. Develop a timetable for testing and
finalization of data collection
forms
4. Determine specifications for drugs
to be evaluated, including meth
ods for bottling, labeling, and
distributing drugs (if a drug trial)
5. Outline main elements of the treat
ment protocol
I
D. Data processing and analysis
1. Outline data processing and data
base management procedures
2. Develop a timetable for implemen
tation of database management
procedures
3. Outline general data analysis plans
4. Outline plans for quality control of
data collection and processing ac
tivities
E. Other activities
1. Perform literature review to iden
tify pertinent background infor
mation on the study treatment,
including review of safety and ef
ficacy data
2. Prepare and submit funding pro
posal
3. Submit research plan and draft con
sent statement to local institu
tional review boards (IRBs) for
preliminary approvals
!
II. PROTOCOL DEVELOPMENT STAGE
A. Patient recruitment and care
1. Identify source of patients
2. Write treatment protocol
3. Determine recruitment goals for in
dividual clinics
4. Develop detailed patient eligibility
and exclusion criteria
5. Establish permissible time windows for prerandomization and
follow-up examinations
6. Establish protocol for manage
ment of clinical conditions
known or suspected to be related
to the disease or therapy under
study
7. Establish procedures for monitor
ing the clinical management of
individual patients
8. Develop patient information and
consent procedures
9. Generate the treatment allocation
schedule and write description of
the methods used to generate the
schedule
10. Develop publicity schemes needed
to facilitate patient recruitment
B. Data processing and analysis
1. Develop procedures for data inter
face between clinics and the data
center
2. Specify contents of data collection
forms
3. Draft and test data collection
forms
4. Submit forms to appropriate study
body for review and approval
5. Develop data management sys
tems, including data collection,
processing, editing, and correc
tion procedures
6. Initiate development of data sys
tem and test computer programs
required for:
a. Verifying patient eligibility
b. Issuing treatment allocations
c. Keying, inventorying, editing,
and updating study data
7. Write study handbook and manual
of operations. Include sections
on:
a. Informed consent procedures
and methods for safeguard
ing patient rights
b. Recruitment and randomiza
tion procedures
c. Patient follow-up procedures
d. Procedures for data collection
and processing
e. Study organizational structure
8. Outline procedures for monitoring
the randomization process
9. Review data storage, backup, and
security procedures
C. Training and communication
1. Develop and implement training
and certification procedures for
key study staff, especially staff at
the clinical centers responsible
for data collection
2. Acquaint staff of all participating
centers with the following:
methods oi
of me
the trial
a. Design and metnoas
iriai
b. Importance of integrity in data
collection
c. Need for data security
d. Study organization
e. Performance and treatment
monitoring procedures to be
employed
3. Initiate regularly scheduled meet
ings of:
a. Study investigative group
b. Steering committee
c. Other study committees
4. Consider distribution of newslet
ters at periodic intervals to
inform staff of participating
centers of study progress and
procedural changes
5. Develop study address directory
and procedures for maintaining
same
6. Outline communication channels
and ground rules for center-tocenter communication
7. Identify center(s) responsible for
coordinating study communica
tions
8. Develop and initiate clinic site vis
iting procedures in order to:
a. Identify and correct problems in
patient recruitment
b. Review administration of study
procedures
c. Identify and correct possible de
ficiencies in data collection
methods
9. Develop procedures for dealing
’ 'inform;
"
la
with requests for study
the
tion originating outside I*,
study
D. Quality assurance
1. Develop procedures for monitoring and reviewing performance
of participating centers in the
trial
2. Outline content of performance
monitoring reports concerning
data collection and data quality
3. Mock up data tables needed for
performance monitoring
4. Develop documentation procedures for.
a. Treatment allocation
b. Reporting primary and second
ary outcomes
c. Classification of primary out
come
d. Modification of data collection
instruments and procedures
e. Data management and forms re
vision
f. Informed consent
g- Modifications of study hand
books and manuals of opera
tions
5. Specify mechanisms for monitoring adequacy of documentation
E. Treatment monitoring
1. Outline content of treatment ef
fects monitoring reports
2. Establish timetable for treatment
monitoring and generation of
treatment monitoring reports
F. Authorship
1. Establish authorship policies for
study papers and for ancillary
studies
2. Establish review procedures for
study papers
3. Establish methods for review and
approval of presentations made
by members of the investigative
group on behalf of the study
G. Management
1. Outline study policy concerning:
a. Informed consent, including
minimum standards for the
I
I
366
I
• t tn 4-. ■
U-
I
HI. Patient Recruitment Stage 367
D. Activities by stage of trial
consent process and mecha
nisms to be used to monitor
clinics for adherence to those
standards
b. Responsibilities for patient care
c. Data security, including specifi
cation of mechanisms to be
used to protect computer
data files against loss or de
struction
d. Access to study data by investi
gators in the study; by people
outside the study
Rights of patients to privacy
and confidentiality
f. Collection and storage of study
records
g. Review and approval of ancil
lary studies performed by
study investigators
h. Communications with persons
outside the study, including
the news media, regarding
study design and results
i. National and local publicity
j. Dissemination of study results
to study patients; to study
staff
k. Acquisition of liability insur
ance for participating centers
and investigators
l. Nature and extent of follow-up
of dropouts
2. Develop guidelines and procedures
required to:
a. Decide when to terminate a
treatment because of adverse
or beneficial effects
b. Modify the general design speci
fications of the trial
3. Create treatment effects monitor
ing committee or some other
body to monitor for treatment
effects
4. Establish system for making work
priority assignments for the data
center and other resource centers
in the trial
5. Designate the group responsible
for dissemination of study infor
mation and results to the lay and
scientific communities
6. Develop safeguards to protect
against premature disclosure of
study results to parties outside
the study
H. Other activities
1. Develop projected budget and staf
fing requirements for the trial
2. Order study drugs and initiate pack
aging, labeling, and distribution
procedures (if a drug trial)
3. Recruit staff at participating cen
ters
4. Develop informational brochures,
official study name, logo, etc.
5. Obtain Investigational New Drug
Application (INDA) if required
by the Food and Drug Adminis
tration (FDA)
6. Negotiate specialized contracts or
agreements, such as study liabil
ity insurance, equipment con
tracts, etc.
7. Designate official repository for
study documents, i.e., minutes of
meetings, performance and treat
ment effects monitoring reports,
completed data collection forms,
etc.
8. Print and distribute essential mate
rials, such as recruitment mate
rials, forms, study handbooks
and other manuals
9. Create necessary committees, be
ginning with the steering com
mittee
10. Review and refine general design
specifications
11. Evaluate preparedness of all cen
ters to initiate patient recruit
ment and follow-up
III. PATIENT RECRUITMENT STAGE
A. Treatment and patient care
1. Establish liaison with appropriate
medical and lay societies to facil
itate patient recruitment
2. Establish channels for patient re
ferral
3. Initiate local and national publicity
campaigns for patient recruit
ment when appropriate
4. Inform local referring physicians of
aims of study and of limits of
study responsibility regarding pa
tient care
5. Provide clinical centers with pa
tient information brochures
6. Review progress in patient recruit
ment
7. Project time requirements for com
pletion of patient recruitment
based on recruitment performance
8. Establish date for termination of
patient recruitment and inform
patient referral sources of date
B. Data processing and analysis
1. Implement procedures for monitor
ing the treatment allocation pro
cess
2. Initiate and maintain procedures
for collecting and processing
study data and related materials
(e.g., ECGs and fundus photo
graphs)
3. Verify that incoming baseline data
forms document patient eligibil
ity for the trial
4. Initiate data editing procedures to
provide checks for accuracy and
consistency within and across
forms
5. Establish and maintain monitoring
procedures to identify deficien
cies in data collection and data
processing
6. Establish procedures to be fol
lowed in maintaining adherence
to the examination schedule
7. Define responsibilities of clinical
centers and of data center for lo
cating patients lost to follow-up
8. Define responsibility of clinical cen
ters in maintaining contact with
dropouts
9. Review data storage, backup, secu
rity, and integrity procedures in
all participating centers
10. Identify topics for analysis and
methods to be used for analysis
11. Outline quality control procedures
for data analyses
12. Complete final editing of recruit
ment and entry data
C. Training and communication
1. Maintain training and certification
procedures for staff of partici
pating centers
2. Continue site visits to participating
centers
3. Initiate regular meetings of the
treatment effects monitoring
committee
4. Continue to hold regularly sched
uled meetings of:
a. Study investigative group
b. Steering committee
c. Other study committees
5. Initiate, if appropriate, preparation
of a newsletter to inform staff of
study progress and procedural
changes
6. Consider central preparation and
local distribution of newsletter
for study patients
7. Update and distribute study ad
dress directory at periodic inter
vals
D. Quality assurance
1. Initiate external monitoring and re
view procedures for all centers in
the trial, e.g., clinical centers,
data center, central laboratory,
reading centers, etc.
2. Review adequacy of the randomi
zation procedure
3. Prepare periodic reports summa
rizing performance of clinical
centers with regard to patient re
cruitment
4. Prepare reports summarizing:
a. Adherence to study protocol
b. Adherence to data collection
and patient examination sche
dule
c. Data quality
E-
W
368
D. Activities by stage of trial
d. Data collection activities at clin
ical centers
e. Data processing activities at the
data center
f. Activities at other resource cen
ters
5. Review documentation standards
and monitor adequacy of docu
mentation for:
a. Treatment allocation procedures
b. Reporting of major events
c. Classification of cause of death
or other major events
d. Informed consent procedures
e. Modification of data collection
forms
f. Modification of study protocol
g. Data management and data
analysis procedures
I
E. Treatment monitoring
1. Develop analytic techniques and
computer programs needed to
monitor study data for evidence
of adverse or beneficial treat
ment effects
2. Begin generating treatment moni
toring reports
F. Authorship and publications
1. Establish paper writing teams and
■
lv ■
i'
I
I
il
schedules
2. Review and, if necessary, revise
guidelines for authorship
3. Write paper(s) on design and
methods of the study
4. Establish procedures for distribu
tion of published papers to the
study group
5. Distribute at periodic intervals up
dated listings of study publica
tions and presentations to all par
ticipating investigators
G. Management
1. Review and, if necessary, revise:
a. Informed consent procedures
b. Guidelines on patient care
c. Guidelines on study publicity
IV. Treatment and Follow-Up Stage 369
d. Guidelines on data access, secu
rity, backup, and storage
2. Continue to review and assign
priorities to activities carried out
by the data center and other key
resource centers in the trial
3. Initiate periodic review of study
committee structure; dissolve
nonfunctional committees
4. Develop patient close-out strategy
H. Other activities
1. Print and distribute data collection
forms and related materials
needed for follow-up examina
tions
2. Initiate reporting procedures with
FDA (if trial involves INDA)
3. Establish a central resource of data
slides on study design and find
ings for use by study investiga
tors in making presentations con
cerning the trial
4. Implement procedures for docu
menting important events in the
trial that may effect data quality
IV. TREATMENT AND FOLLOW-UP
STAGE
A. Treatment and patient care
1. Monitor and report adverse treat
ment effects
2. Review procedures for monitoring
and evaluating the clinical man
agement of individual patients
B. Data processing and analysis
I. Continue procedures for maintain
ing adherence to follow-up visit
schedule, including special proce
dures for patients classifed as
inactive or lost to follow-up
2. Review and expand data edit proce
dures to provide checks for con
sistency within and across forms,
including forms used for preran
domization visits
3. Review and expand, if necessary,
monitoring procedures to iden
tify deficiencies in data collec
tion and processing
4. Initiate central coding procedures,
if required, for cause of death or
other primary outcome variables
5. Revise data collection forms as nec
essary
6. Reproduce and distribute data col
lection forms as needed
7. Develop and test data collection
forms and data management sys
tems required for patient close
out stage
8. Prepare to process data collected at
close-out visit(s)
9. Revise data management proce
dures as appropriate
10. Monitor patient adherence to as
signed treatment
C. Training and communications
1. Continue site visits to participating
centers to:
a. Review administration of study
procedures
b. Identify and correct the prob
lems in data collection and
processing
2. Continue to hold regularly sched
uled meetings of:
a. Study investigative group
b. Steering committee
c. Treatment effects monitoring
committee
d. Other study committees
3. Continue to publish newsletter (if
initiated):
a. To inform study investigators of
study progress and proce
dural changes
b. To inform study patients of
study progress
4. Continue (modify if necessary) cer
tification procedures for staff of
participating centers
5. Update and distribute revised study
handbooks and manuals of op
erations
6. Continue to update and distribute
study address directory, mailing
labels, forms (when necessary),
etc.
7. Communicate with personnel of all
participating centers concerning
timetable and procedures for
close-out stage, especially in the
use of new data collection forms
or techniques
D. Quality assurance
1. Review and revise study quality as
surance procedures as appro
priate
2. Continue to prepare monitoring reports summarizing:
a. Adherence to study protocol
b. Adherence to data collection
and examination schedule
c. Quality of data generated
d. Data collection activities at clin
ical centers
e. Data processing activities at the
data center
3. Continue periodic review of ade
quacy of documentation for:
a. Reporting of primary and sec
ondary outcomes
b. Classification of causes of death
or other major outcomes
c. Modification of data collection
forms
d. Modification of study protocol
e. Ongoing data management
procedures
E. Treatment monitoring
1. Revise data analysis and reporting
procedures required to monitor
study results for evidence of ad
verse or beneficial treatment ef
fects
2. Initiate procedures for review and
implementation of protocol mod
ifications recommended by the
treatment effects monitoring
committee
F. Authorship and publications
1. Paper writing activities:
a. Establish and activate writing
teams
370
V. Patient Close-Out Stage 371
D. Activities by stage of trial
b. Develop papers on study find
ings
c. Develop papers on natural his
tory and ancillary studies
2. Draft paper containing main study
conclusions
3. Prepare stand-by press release for
use in the event study findings
appear in the news media prior
to publication in a scientific jour
nal
4. Continue procedures for distribu
tion of published papers to study
investigators
5. Update and distribute listing of
study publications to study in
vestigators
G. Management
1. Review and, if necessary, revise
study guidelines concerning:
a. Patient care
b. Inquiries from outside the study
regarding study design or
study results
c. Rights of access to study data
d. Informed consent procedures
e. Information sources to be used
in classification of major
events
2. In the event of a major protocol
change or in preparation for
close-out discuss and establish
procedures for:
a. Informing patients of close-out
b. Informing referring physicians
of close-out
c. Developing special data collec
tion forms for close-out
d. Developing special patient con
sent forms for post-trial fol
low-up (if planned)
3. Develop detailed patient close-out
procedures
4. Establish timetable for close-out ac
tivities, including timetable for
termination of funding for data
collection at participating clini
cal centers
5. Recommend procedures to be fol
lowed at clinical centers in termi
nation of patient follow-up
6. Prepare special informational mate
rial to be dispensed to patients at
the close-out examination
7. Develop plan to monitor for possi
ble adverse effects associated
with treatment in the close-out
stage
8. Review data storage, backup, secu
rity, and integrity procedures
9. Review and revise organizational
structure of the trial
10. Continue making priority assign
ments for data analysis activities
H. Other activities
1. Continue procedures for distribu
tion of study drugs during the
follow-up phase
2. Continue FDA reporting proce
dures (if trial involves INDA)
3. Reproduce and distribute data col
lection forms to be used during
close-out stage
4. Review timetable for remainder of
study and prepare request for ad
ditional funding (if necessary)
5. Update central repository of data
slides for use by study investiga
tors in making study presenta
tions
V. PATIENT CLOSE-OUT STAGE
A. Treatment and patient care
1. Initiate procedures for unmasking
treatment assignments, when
masking is involved
2. Monitor for adverse effects due to
treatment termination (if appro
priate)
3. Inform the study investigators of
study results and their implica
tions for patient care
4. Formulate treatment recommenda
tions to be made to patients on
close-out
5. Inform patients of study results
and recommended future treat
ment
6. Inform patients’ primary care phy
sicians of study results
7. Update patient identifying informa
tion to facilitate post-trial fol
low-up
8. Inform patients of plans for post
trial follow-up (if any) and of
methods to be used to maintain
contact with them after termina
tion of regular follow-up
9. Request data center to prepare
summaries of accumulated fol
low-up data needed by clinic per
sonnel when reviewing care
needs of individual patients and
for facilitating transfer of perti
nent data to primary care physi
cians
10. Effect transfer of patient care re
sponsibilities to appropriate
sources
11. Document that all active patients
have been informed of study re
sults
B. Data processing and analysis
1. Perform final edit of accumulated
study data. Identify items that
need review or corrections
2. Initiate special search procedures
to locate patients classified as
lost to follow-up for final data
analysis
3. Develop and carry out data anal
yses that summarize study find
ings, including results from the
close-out process
C. Quality assurance
1. Prepare final reports summarizing:
a. Adherence to study protocol
b. Adherence to data collection
and examination schedule
c. Quality of data generated
d. Performance of all participat
ing
D. Treatment monitoring
1. Prepare final report on treatment
effects
2. Hold final meeting of treatment ef
fects monitoring committee
3. Monitor close-out process to en
sure adherence to indicated pro
cedures for separation of pa
tients from the trial
E. Authorship and publications
1. Write and submit for publication
papers) summarizing study re
sults
2. Supply advance copy of paper(s)
on results to participating clinics
for distribution to staff and re
ferring physicians
3. Establish coordinated approach to
dissemination of information to
the medical public in conjunc
tion with publication of study re
sults, including press releases, if
needed
4. Develop paper writing and analysis
plans for termination stage of
study, including procedures for
review of papers prepared in that
stage
5. Develop mechanism for support of
travel and work of writing teams
and leadership committees dur
ing termination stage
6. Develop mechanism for distribu
tion of papers published during
termination stage
7. Update and distribute list of publi
cations to study investigators
F. Management
1. Develop plans and policies for:
a. Disposition of equipment pur
chased with study funds
b. Disposition of unused study
drugs and other supplies
c. Disposition of centrally stored
study materials, such as
frozen serum specimens,
ECGs, fundus photographs
d. Disposition of patient medical
records and materials accum
ulated at clinical centers in a
manner consistent with local
statutes
372
e. Final disposition of study data
on completion of the termi
nation stage
2. Review study organizational struc
ture and propose revisions for
termination stage
3. Review timetable for termination
stage and develop plans for addi
tional funding (if necessary)
4. Develop plans for disengaging
study investigators and related
staff from the trial
I
VII. Post- Trial Follow- Up Stage
D. Activities by stage of trial
G. Other activities
1. Update central repository of data
slides for use by study investiga
tors in making presentations of
study findings
2. Update and distribute address di
rectory
2. Implement plan for authorship and
review of manuscripts prepared
during termination stage
3. Distribute copies of finished papers
to study investigators
4. Distribute updated list of study
publications to participating in
vestigators
D. Management
1. Establish policy on type and
amount of data to be made avail
able outside the study structure
during and after the termination
stage
2. Establish approach for dealing
with inquiries regarding the
study and the study results
3. Establish study stewardship body
after termination of the trial
E. Repository
VI. TERMINATION STAGE
A. Treatment and patient care
J. I
1. Verify that all clinics have com
plied with close-out procedures
(e.g., verify that all patients have
been informed of study results,
have been taken off their as
signed treatment, and have had
care responsibilities transferred)
2. Review and revise, if necessary, con
clusions of the study after com
pletion of final data analysis
B. Data processing and analysis
i
I.-'
■
1. Complete final edit procedures and
create final data file
2. Prepare data listings or data tapes
(disks) needed by writing teams
3. Update backup tapes and reposi
tory listings of data files
C. Authorship and publications
1. Implement plans for support of
travel and work of writing teams
and leadership committees re
maining in operation in termina
tion stage
il
1. Establish a suitable repository for
all study materials to be retained
beyond the termination stage,
such as:
a. Patient medical records
b. Completed data forms
c. Associated records, such as
ECGs and fundus photo
graphs
d. Serum samples, biopsy slides,
etc.
2. Establish a central repository for
patient identifying and locator in
formation
3. Store computer data files and
backup copies
F. Other activities
1. Dispose of unused study drugs (if
any)
2. Complete phase-out of study per
sonnel
3. Dissolve all remaining study com
mittees, except for stewardship
committee
4. Submit final report to FDA (if trial
involved INDA)
5. Inform IRB of completion of study
6. Prepare final version of data slides
for use by study investigators in
presenting results of the trial
7. Maintain communications with
participating investigators
VII. POST-TRIAL FOLLOW-UP
STAGE (optional)
A. Preparatory steps
1. Re-establish contact with clinics to:
a. Inform them of the proposed
follow-up
b. Update address directory
c. Address special procedural and
ethical questions posed by the
follow-up
2. Assemble roster of patients to be
followed
3. Outline approach to be used in the
follow-up and the amount of
data to be collected
4. Obtain IRB approval for the pro
posed follow-up study
373
B. Follow-up
1. Initiate mechanisms to establish pa
tient contact
2. Implement special procedures to lo
cate patients who do not re
spond to initial contact or who
cannot be located
3. Update patient identifying and lo
cator information
4. Assemble data collected from the
post-trial follow-up
C. Data analysis and publication
1. Link added follow-up data with ex
isting patient files
2. Carry out analyses to summarize
post-trial follow-up results
3. Prepare manuscript summarizing
results
D. Other activities
1. Store updated patient identifying
and location information for fu
ture follow-up
2. Distribute a copy of any manuscript(s) produced to study inves
tigators
E.l Consent Statement for MPS 375
E. Sample consent statements
Table E-l Content checklist* for sample consent statements
Topic
E.l
E.2
E.3
Consent statement for the Macular Photo
coagulation Study (MPS): Senile Macu
lar Degeneration Study
Consent statement for the Persantine As
pirin Reinfarction Study (PARIS)
Consent statement for the Hypertension
Prevention Trial (HPT)
Table E-l
Content checklist for sample con
sent statements
This appendix contains consent statements from
three of the multicenter trials sketched in Ap
pendix B:
• Macular Photocoagulation Study (MPS): Se
nile Macular Degeneration Study
• Persantine Aspirin Reinfarction Study
(PARIS)
• Hypertension Prevention Trial (HPT)
'P
T. :. ;
The statements used by the individual clinics may
have differed. All clinics could expand on infor
mation contained in the prototype statements,
but could not delete or abridge information.
Table E-l provides a content analysis of the
three statements. The checklist is based on Table
14-4. None of the statements covered all of the
items in the checklist. The MPS statement had
the largest number of content deficiencies—16.
There were 9 noted for the PARIS and 7 for the
HPT statements.
The last line in Table E-l gives the reading
grade level of the text, derived using a scoring
system developed by McLaughlin (1969). The
PARIS statement had the highest reading level.
It also contained language designed to speak for
the patient (e.g., as in use of the phrase / under
stand)—a defect largely avoided in the other two
statements.
E.l CONSENT STATEMENT FOR THE
MACULAR PHOTOCOAGULATION
STUDY (MPS): SENILE MACULAR
DEGENERATION STUDY
Macular degeneration is a major cause of visual
loss in the U.S. It results from aging changes
374
that affect the pigment cells and small blood
vessels behind the retina. There is a familial
tendency. No form of treatment is known to be
effective.
Sometimes, a progressive deterioration of the
pigment cells results in a slow reduction in vi
sion. At other times, a break in the membrane
behind the retina permits a blood vessel to grow
through and leak fluid and/or blood beneath the
retina and into the retina causing a rather sud
den loss of vision.
Photographs taken after fluorescein dye injec
tion locate the position of the leaking blood
vessel. The closer this vessel is to the center of
the retina, the greater the threat to vision. If the
vessel is somewhat removed from the center, it
may be possible to close the vessel with the laser
and prevent further growth and further bleeding.
If the vessel is directly central, laser treatment is
not recommended.
Prior to laser treatment, numbing medication
is given to prevent discomfort and motion of the
eye. Rarely, this injection may cause some swell
ing behind the eye and some visual loss.
Possible complications of laser treatment in
clude bleeding, retinal wrinkling, pigment loss,
and damage to the center of the retina which
may cause vision to be worse than before treat
ment. It is important to recognize, however, that
each of these problems can occur without the use
of the laser.
At present, it is not known whether your
chances of maintaining good central vision are
better with or without laser treatment, and your
doctor has agreed to participate in a randomized
trial to find an answer to this question. If your
eye is eligible and if you agree to participate, you
will be assigned in random fashion to a treat
ment group or to a non-treatment group. Ran
domization is similar to flipping a coin so that
there is one chance in two of being treated or not
treated with the laser. This provides an opportu
nity to balance the risks and benefits of laser
treatment.
There is another important benefit. Patients
with a second eye at risk have much to gain from
participation since we anticipate the results of
this study may provide information that will
help the ophthalmologist with the management
of the second eye.
MPS
PARIS
HPT
General descriptive and design information
• Disease or condition to be studied
• Type of people to be studied
• Length of follow-up
• Method of follow-up
• Description of data collection procedures and
schedule
Treatment procedures described
• Test treatment
• Control treatment
• Method of treatment administration
• Method of treatment assignment
• Treatment masking
• Rationale for choice of treatment
• Alternative treatments
Risk-benefit information detailed
• Nature of treatment side effects
• Risk-benefit of test treatment
• Risk-benefit of control treatment
• Potential side effects of test treatment
• Potential side effects of control treatment
• Risk-benefit of special procedures to be per
formed
Patient responsibilities and safeguards
• Patient follow-up responsibilities detailed
• Provisions for protecting patient from prolonged
denial of a beneficial treatment or exposure to
a harmful treatment detailed
• Safeguards for protecting patient's right to
privacy detailed
• Patient’s right to withdraw from trial stated
• Statement of right to have questions answered
before enrollment provided
• Types of information that will not be disclosed
during trial detailed
• Policy on care and payment for study related
injuries detailed
• Statement of where and how personal identifiers
will be used
• Amount and type of information provided to
patient during trial indicated
• Amount and type of information provided to
patient at end of trial indicated
SMOG gride
NA
NA
NA
NA
12
NA
NA
NA
NA
14
12
•A checkmark indicates the item was covered. A dash indicates it was not. NA indicates not
applicable.
376
E. Sample consent statements
Results of the study will be analyzed regularly;
and any significant findings will be made known
to the patient, especially findings which would
alter the management of either eye.
To be certain that the physician will be able to
locate all patients in case there are important
study findings, we ask you for the names of
relatives and others who may be contacted in
case the patient cannot be located at some future
date.
My signature below indicates that I under
stand all of the above and agree to participate in
this program. I recognize that I am under no
obligation to join this study, that I am free to
withdraw at any time, and that neither failure to
join nor withdrawal will prejudice the medical
care which I receive at the Johns Hopkins Medi
cal Institutions.
Patient’s Signature
__________________
Doctor’s Signature
__________________
Witness
__________________
Date
E.2 CONSENT STATEMENT FOR THE
PERSANTINE ASPIRIN REINFARCTION
STUDY (PARIS)
I agree to permit Dr.and the
study physician to treat me,, with
Persantine (dipyridamole) and/or aspirin (ace
tylsalicylic acid) or placebo (inactive substance)
for coronary heart disease. It has been explained
to me that some studies have shown that aspirin
may produce favorable results in the treatment
of heart attacks and blood clot disorders by
affecting the function of the blood platelets (the
small cells in the blood that are necessary to
keep blood from clotting [sic]). There is also
some indication that the addition of Persantine
may enhance these effects. However, at present
there are no clear-cut data showing that people
having one heart attack will not have another
heart attack if they are given drugs which change
platelet function. This study is designed to test
the possibility that these drugs may help to pre
vent a recurrent heart attack.
It has been further explained to me that all of
the persons participating in this study will be
assigned at random to one of the following three
groups. Neither I nor my physician will know to
which group I belong.
E.3 Consent Statement for the HPT 377
1. One group will be given aspirin. Along
with aspirin this group will receive a
placebo pill (inactive substance) resem
bling the second drug, Persantine. Per
sons in this group will take 325 mg of
aspirin three times a day.
2. A second group in this study will take as
pirin and Persantine. Persons in this
group will take 325 mg of aspirin and
75 mg of Persantine three times a day.
3. The third group will be what is called the
control group. Those selected at random
for this group will be given two placebos,
that is, pills appearing like the other
drugs in the study but with no active
medication in them. This group will be
compared with the other groups to see if
there is a difference between the groups
taking medication and the group taking
placebos.
I understand that aspirin may cause stomach
irritation or a bleeding tendency. No one who
has had a personal sensitivity or history of signif
icant problems taking aspirin will be asked to
participate. I also understand that Persantine
may cause lightheadedness or dizziness as a re
sult of lowered blood pressure. I understand that
there are no data in humans to suggest that its
long-term use will lead to the development of
any serious illness. Studies are in progress to
determine whether or not this drug has any po
tential to cause tumors in animals. Should infor
mation become available suggesting any adverse
effects, I understand that I would immediately
be notified by the study physicians and that any
such measures deemed necessary for the protec
tion of my health would be carried out
promptly. I also understand that if I should have
any reactions to the medication during the
study, I should notify the study physician so that
appropriate treatment could be carried out
promptly.
I understand that the reason for the placebo
group is that there is always a chance, in fact a
reasonable possibility, that the drugs may be
ineffective and may cause some side effects
which outweigh any beneficial effects. It has
been explained to me that if at any point during
this study it becomes evident that patients on
one treatment are not doing as well as the pa
tients on the other treatments, they will be
switched to one of the superior treatments.
It has also been explained to me that there are
other methods of possible prevention of recur-
rent heart attack including diet, weight reduc
tion, exercise programs, drug control of high
blood pressure, cessation of smoking, various
types of heart surgery, or use of anticoagulants
which reduce the clotting ability of blood with
out affecting blood platelets. I understand that
all of these forms of treatment, with the excep
tion of anticoagulant drugs which cannot be
given together with PARIS medications, are
available to me with the advice of my personal
physician. I also understand that I must avoid
taking aspirin and aspirin-containing drugs
during my participation in this study.
I have read and have understood the forego
ing explanation and I fully understand the pro
gram of study. I also understand that my partici
pation in this study is of great value to me and to
others like myself who have coronary disease
and have survived a heart attack. I have had an
adequate chance to ask questions and I may ask
questions at any time while the study is in pro
gress. I voluntarily consent to participate, or to
continue my participation, in this study and to
treatment with these drugs or placebo. I under
stand that I may withdraw my consent and dis
continue my participation in the study at any
time.
Witness:
Signature
Address:
Street
City
Zip Code
State
Signed:
Patient or person authorized to consent
for the patient
Address:
Street
City
Today’s Date:
Zip Code
State
____
Month
Day
Year
Source: Reference citation 375. Reprinted with permission of the
American Heart Association. Inc., Dallas. Texas.
E.3 CONSENT STATEMENT FOR THE
HYPERTENSION PREVENTION TRIAL
(HPT)
The Hypertension Prevention Trial (HPT) is de
signed to help us determine whether or not peo
ple can avoid high blood pressure through a
change in diet. Only volunteers who are not
being treated for hypertension and who do not
now have elevated blood pressure are eligible for
the study. The main aim is to determine if people
can make, and then maintain, the diet changes
proposed. Each person will be studied for at
least two years. The amount of change in blood
pressure will be observed over that time period.
The specific diets to be studied are:
• Low-sodium diet—participants assigned to
this diet will be asked to reduce the amount
of foods they eat which contain high levels
of sodium. Sodium is a mineral required by
the human body in small amounts. The
most common source of sodium is table
salt (sodium chloride).
• Low-sodium and high potassium diet—par
ticipants assigned to this diet will be asked
to reduce the amount of high-sodium foods
in their diet and also increase their use of
foods which are high in potassium. Potas
sium is also a mineral required by the
human body. The most common sources of
potassium are fresh fruits and vegetables.
• Weight loss diet—participants assigned to
this diet will be asked to reduce body
weight by eating less and through increased
exercise.
• Weight loss and low-sodium diet—partici
pants assigned to this diet will be asked to
lose weight by eating less, increasing exer
cise, and to reduce consumption of highsodium foods.
People entering the study will be placed in one
of two weight groups. Approximately half the
people enrolled are expected to fall into each of
the two groups. People in the higher weight
group will be assigned to one of the above diets,
or to no diet change. Those in the lower weight
group will be assigned to one oi the two diets
above that do not involve a weight loss, or to no
diet change.
The specific diet you will be asked to follow
depends on an assignment made by the Data
Coordinating Center located in Baltimore, Mary
land. The assignment is made using a chance
378
E. Sample consent statements
procedure, like rolling dice, but carried out with
a computer. Hence, neither we nor you know in
advance the group to which you will be assigned.
Everyone taking part in the study will need to
return to the clinic for blood pressure measure
ment, every six months after enrollment. A 24hour food record and overnight urine collection
will be required prior to each of these visits. An
electrocardiogram and a blood sample will be
drawn once a year.
Participants assigned to a diet group will need
to make more frequent visits to the clinic to
receive instructions on how to make necessary
diet changes. As a rule, these visits will be organ
ized as group instruction sessions. Each session
will last about one and one-half hours. The
schedule will be:
• One session a week for the first ten weeks
after enrollment
• Six additional sessions during the rest of the
first year
• A minimum of four sessions in the second
year, and in each year thereafter so long as
the study continues
I
a
fl' »I . >. gf
I ’•?
PH
Entry into the study is voluntary. If you do
decide to enroll we will keep in contact with you
for the duration of the study. We may even wish
to maintain written or telephone contact with
you after the study is finished.
You are not obligated to continue in the study
if you change your mind about participating
later on. Withdrawal from the study will in no
way affect any care or treatment you are receiv
ing from other clinics in the University. How
ever, anyone who drops out after enrollment
reduces the scientific value of the study. Hence,
you should not enroll if you are uncertain about
the value of the study or if you know now that
you cannot fulfill the study requirements. If you
do decide to drop out, we will contact you by
telephone or letter about once a year.
Data, including peronal information such as
name and address, will be stored in a computer
at the Clinic and at the Data Coordinating Cen
ter in Baltimore. All personal information is con
fidential and available only to study personnel.
No reports will be presented or published which
reveal your identity.
There will be no charge to you for any of the
study procedures or clinic visits. The examina-
’ r are not in
tion procedures used’ in this study
tended for individual medical diagnosis or care.
If you have any medical problems that come to
our attention, we will refer you to proper sources
of care.
The University has no general provision for
compensation in the event of physical injury
resulting from research studies. In the unlikely
event that injury should occur, medical treat
ment is available, but may not be provided free
of charge.
We believe the risks associated with participa
tion in this trial are small. The diets being
studied are consistent with general nutritional
recommendations. The tests and procedures are
standard and in common use.
If you are satisfied with the explanation of the
Hypertension Prevention Trial and wish to take
part in it, please sign below. Please do not sign
until you have had all questions concerning the
study and your participation in it answered to
your satisfaction.
Participant
F. Data items and forms illustrations
F.l3 Unit specifications
F. 14 Precision specifications
F.l5 Calculation items
F.16 Instruction items
F.l7 Age and birthdate items
F.l8 Reminder and documentation items
F.l9 Full-page versus two-column layout
F.20 Layout for SKIP items
F.21 Instructional information
F.22 Unformatted responses
F.23 Formatted responses
F.24 Layout for check positions
F.25 Field designations and precoded responses
Item numbering
Items that indicate presence or absence of
a finding or condition
F.3 Unnecessary words
F.4 Double negatives
F.5 Compound questions
F.6 Comparative evaluations
F.7 Inverted meaning of a yes reply
F.8 Presence versus absence of a condition
F.9 Time references
F.10 Direction of response
F.ll Leading questions
F. 12 Vertical versus horizontal response lists
F.l
F.2
This appendix contains illustrations referenced in Chapter 12. Many of the items are•
fr°m study
Name (print)
Signature
Date
aCThe cat*tion of an illustration as satisfactory or unsatisfactory relates to the point constdered.
Witness
A satisfactory classification in one context may be unsatisfactory in another.
Name (print)---------
Signature
Date
Should you have any complaint after enrollment, you
may contact the HPT Clinic Director, or the University
Human Volunteers Office. The Human Volunteers Office
should be contacted if you desire to make the complaint
without knowledge of the clinic staff.
Clinic Director:
Harry J. Jones, M.D.
684 North Fox Street
Any Town, MD 47150
Phone: 872-8420
Jane V. Moore, M.D.
Human Volunteers Office:
7844 East Sth Street
Any Town, MD 47150
Phone: 962-7741
CLINIC STAFF: Please make a copy of the signed con-
sent form for the participant
F.l ITEM NUMBERING
F.1.1 Unnumbered items (Unsatisfactory; photo reproduction)
Study No.:
Date of
Record Rev.:
/
/
(mo.) (day) (yr.)
Name of
Reviewer:
Name of Hospital:
Name of Patient:
Address:
Name in
Which listed:
Telephone No.:
History
No. :
Date of Birth:
Medicare
Number:
Soc. Sec.
Number:
/
/
(mo.) (day) (yr.)
Age:
379
Insurance
Number:
F.2 Items that indicate presence or absence of a finding or condition
380
F. Data items and forms illustrations
F.1.2
Numbered items (satisfactory; photo reproduction)
(Unsatisfactory; photo reproduction)
F.2.2
15.
*. OPO NO.
4. HOSPITAL NO.
381
had?
Could you list all the >eta (types of animals) you/
Index Subject
0 ■ C
I
tea thia pat
B. SPKCIAL NO.
aver alcA
prior to yoar/
a.
B. PIRBT NAMK
. 7. MIOOLK
1 ». MAIOCN
COMPLETE ONLY
IF NEEDED BY
HOSPITAL
B. AOORKSB (Str—t and Nmbr)
(City, Zana and Stalo)
10. TCLKPMONK NO.
II. IOC
Did you haw*
clow oontact
with thlB pat?
•.9., did ha ait
Pet
on your 1r>. •1®bPin
on your bad?
hoepl tAllzetlai/
fiiyalcian vialt
?
iEntK Year
YES
ia. date RsataTCRCo
W®.
Day
1
11.DATE FORM INIT1ATCO
W®.
Day
Yoar
□w
□D
□ SEP.
MARITAL STATUS
□s
LMP
,
Day
Mo.
us
IAMPLINR PRAMS PATIENT
□ PrtTOte
SELECTED FOR
STUDY
□w
F.1.3
□ OR
4
NOT SELECTED
FOR STUDY
i Syatamatle Smnpllnf
□ Baaed on
1 Special Sempllnf (Spodty)
□ Boted on
1 Sam* 11 n* Oaaign
□ For Othor Reotona
1 (Spoelty bolom)
Comment
F.2.3
ITEMS THAT INDICATE PRESENCE OR ABSENCE OF A FINDING OR CONDITION
F.2.1
15.
(Unsatisfactory; photo reproduction)
GENERAL MEDICAL HISTORY: Have you ever had any of tha following conditions?
For each yes in column 1, pleasa fill In columns 2 to 7.
(1)
(2)
First
occurrence
Condition
Cataracts
Any other eye
problems (specify)
Heart trouble of
any kind
Stroke
Check
if yes
(Yr.)
(3)
First
seen by
physician
(Yr.)
UNK
□ Other
□ PR
Item numbering is mandatory if the respondent is required to skip certain items. However, it is useful
even when skips are not required. The numbers serve as convenient references for data editing and
processing.
F.2
Oouia you tall ao
tho we
location
of tha vatarlnarlai
traatad thia pat?
*«K1 OP
aESTATION
□N
□Baa®d on
□ Clinic
Wwt ymr wm
this pet BicA?
Yoar
RACC
□ cu
>
I.
ACC
IS. DATE OP BIRTH
Yoar
to
F.
Oould you tall aa
tha naaa of tha
dlaaaaa?
(4)
Treated
currently
(yes or no)
(5)
Current or most
recent physician
and/or clinic
(Name & address)
(Satisfactory; photo reproduction)
137 Has a doctor ever
told you that you had
ASK (a-c) FOR EACH OF
THE FOLLOWING:
Varicose veins_________
Phlebitis (Inflammation of
veins usually in arms or
legs)
_____
Repeated Vaginal Infections (more than 3/year)
Repeated Pelvic and/or
Uterine (female) Infections
Venereal Disease
Stroke
High Blood Pressure
Heart Disease
(Specify:
Anemia (poor blood)
1
(A) IF YES
ASK (B-C)
YES
NO |UNK
(B) IF YES
At what age?
(C) IF YES TO
(A): Were you
hospitalized?
YES NO‘ UNK
382
F. Data items and forms illustrations
F.2.4
(Satisfactory; photo reproduction)
Have you ever been pregnant? NO O YES O
F.5 Compound questions
F.4
b. How many times?----- --------------- ------------------------------------
(If yes, please complete table below listing all pregnancies, beginning with the first pregnancy. Include miscarrii
(If no, go to page 18)
Pregnancy
order:
No.
Child's
first
name
Date
pregnancy
ended or
date of birth
Residence during preg
nancy, list all If
more than one
(No. of rr.os, in each)
Physician and/or
hospital
(Nama ft address)
383
DOUBLE NEGATIVES
F.4.1
F.4.2
(Unsatisfactory; facsimile)
05:38. In the past week, did you have any
injuries or accidents that didn't cause you to
cut down on your normal activities?
0
1
NO
(Unsatisfactory; contrived)
The answers to items 28 thru 32 have been
reviewed by a nutritionist and they do not
disqualify the candidate from enrollment.
(
YES
Pregnancy outcome
and no. o< months pregnant*
F.4.3
Yes
)
(
No
)
(Satisfactory; contrived from F.4.1)
(Unsatisfactory; contrived)
F.4.4
Since your stroke are you (check one)
week did you have any accidents
In the past
[
or injuries?
(
(
(
(
3.
)
)
)
)
Unable to walk
Barely unable to walk
Able to walk fairly well
Able to walk normally
0
NO
1
YES
If yes.
Did they cause you to cut down on your
normal activities?
F.2.5
Comment
0
F.2.1 is deficient in that it allows the respondent to skip conditions that are absent. This design can
lead to ambiguous results, especially when none of the conditons are checked. There is no way in such
cases to know if the entire item was overlooked or if it was left unanswered because none of the
conditions were present.
F.2.2 is defective in that it provides no indication of what the index subject is to do when the answer
to the lead question is “No.” Taken literally, a “No” reply means the respondent is unable to list all his
pets. Further, the time period to which the question refers is unclear. Is the intent to list all pets the
index subject has ever had or only those for some specified time period? Presumably the entire set of
questions is to be skipped for an index subject reporting no pets, creating the same ambiguity as cited
for F.2.1.
The items shown in F.2.3 and F.2.4 avoid this problem by requiring an answer even if the condition
is absent. Example F.2.3 requires a “Yes,” “No,” or “Unknown” answer for each disease condition
listed; example F.2.4 has a lead question that, if checked “No,” allows the respondent to skip the rest
of the item.
[... 1
R
I
■I
F.3
F.3.1
(Unsatisfactory; facsimile)
Would you please tell me about how much
income you and your family will get during
1971, January
Januar through December? I mean your
total family income - before taxes from all
sources.
1. Under $3,999
2. $4,000 - $7,999
3. $8,000 - $11,999
F.3.3
F.3.2
(Satisfactory; contrived from F.3.1)
Please indicate your total, before tax, family
income for 1971.
1. Under $3,999
4. $12,000 - $15,999
2. $4,000 - $7,999
3. $8,000 - $11,999
5. $16,000 - $24,999
6. Over $25,000
-
(
) (
Yes
No
)
Comment
F.4.1 is almost impossible to comprehend in this form. Compare it with the reworded version in F.4 4.
F.4.2 is difficult to understand because it consists of two parts and a double negative phrase, do
not disqualify.” The reworded version in F.4.5 divides the question into two parts and eliminates the
use of the double negative.
,
Example F.4.3 uses two negative terms, “barely” and “unable.” F.4.6 avoids the confusion created
by their use.
F.5
COMPOUND QUESTIONS
6. Over $25,000
F.5.1
(Unsatisfactory; contrived)
Comment
Unable to walk
Barely able to walk
Able to walk fairly well
Able to walk normally
Do the answers qualify the candidate for
enrollment in the study?
4. $12,000 - $15,999
5. $16,000 - $24,999
F.3.1 contains unnecessary words that are not helpful, and perhaps even confusing, particularly since
the question simply refers to income and the “clarifying” remark following the question refers to total
family income. F.3.2 conveys the same meaning as F.3.1, but with fewer words.
It is easy to find examples of items with unnecessary words. An economy of words cannot be
achieved without a considerable investment of time and effort in the development, review, and testing
processes involved in constructing forms.
No
1
)
)
)
)
If Yes,
F.4.7
UNNECESSARY WORDS
Yes
) (
YES
Since your stroke are you (check one)
Were items 28 through 32 reviewed by a
nutritionist?
(
1
(Satisfactory; contrived from F.4.3)
F.4.6
(Satisfactory; contrived from F.4.2)
F.4.5
NO
The answers to items 28 thru 32 have been
reviewed by a nutritionist and they quali fy
the participant for enrollment.
(
Yes
) (
No
)
385
F.6 Comparative evaluations
384
F. Data items and forms illustrations
F.5.2
(Unsatisfactory; facsimile)
F.6
YEAR
MONTH
Why did you see a doctor?
00
01
02
03
I mean, what was wrong with you?
COMPARATIVE EVALUATIONS
F.6.1 Unstated standard
F.6.1.1 (Unsatisfactory; contrived)
(
ROUTINE CHECK-UP (GENERAL HEALTH)
ROUTINE CHECK-UP FOR PREGNANCY
JUST DIDN’T FEEL GOOD
RECEIVING TREATMENT FOR ILLNESS, SPECIFY ILLNESS:
05
(
Undefined standard
____-----------------
F.6.2.2
(
contrived from F.6.1-1)
) ( No )
(Unsatisfactory; contrived)
F.6.2.1
Do you have any major diseases?
RECEIVING TREATMENT FOR INJURY OR ACCIDENT?
SPECIFY WHAT WAS WRONG.
Yes
(Satisfactory;
Do you get^more exercise now than
did'
yOU d
’. J ’before
-3 “““ ,your illness?
Do you get more exercise now?
F.6.2
04
F6.1.2
Yes
) (
Yes
) (
No
)
(Better; contrived from F.6.2.1)
Do you have any major diseases,
such as cancer or heart disease?
)
No
(
Yes
) (
No
)
RECEIVING TREATMENT BECAUSE RESPONDENT WAS A VICTIM
OFCRIME.
SPECIFY.------------------- - ----------------------------------
F.6.2.3
(Satisfactory;
contrived from F.6.2.2)
following diseases?
Do you have any of the
(Check as many as apply):
06
F.5.3
OTHER, SPECIFY
(
(
(
(
(
(Satisfactory; contrived from F.5.1)
Were items 28 thru 32 reviewed by a
nutritionist?
(
Yes
) (
No
)
If Yes,
F.6.3
F.6.3.1
) Cancer
disease
) Heart
) Diabetes
) Arthritis
) Other (specify)
Positive versus negative standards
(
Yes
) (
No
(
(Satisfactory; contrived from F.5.2)
Why did you see a doctor?
00
01
02
03
F.5.5
■ j be met in order to check “Yes.” Technically, a “No”
.The question in F.5.1 requires two conditions to
either that the list of questions
No
(
Yes
(
)
F.6.3.4
) (
No
(Satisfactory;
Yes
) (
)
No
contrived from F.6.3.3)
Would you say you are in better health
than you were a year ago?
Would you say you are in worse health
than you were a year ago.
)
(
Yes
) ( No )
hems requiring a
to^be made. F.6.1.2 specifies
provldesa generti
F.6.3.1 and F.6.3.3).
“clarifying” question, as in example F.5.4.
) (
(Unsatisfactory; contrived)
ROUTINE CHECK-UP (GENERAL HEALTH)
ROUTINE CHECK-UP FOR PREGNANCY
IUST DIDN’T FEEL GOOD
RECEIVING TREATMENT FOR ILLNESS, SPECIFY ILLNESS:
Comment
Yes
)
F.6.3.3
(Satisfactory; contrived from F.6.3.1)
Are you taller than your mother?
Are you shorter than your mother?
the candidate for
Do the answers qualify
n
enrollment in the study?
F.5.4
F.6.3.2
(Unsatisfactory; contrived)
by^major" disease. F.6.2.3 .s even more expired rn
386
F Data items and forms illustrations
F.9 Time references 387
F.7 INVERTED MEANING OF A YES
REPLY
(Unsatisfactory; contrived)
F.7.1
F9.1.3
Patient uses (answer a through c):
(
(
(
Insulin
Oral hypoglycemic agents
Digitalis
a.
b.
c.
Were you taking any prescription drugs
when you had your last MI?
(Satisfactory; contrived from F.7.1)
F.7.2
Patient does not use (answer
a through c):
Time frame specified in terms of a defined event (Satisfactory; contrived)
Yes
) (
) (
) (
(
(
Yes
No
)
)
)
a.
b.
c.
(
(
(
Insulin
Oral hypoglycemic agents
Digital is
Yes
) (
) (
) (
No
)
)
)
F.9.2
F9.2.1
No
>2
Time interval references
Time interval defined by two events (Satisfactory; facsimile)
Comment
F.7.3
110.
F.7.1 requires the respondent to check “Yes” if the patient does not use an indicated drug. The
question is less confusing if the use of a drug requires a “Yes” reply, as in F.7.2.
Did your doctor advise you not to take the pill or IUD
between the start of your periods and your first pregnancy?
YES, IUD
F.8 PRESENCE VERSUS ABSENCE OF A
CONDITION
F.8.1
YES, PILL
F.8.2
(Unsatisfactory; contrived)
(
F.8.3
Yes
) (
No
What were the reasons?
(Satisfactory; contrived from F.8.1)
NO
Has a doctor ever told you that you
have heart disease?
Are you free of heart disease?
ASK:
)
(
Yes
) (
No
F 9.2.2
)
UNKNOWN
Time interval defined as periodfrom last study examination to present (Satisfactory; photo
reproduction)
8) Since the patient’s last completed
follow-up visit, has he had any of the
following (answer each question):
Comment
F.8.1 requires an affirmative response to indicate the absence of a condition. In a sense, the question is
impossible to answer since there is no way to know if one is free of heart disease.
Item F.8.2 is stated in positive terms and avoids the problem of F.8.1 by relying on an operational
definition of heart disease.
Y«
( ) (
B) An obvious stroke?
• ( ) ( )«
C) Weakness or paralysis of any part
F.9
TIME REFERENCES
F.9.1
Time point references
F.9.1.1
9.
F.0.2.3
Present time point (Satisfactory; facsimile)
Timeframe defined in terms of a specified calendar date (Satisfactory; photo reproduction)
Approximately what was your weight in
?
pounds
(2 YEARS
PRIOR TO
REF. DATE)
10.
Time interval defined as entire "study period" (Satisfactory; facsimile)
List all in-patient admissions during study period (for any diagnosis, to this hospital only).
( No
r;:cF.9.1.2
( ) (
(AAW) Are you presently taking any
drugs prescribed by a physician?
Yes
9.
)«’
of his body? --- --------------------
(
ii
N.
A) Cardiac asthma?
Approximately what was your weight in
?
REFERENCE DATE
POUNDS
(
KG)
(
KG)
UNKNOWN
UNKNOWN dj
Identify with a check
the admission selected
for review sample____
1st_________________
2nd_________________
3rd_________________
4th_________________
Sth_________________
6th
____ Admission
Month i Day i Year
Date of
_______Discharge
Month i Day t Year
_________ I__________ I_________
i
i
t
I
l
l
If Study Admission is the First in the Study Period, give most Recent Date of Hospitalization
(this hospital only) before Study Period:
None
F
a^-
V^r
388
F.ll Leading questions
F. Data items and forms illustrations
F 9.2.4
F.10.2
Time interval definedfrom some calendar time to present (Satisfactory; photo reproduction)
Mixed attitude expression (Unsatisfactory; facsimile)
(READ CHOICES)
How often did you feel
FHY3ICJAN OR CUNIC VISITS SINCE 1W0
18.
Data
yrJ
ONCE
(1 TIME)
Pleased about having
accomplished some
thing?
0
1
2
3
44.
Proud because someone
complimented you on
something you had done?
0
1
2
3
45.
Depressed or very
unhappy?
0
1
2
3
46.
Bored?
0
1
2
3
03:43
Spedahy
1
F9.2.5
18.
Time interval from present to recall limit (Satisfactory; photo reproduction)
47.
Particularly excited
or interested in
something?
0
1
2
3
48.
Downcast and
dejected?
0
1
2
3
Could you plmm toll m a faw th1n«s about each tlua you worn x-rayod? Start with the first x-ray
you ramubor, and than tall m about later ones. (ASK A-F)
(A) What part of the” (I) In what
head aat x-royod?
year ms
skull, tonsils,
thh
slmaos, etc.)
dooaT
(0) Why ms the (E) About
x-ray taken? how mny
(C) In what hospital ®r doctor's
office ms this dora?
NNC
film ware
taken?
AOOEESS
F.10.3
tz
F.9.3
SEVERAL
TIMES
OFTEN
(2-3 TIMES) (3+ TIMES)
NOT AT
ALL
(0)
Pkm list all phyiidan and/or dink: visits since 1950 other than routlna amptoymant axarrn.
Physician and/or dink
(Name & addrm)
389
Comment
The difficulty with F. 10.1 arises from the mixture of disease and nondisease states. This problem could
be corrected by replacing “euglycemic” with “diabetic” in the check list.
F.10.2 includes a mix of positive and negative states of mind. The first two items are phrased in
positive terms. The next two are stated in negative terms. Intermixing of this sort is ill-advised unless
there are good reasons for doing so (e.g., to keep a respondent from automatically checking the same
category for all items without thinking).
Comment
The examples in F.9.1 all refer to some point in time—the present, a calendar date, or the date on
which a defined event occurred. The examples in F.9.2 are for defined time intervals. The interval in
F.9.2.1 is defined by two events—the start of menstruation and first pregnancy. A common frame of
reference for follow-up studies is illustrated in F.9.2.2 where the respondent is required to cover the
time period from the last follow-up visit to the present visit. The time frame in F.9.2.3 is the entire
“study period.” This mode of reference is useful only when it is clear what is meant by the phrase.
Example F.9.2.4 covers a time interval from 1950 to the time of the interview. The interval in F.9.2.5 is
defined by the limits of a respondent’s recall.
F.ll
LEADING QUESTIONS
F.ll.l
(Leading; contrived)
F.11.2 (Not leading; contrived)
Have you ever had allergies
to ragweed?
Do you still have allergies
to ragweed?
(
Yes
) (
No
(
)
Yes
) (
No
)
If Yes,
F.10
F.10.1
DIRECTION OF RESPONSE
Do you still have them?
Mixed disease categorization (Unsatisfactory; contrived)
(
Is the patient (answer each question):
Hypertenslve?
Euglycemic?
Obese?
Hyperlipidemic?
(
(
(
(
Yes
)
)
)
)
No
(
(
(
)
)
)
F.11.3
Yes
) (
No
)
Comment
F. 11.1 not only leads the patient but assumes he has had an allergy to ragweed in the past. As written,
there is no way to answer the question if the patient has never had ragweed allergies. F.l 1.2 avoids the
problem by beginning with a general question that determines whether the patient has ever had
ragweed allergies before inquiring about current allergies.
Ip
I
F.12 Vertical versus horizontal response lists
390
F. Data items and forms illustrations
F.12
VERTICAL VERSUS HORIZONTAL RESPONSE LISTS
F.12.2
FI2.2.1
Horizontal lists
F.12.1
F. 12.1.1
(check) |
| WM; |
~| WF;
NWM; ,
| NWF;
OM;
OF
(
(
(
(
(
(Unsatisfactory; photo reproduction)
11. How many times a week do you eat each of the following foods?
Fish
; Meat
; Poultry--- ; Eggs---------- ; Cheese------Breads & cereals
; Potatoes---------- ; Rice ——; Vegetables .
Fruits and fruit juices----------; Sweet desserts---------- ; Candy---------
Butter or margarine
; Green salads —
F. 12.2.2
F. 12.1.3
16.
F.12.1.4
22.
(Unsatisfactory; photo reproduction)
(Unsatisfactory;photo reproduction)
If yes, please complete the table below: (Start with most rt
(Satisfactory; contrived from F.12.1.4)
F. 12.2.3
(
iSil
F.12.1.8
Yes
) (
No
)
$
36 thru 45,999
46 thru 55,999
55,000 and over
declines to answer
F. 12.2.4
Never
Insomnia
Sense of exhaustion (except after exercise)
Periods of alternating gloom and cheerfulness ...
Periods of being particularly self-conscious
None
1-
Erythema
Edema
3.
Vesicles/oozing
4-
Buila/denudation
5.
Delayed flare
(Satisfactory; contrived)
Does the patient have any of the following diseases?
(Check all that apply.)
(Satisfactory; photo reproduction)
Sensation of heart beating (except after exercise)
0.
IZI 2-
(Unsatisfactory; photo reproduction)
16. How often do you
experience:
16 thru 20,999
21 thru 25,999
26 thru 35,999
(Satisfactory; contrived from Fl 2.1.3)
|“|
14. How often do you use the following types of medicine?
Never
Occasionally
Frequently
Aspirin, Bufferin ..
Never
Occasionally
Frequently
Vitamin pills
Never
Occasionally
Frequently
Sleeping pills
Never
Occasionally
Frequently
Tranquilizers
Never
Occasionally
Frequently
Laxatives
Never
Occasionally
Frequently
Anti-acid medicine
I’M
< $5,000
IV,
5 thru 10,999
11 thru 15,999
1
Check the reactions at test sight 7-14 days after application.
(Satisfactory; contrived)
Are you taller than your mother?
F.12.1.7
rj5
(
(
(
(
(
(
(
(
(
(
RADIATION THERAPY: Have you ever had any treatments with radium, cobalt 60, cobalt bomb radio is
| | NO
| | YES
p1] DON'T KNOW
If yes, please complete the table below: (Start wi
F.12.1.6
White
Black
Hispanic
Oriental
Other (Specify)
(Satisfactory; facsimile)
RADIATION THERAPY: Have you ever had any treatments with radium, cobalt 60, cobalt bomb radic
F. 12.1.5
£p
(AAW) What is the combined income of all members of
your •household?
■ / ■? (Use list below to categorize
candidates response). (Use HPT flashcard 03)
Check the reactions at test sight 7-14 days after application.
Q 2. Edema | | 3. Vesicles/oozing
I I 0. None
I 1 1. Erythema
Qs. Delayed flare
r"| 4. Bulla/denudacion
P~j NO (2J YES P~| DON'T KNOW
'I'
(Satisfactory; facsimile)
(AAW) How would you characterize your ethnic
origin? (Use HPT flashcard 01)
(Unsatisfactory: photo reproduction)
Sex and Race:
F.12.1.2
Vertical lists
Occasionally
Frequently
(
(
(
(
(
F
J2
F
r
Diabetes
Cancer
Hypertension
Heart disease
None of the above
391
392
F13 Unit specifications 393
F. Data items and forms illustrations
F.12.3
Fl3.2.2 (Fair; photo reproduction)
Comment
-If a check mark was placed after item 20-R,
answer items z\ through E below relative to
the episode which resulted in a diagnosis of
myocardial infarction:
A list that is horizontally arrayed requires less space than one that is vertically arrayed (e.g., compare
F. 12.1.3 and F.12.2.3). However, vertical layouts are generally less confusing than horizontal layouts
to use (compare items in F.12.2 with those in F.12.1). The main difficulty with the examples in F.12.1
stems from the confusion the respondent is likely to have in locating the proper check space. Items
F. 12.1.2 and F. 12.1.4 are especially defective in this regard. The uniformity of spacing makes it
difficult for the respondent to decide whether the check space for the response is in front of or behind
the designated reply.
A horizontal layout is acceptable for short lists. However, even in such instances it is important to
use a layout, as in F. 12.1.5 and F. 12.1.6, that makes it easy to associate a response with the
appropriate check space. The association is not obvious for F. 12.1.1 and F. 12.1.4 and only moderately
so for F.12.1.7.
The amount of space provided for making a check in a vertical list should be adequate to avoid the
confusion that can result if the check is not registered squarely in the center of the check space. For
example, the vertical separations in F.12.2.3 are better than those in F. 12.1.7 and F. 12.1.8.
A) Highest SCOT recorded (state units): .. ------ —55
not done,
done, check
check here:
here:------------------------ (STOP)
IfIf not
B) Highest IJJH recorded (state units): — _______JG
( )
If not done, check here: —
C) Highest sedimentation
(mm/hr):-------
S7
------- ( )
(Satisfactory; contrived from F 13.2.1)
Laboratory Tests:
UNIT SPECIFICATIONS
recorded
If not done, check here:------
Fl 3.2.3
F.13
rare
YES
NO
NR
DATE
DAY
RESULTS OF TEST
UNITS
(if menstruating
day of cycle)
Time units
F.13.1
Time and date of BP measurement
a. Date of patient's next appointment?
Mo
Serum prolactin
F.13.1.2 (Satisfactory; facsimile)
(Unsatisfactory; contrived)
F. 13.1.1
Tay
AM
PM
a. Time of day
b. Time of patient's next appointment?
LH
b. Date:
ITo
Tay
Estradiol
I
I !.. 1
F.13.2
Laboratory units
F. 13.2.1
(Unsatisfactory; photo reproduction)
14.
Laboratory Tests:
Serna prolectin
■
I
LH
Estradiol
YES
NO
NR
DAY
(if menstruating
day of cycle)
DATE
RESULTS OF TEST
F.13.3
F. 13.3.1
Height-weight units
(Unsatisfactory; photo reproduction)
Fl3.3.2
(Satisfactory; facsimile)
Height:
nrEJ
13. Ht (shoes off):
NR | I
14.
Weight:
Wt (outdoor garments and
shoes off):
inches
“IS?.-
F.15 Calculation items
394
F.14.5
F.13.4
Solid line with no indication of required precision (Unsatisfactory; facsimile)
Unconventional units (Unsatisfactory; photo reproduction)
1) Results for the FIRST SO minute collection period.
a) Scrum creatinine determination,. The blood sample for this determination
should be taken at the start of
c2 the FIRST 90 minute collection period.
o-.
Carry the accuracy of the determinations
Record the results in mg./ml.
four (4) places beyond the decimal point, e.g., 0.0185 mg./ml. Record the
readings from both aliquots.
mg./ml.
Aliquot 2
mg./ml.
"
‘
Record
b) Urine creatinine concentration for the FIRST collection
period,
the determinations
three (3)
results in mg./ml. Carry the accuracy of -----places beyond the decimal point, e.g., 0.155 mg/ml. Record the readings
from both aliquots.
mg./ml.
Aliquot 1.
mg./ml.
Aliquot 2
inches
lbs.
Comment
• ■ precision
‘
’
Ji use of a series of dashed lines and decimal points. This
F.14.1 indicates the required
through
format is preferred to those illustrated in FJ4.2 and F.14.3 for reasons indicated in Section 12.6.8.2.
The main problem with F. 14.1 is that it requires more precision than is attainable with ordinary body
height and weight measurements.
A format involving use of a solid line with no indication of required precision, such as illustrated in
F.14.5, should be avoided if possible. The instructions for the item should indicate the desired
precision if this format is used, as in F.14.4.
F.15
CALCULATION ITEMS
F.15.1
A
URINE
Height
Weight
F.14.6
Aliquot 1
A
SERUM
F.13.5 Comment
F.13.1.1 does not specify how time is to be recorded—on a 24-hour basis or on a 12-hour basis. If the
latter, there will be ambiguity in the information supplied unless the respondent indicates AM or PM.
F.13.1.2 avoids this difficulty by requiring the respondent to indicate AM or PM.
The unit of measurement should be part of the item, as in F. 13.3.2 and F. 13.4, when measurements
are to be made using a specified unit. The item should provide space for recording the unit oi
measurement when it is not specified (e.g., as in F. 13.2.3). The format for F. 13.2.3 is better in this
regard than for F.13.2.2. Items F.13.2.1 and F.13.3.1 are defective because they do not provide an
indication of the unit of measurement.
The illustration in F.13.4 is taken from a form used in the University Group Diabetes Program
(UGDP). Creatinine determinations are typically recorded in lOOmg/ml. The UGDP form required
recordings in mg/ml. The requirement resulted in a large number of recording errors.
.. I
395
F. Data items and forms illustrations
Bad examples
Fl5.1.1
Blood pressure measurement (Facsimile)
INITIAL MEASUREMENT
(Record 3 measurements)
□□□ □□□ >■ —'—
SBP
DBF
2
/
(AVERAGE OF 3 READINGS) ' -----------------3-
SECOND MEASUREMENT
/
(Record 3 measurements)
□□□ □□□ i-__ L__
SBP
DBP
2
/
(AVERAGE OF 3 READINGS) * --------3/
F. 15.1.2
Blood pressure measurement (Photo reproduction)
7. Blood Pressure Jteasurerents:
F.14
F.14.1
PRECISION SPECIFICATIONS
Dashed lines with decimal points (Satis
Boxes with decimal points (Satisfac
F.14.2
tory; contrived)
factory; facsimile)
Height:
inches
Weight:
pounds
Height:
□II
inches
F.14.3
Hatched line with indication of required
precision (Satisfactory; contrived)
F.14.4
Seated:
Systolic
1st R-Z
Zero
^8)
Solid line with written indication of
required precision (Satisfactory; fac
(13)
Ul)
r- 1st Corrected:
2nd R-Z
Zero
- 2nd Corrected:
Weight:
Sth Phase
(Disappearance)
Diastolic
4th Phase
Diastolic
= Sun of 1st & 2nd
?14)
(12)
(10
rm Hg
nro Hg
run Hg
rrm Hg
rm Hg
tnn Hg
/
mm Hg
7
simile)
Blood pressure (record to nearest even number)
Systolic blood pressure
Diastolic
_____ mHg
Systolic
mmHg
mu Eg
Average 1st & 2nd:
SBPOR-Z
nrnHg
Note:
sth DE? R-Z
These are the values to check for eligibility,
goal, or escape, as is apprrnriate.
Fl5 Calculation items
396
397
F. Data items and forms illustrations
F15.1.3
Cholesterol readings (Photo reproduction)
F.15.2
Both 1" aliquots
taken (
)
5) Serum cholesterol (recorded in mg. per cent). Determina
tion made at the University of Minnesota on the basis of two
.1 mL aliquots taken at both the initial and final examina
tions. Wlien aliquots have been taken place a check in the
appropriate parentheses at the right and record the date tak
en below.
Initial taken----------■■ -------------- -------------■n.
FU.
Final taken
Tn
Mat
F15.1.4
Both “F* aliquots
taken (
)
||
| - Avg. -> J
ZT
I-F-—I-
Overall -*■
i
I
Good examples
Fl 5.2.1
Blood pressure measurement (Facsimile)
RZ RP measurements
__ Lr
RP in nwHg
$RP
TO?
Reading
e- DiS.
1st RZ
«- Avg.
a. Reading
Haserved for Recording Reading
b. Zero value
c. a-b
Oscillometric readings (Photo reproduction)
2nd RZ
1) Record the oscillometric readings for the RIGHT leg, for the two sites indicated.
d. Reading
a) Two inches below the lower margin of the patella.
jnm.
e. Zero value
b) Two inches above the apex of the internal malleolus
jnm.
f. d-e
Avg RZ
2) Record the oscillometric readings for the LEFT leg, for the two sites indicated.
a) Two inches below the lower margin of the patella
g. Sum (c + f)
jnm.
h. Avg (g ♦ 2)
jnm.
b) Two inches above the apex of the internal malleolus--------
F. 15.2.2
3) Total of all four readings on PRESENT examination (la+lb+2a+2b)-----
jnm.
4) Total of all four readings on PREVIOUS examination (la+lb+2a+2b).
jnm.
F. 15. /.5
Ji
Did you
perform
this
activity?
_L _L
c 49
48
50
51
_L
J_
54
53
52
C
55
C
_L
_L
60
59
58
C
61
__L
57
56
L_ L_
62
63
For Clinic Personnel Use Only
Month of Activity
Average
number
of times
per
month
1
Sum of readings 2 and 3
F.15.2.3
Blood pressure measurement (Photo reproduction)
SECTION A: walking and Miscellaneous
5th Phase
Diastolic
Systolic
Walking tor Pleasure and/or to Work
rm
First
Second
Back Packing
Mountain Climbing
47
C
Reading 3
No ' Yh Jan Feb Mar Apr May.Jun* July^j Aug!Sept| Oct Nov Dec
(2) I (3)
Cross Country Hiking
46
jnm.
Activity assessment (Photo reproduction)
Using Stairs When Elevator is Available
_1 _L
Reading 1
Item 4
ACTIVITY (1 >
Diastolic (Sth Phase)
Systolic
C
5) Item 4 — Item 3 —
6) Item 5
Blood pressure measurement (Photo reproduction)
Third
Sum ol
2nd & 3rd
Average of
2nd & 3rd
I
I
I
I
I
I
I
I
I
I
| |
(29)_
| 1(30)
mrror
I
rm (3t) rm <34, II
I
I
I
I
SB?
2.3
(55^
2.3
Diastolic
Systolic
I I _[
|||| (32)_
^l(37)_ | | | 1 (40)
mmHg
i----------- •_
I
I
I
I
I
I
I
I
I
I | I I (38) | Cl ] (41) • mmHg
Illi (39)
rm
(42)
I
| mm Hg
I
I
I
Auto
2.3
Auto D0P 2.3
I
I
I
USKtSlS^i^
398
El7 Age and birthdate items
F. Data items and forms illustrations
F. 15.2.4
28.
Comment
F.16.3
Pulse (Photo reproduction)
See Section 12.5.8 for discussion of STOP and SKIP items.
Standard Pulse
1(28) beats/30 secs.
I 1
beats/minute
x2 =
F.17
AGE AND BIRTHDATE ITEMS
F.17.1
F.15.3 Comment
(Satisfactory; facsimile)
9. (AAW) What is the month, day
and year of your birth?
F. 15.1.1 is bad because there is no space for calculating the average systolic and diastolic blood
pressures. F. 15.1.2 has a confusing layout as well as a shortage of work space. The three blood
pressure examples in Section F. 15.2 avoid the problems of F. 15.1.1 and F. 15.1.2.
F. 15.1.3 is defective because of the absence of any work space for performing the additions,
subtractions, and divisions required for the item. As written, the operation must be done on scratch
paper. A vertical layout, such as displayed in F. 15.2.1, would be required to enable the respondent to
use the form to make and record intermediate calculations needed to complete the item.
The layout in F. 15.1.4 could be improved by providing work space for the required additions and
by having item 4 appear above item 3 to facilitate the subtraction required in item 5.
The layout of F. 15.1.5 is not well suited for the averaging process, since users would almost
certainly have to copy the data from the item onto a separate worksheet to perform the required
additions and divisions.
F.16
F.16.1.1
11. (0A) Does the birthdate fall within
the interval defined below?
(
Eligibility STOP item (Satisfactory;
facsimile)
F.16.1.2
Temporary STOP item (Satisfactory;
facsimile)
),
1
(st°P )_
No 2
(
shoes off):
— 1Ks—
15. Q.I. « Wt/Ht2
Candidate must be 25 or older but
49 or less at the time of registration
to be eligible for enrollment into the
HPT. Permissible birthdates for visits
completed in 1982 are (use month and
day recorded in item 8):
24. Serum specimen for central labor
atory determinations collected?
TncFes
14. Wt (outdoor garments and
(stop )*
)
Yes
1
No 2
* Remove by drawing required blood.
(use HPT Chart 11)
0.
(
1
“Uay
Mo
“Uay
-_2_2
Yr
No
7
5
(stop )*
)
Yes
Mo
thru
_____ Ibs/in?
16. Q.I. <0.0500 Ibs/in2
:-
Yes
STOP items
13. Ht (shoes off):
F.16.2
“Yr-
“Uay
10. (AAW) What was your age at your
last birthday?
INSTRUCTION ITEMS
F.16.1
'I1
“fTo
2
F.17.2
See Section 12.5.10.
SKIP item (Satisfactory; facsimile)
6. (AAW) Have you ever had your blood
pressure measured before your
first visit here?
(
If No, go to item 7.
If Yes, answer a and b
Comment
Yes
) (
’
No
)2
Yr
399
F19 Full-page versus two-column layout 401
400
F Data items and forms illustrations
F.18
REMINDER AND DOCUMENTATION ITEMS
F.19
F.18.1
F18.1.1
Reminder items
Full-page layout
F.19.1.1
F18.1.2
Time window check (Facsimile)
Procedure reminder (Facsimile)
24. Serum specimen for central labor
atory determinations collected?
5. Time window check
a. Date of BL 1 (item 20b, Form 01CP)
“Ro
FULL-PAGE VERSUS TWO-COLUMN LAYOUT
F.19.1
~Day
(
), (st°P )*
No 2
Yes 1
^?r—
MONTH
3. Sex:
(
Yes
’
Male
C Female
Z Hispanic
Z Black
4. Race: Z White
0123456789
io
12
ii
6. Have you smoked at least 100 cigarettes (five packs) in your entire life?
Asian
C Other
u
13
15
16*
IF YES^|
Yes
Z No
16
a) How old were you when you first started regular smoking?years old
b) On the average or the entire time you smoked, how manv agarettes did vou smoke per dav’
agarettes per dav
c) Did you ever stop smoking agarettes fora year or more, and then start again’ Z No Z Yes. stopped fot _years
d) Do you smoke agarettes now?
IF NO. how old were you when you stopped?years old
No
IF YES, how many agarettes a day do you smoke now’
agarettes
Yes
No 2
e) Do you (or did you) inhale?
No. rarely
Yes. usually
0 Do you (or did you) smoke filter agarettes
No. rarely
Yes. usually
g) Did you ever smoke agarettes which you rolled yoursetf?
□No
Yes. tor
Documentation items (Facsimile)
32. Form reviewed by:
7.
a. Name:
Have you ever smoked a pipe regularly ?
No
Yes
.years
IF YES
a) How old were you when you started regular pipe smoking?years old
b) On the average of the entire time you smoked, how much pipe tobacco did you smoke?.
b. HPT Cer. No.
c) Did you ever stop smoking a pipe for a year or more and then start again?
c. Date:
Z
Z No
5. Please circle the highest grade in school you have completed:
* Candidate should be rescheduled to
remove this stop condition.
F.18.2
2. Were you bom in the U.S.?
»t*«
American Indian
(stop )*
)
/
OAV
IF YES. in what state?
* Remove by drawing required blood.
b. Is this visit at least 7 days after
the date in item a.
(Unsatisfactory; photo reproduction)
1. When were vou bom? ______
“Ro
d) Do you smoke a pipe now?
IF NO. how old were you when you stopped?
No
IF YES. how much tobacco do you smoke now’.
Yes
“Hay
jjipefubper_______
No Z Yes. stopped tor
years
I
.years old
__pipefuls
33. Data entered by:
8.
a. Name:
“Ro
Comment
See Section 12.5.13.
i:-
Yes
IF YES^|
c) Did you ever stop smoking agars fora year or more, and then start again?
c. Date:
F.18.3
No
a) How old were you when you started regular agar smoking?
years old
b) On the average of the entire time you smoked, how many agars did vou smoke?.
b. HPT Cer. No.
I.. I
Have you ever smoked cigars regularly?
“Day
-7r~
d) Do you smoke agars now?
No IF NO. how old were vou when you stopped’years old
□ Yes IF YES. how manv agars do you smoke per day or per week?
No
.agars pet
Z Yes. stopped fot
.agars per^ ,,
years
F.19 Fuil-page versus two-column layout
402
F. Data items and forms illustrations
F.19.2
F.19.1.2 (Satisfactory; photo reproduction)
F. 19.2.1
Two-column layout
Right-hand justification for response boxes (Satisfactory: photo reproduction)
PART HI: Clinical Findings.
Na
Right leg
1) Is the femoralis artery palpable?
Left leg
( ) ( ) ( )
( ) ( ) ( )
Na
2) Is the dorsalis pedis artery palpable?
.....
Right foot
Left foot
pwaibla
to
tert
Not
poaaiUa
to
tort
( ) ( ) ( )
( ) ( ) ( )
Item 6-1 continued:
ii) Any evidence of ihock?
Item 6 continued:
C) How much exertion would it typically
take to precipitate such an episode (check
only one)?
Walking at les* than ordinary pace
Walking at an ordinary pace
Walking hurriedly or up hill,.. or climbing stain
Not related to exertion .
iii) Arrhythmii?
<*
‘J*
iv) Leucocytoeu?
If YES, what was the
highest recorded value
(cell*/mm’)? —™_—
D) Can excitement, emotion, or meals
precipitate sudi an episode?-------
v) Elevated sedimentation
rate? ...——
——
E) Does rert typically relieve auch an epuode?
PART IV: Oscillometric Examination. The upper margin of the cuff for the knee measurement should be placed
two (2) inches below the lower margin of the patella. The lower margin of the cuff for the ankle reading should
be placed two (2) inches above the apex of the internal malleolus. The patient should be reclining and the knee
undexed with the ankle in a neutral position. If a particular reading cannot be taken because the site is missing or
the patient has a lesion in the area of the site of the measurement, please indicate this by writing "Not Possible”
in the space reserved for recording the measurement Record the results in mm.
1) Record the oscillometric readings for the RIGHT leg, for the two sites indicated.
a) Two inches below the lower margin of the patella.-------------
mm.
b) Two inches above the apex of the internal malleolus-----------
.mm.
Item
If YES, what was the
(*)
mm.
)
( )*
No
I) Did the patient get medical atten
tion in connection with any episode
of pain, aching, etc., during this
period?
—.-—™.~—~. )
If NO, proceed to item 6-J.
If YES,
such medi1!
1 to, pleaae
picaac give
gnc the
uic place
piew where
«
cal information may be found:
p,la4*C
1
I
1
”!i
Si
'
I
J
highest recorded value
(state units)?---------- -----------------------vii) Abnormal LDH? ------ P )
C) What ha* been the longert duration of such
an episode?
(■)”
I-ess than 10 minute* .......
10 to SO minutes
—
<’)
More than 30 minute*
H) Have any of the episode* been *uch
that rest or nitroglycerin did NOT
bring relief in the typical manner?
a) Two inches below the lower margin of the patella
If rES. wh»i «u the
highest recorded ealue
(mm/hr)?
vi) Abnormal SCOT?
F) Doe* nitroglycerin typically relieve such an
episode?
Not at all
After more than 10 minute*
—
iu less than 10 minute*
Nitroglycerin not used
(*)
2) Record the oscillometric readings for the LEFT leg, for the two sites indicated.
b) Two inches above the apex of the internal malleolus
Not at all
..... —• ( ’ I*7
After more than 10 minute*
........... ( * )
In lew than 10 minute*
— ( *)
Rest not u*ed ----------- ---------- —----- ( 4 )
Then avail yourself of this information and an
swer items i through viii below. (H medical
attention was obtained on more than one occa
sion. answer the questions in connection with the
most serious of the episodes.)
(^)
(£)*
Dw
If YES, what wa* the
highest recorded value
(state units)? —~viii) ECG evidence of a new myocardial
infarction?
ECC not done
Negative ----------
Suggestive ----------Definite--------- —
— ( ,)»
El’!
J) Did any of the episodes
episode* sir
wnce
the last completed followfollow-up
visit result in a diagnosis of:
i) Mvocardial infarction? ( ’ )
ii) Acute coronary insuffi
ciency? -- ----------------- ( 3 )
iii) Angina pectoris?
’ )
(*) (*)*
SB
7) Ha* the patient required
rnitroglycerin
since his last completed
follow-up visit? (^)
compl
(IU)«
403
F.20 Layout for skip items 405
404
F. Data items and forms illustrations
F 19.2.2
12.
Left-hand justification for response boxes (Satisfactory; photo reproduction)
[1 ]
£ 1 minute
[2 ]
1-4 talnutM
(Continued)
Wo
[1 ] [2 ]
Inadequate
Inf ortsa clou
(3 ]
M. Other noa-neoplastlc renal
pathology
(3 ]
5-29 nlnutaa
30-59 minutes
[1 ] [2 ]
[3 )
h ]
N. Ma 11gnant
neoplasa
(5 ]
1-24 hours
[X 1 [2 J
[3 ]
O. Benign ueoplasn
[S 1
> 24 hours
[1 1 (2 1
(3 ]
[7 ]
Unknown
P. Recent acuta
myocardial
Infarction
[a ]
Death witnessed
(1 J tz ]
(3 ]
Q. Old myocardial
infarction or
scar
PAR" V:
13.
13.
If death was not witnessed, what is the
latient
estimated tine between when the par
’
waa last known to be alive and thei death?
Pathology Noted on Aucopay
[X ] (2 1
Uaa an autopsy performed?
[X 1
(2 1
TM
[1 J [2 1
(3 ]
Ho
tl 1 (2 ]
Were therm findings of the
following? (Check each tc«)
Yes
(3 ]
Ho
(1 ] [2 1
(1 1 (2 ]
[X 1 (2 1
Inadequate
Inforaatioa
(3 ]
(3 ]
(3 ]
(1 ) (2 ]
A. Esophagitis
*9
B. Gastritis
!SO
C. Superficial
! SI
gastric eroeioa(s)
(1 ] (2 ]
(3 )
[3 ]
(3 ]
F.20
LAYOUT FOR SKIP ITEMS
F.20.1
Arrows and pointers
F.20.1.1
(Unsatisfactory; photo reproduction)
1 Oyet
2
no
2. 0«1 vuu peisonally leques. the diet information from your physician?
3
no
10
4
For each item below indicate whether it was for that reason that the physician asked you to
follow the special diet.
a
Diabetes
1
vet
2
no
b
Overweight
1
yet
2
no
c.
H<gh Blood Pressure
1
yet
2
no
1 O yet
2
no
d. High Blood Fat or Cholesterol
e
Food Allergy
1
yet
2
no
S. Myocardial
aneuxysn
f.
Ulcer
1
yet
2
no
g.
Other
1
yet
2
no
U. Recent coronary
occlusion by
hemorrhage
into plaque
V*» 2
CC USE
R. Rupture of
myocardiim
T. Recent coronary
occlusion by
thrombosis or
emb o Lisa
’
P:.,.rsr summarize the food changes your physician advised you to make.
Specify___________________
5. Were you given printed instructions describing the special diet?
1
¥•» 2
no
6. Was the special diet explained to you by the physician or his staff?
t □»«
7. Check the following people who explained the diet to you.
a.
2 Dno
1 O v*» 2
no
1 O yea 2 O no
Phytician
b. Nurse
c.
V. Pulmonary
eabollSB
Dietitian or Nutritionist
d. Other Staff
1 O yet
2
no
1
2
no
yet
Specify
8. How well did you understand the diet changes the physician advised you to make? (Check one)
F.19.3 Comment
The layout for F. 19.1.1 is confusing and cluttered. The lack of a standard location for check spaces
adds to the confusion. The layout in F. 19.1.2 is cleaner. It requires more page space but makes the
response positions of the items easier to locate than in F. 19.1.1.
The two-column layout in the examples shown in F.19.2 makes more efficient use of page space
than is the case in F. 19.1.2. The examples in F.19.2 differ with regard to location of check positions;
either arrangement is acceptable.
1
Very well I understood
what changes to make
2
Fairly well. I understood
some of the changes required
but had further questions
3
Not very well. I didn't know
what changes to make
9 . Have you started making the diet changes the physician advised you to follow?
1 Oves
10. Approximately how long has it been since you started making these
diet changes?
2
no
1
lets than
one month
4 O seven -nine
months
11
2 C one-three
month!
3
four-six
montht
ten-twelve
montht
6
more than
twelve montht
5
In general, how closely have you been following this diet during
the past year?
1 C have changed eating
habits consistent with
diet .md very rarely
go off diet
2
follow diet
mott of the
time
3 O have not been
able to ttkx to
the diet contitrently
Continue with quemon 12.
1
F20 Layout for skip items
F. Data items and forms illustrations
406
F20.I.4
F.20.1.2 (Unsatisfactory: photo reproduction)
54.2.
407
(Satisfactory; photo reproduction)
Did you work at ■ job or business at any time in the past three months?
a Date ot Death
NO
1 [
54.3.
2.
|
YES
8.
If the
shipyard
worker is
DECEASED
~T
Full time or Part-time?
1.
2.
F-T
8.
P-T
Year
Day
Mo
NA
b Place ot Death
State
City
c Cemetery
State
City
Name
d Cause of Death
NA
* I (YELLOW FORM) and answer the questions ABOUT THE DECEASED
Please
I
54.4
Did you work at any time last week or the week before not counting work around the house?
[
2.
NO
YES
8.
NA
F.20.2
SKIP or GO TO items
F20.2.1
(Satisfactory; facsimile)
9. (AAW) Are you presently taking any
drugs prescribed by a physician?
I
(
54.5
Even though you did not work during that time, do you have
a job or business?
a
YES
6
o
2.
'i'
No 2
If Yes, answer items a thru d
8.
NA
I
F.20.2.2 (Satisfactory; photo reproduction)
o
o
3
6
o
o
1
If No, go to Item 10, Page 4
IO
NO
(stop )
)
Yes
11.
Have you ever had any surgery or operations, even minor ones such as tonsillectoaty?
Have you ever had any surgery
SKIP TO Q. 13
NO
ASK Q. 12
UNKNOWN|
I SKIP TO Q. 13
0
IF YES, ASK:
12.
F.20.13
7.
(Satisfactory; photo reproduction)
Have you ever smoked a pipe regularly?
No
Yes
IF YES
Could you tell me:_______
(A) What type of surgery?
__ ________________________________
IF YES, ASK B, C:
(B)TIow old were you
at the time?
YES
NO
UNK
(C) Were you
hospitalized?
YES
NO
UNK
a) How old were you when you started regular pipe smoking?-------------- years old
Tonsillectomy
b) On the average of the entire time you smoked, how much pipe tobacco did you smoke?------------- pipefuls per ^
c) Did you ever stop smoking a pipe for a year or more and then start again?
d) Do you smoke a pipe now?
IF NO, how old were you when you slopped?—
No
IF YES, how much tobacco do you smoke now?.
Yes
I
8.
Have you ever smoked dgari n^ulariy?
I
No
Yes
No
Yes, stopped for
.years old
pipefuls per
years
Appendectomy _____
‘Hysterectomy
(removal of uterus)
•Removal of ovaries
(d«y ot wnk)
IF YES
a) How old were you when you started regular dgar smoking?-------------- years old
F.20.3 Comment
The layouts for F.20.1.1 and F.20.1.2 are cluttered. Those for F.20.1.3 and F.20.1.4 are better.
The illustrations in F.20.2.1 and F.20.2.2 involve use of GO TO or SKIP instructions in place of
are preferable to arrows or pointers when the form involves a lot of skips on the same page or when
the skips are to other pages of the form.
408
F. Data items and forms illustrations
F.21
INSTRUCTIONAL INFORMATION
F.21.1
F.22 Unformatted responses 409
F.22
F.22.1
Shading or highlighting (Photo reproduction)
PLEASE COMPLETE THE ONE SECTION BELOW WHICH APPLIES TO YOU.
Did you •vur work In ■ shipyard?
If you ARE
the person
named above
1
I
34
(Unsatisfactory; photo reproduction)
IS THERE A PREVIOUS HISTORY OF STROKE?
2._
.YES
1-
35
IF YES, MONTH
— DAV—
IF THERE WAS A CEREBROVASCULAR DISEASE DIAGNOSIS;
36
APPROXIMATE DATE OF ONSET: MONTH
DAY
—YR.
ACTIVITY AT ONSET OF STROKE----------------------------- -----------------------
I YES Please
and complete the yellow form. Skip page 2.
—NO
3.
UNKNOWN
unknown'
________YEAR
DATE UNKNOWN
_—NO STROKE ------
UNKNOWN
NO STROKE —
UNKNOWN
_NO STROKE —
37
PLACE AT ONSET OF STROKE
NO Please RETURN THIS QUESTIONNAIRE in the enclosed envelope so we can correct our
records and prevent another mailing to you Do not complete the remainder ot this questionnaire.
0
38
UNKNOWN—.NO STROKE
39
If you ARE NOT
the person
named above
UNFORMATTED RESPONSES
I
1 Please check box and
COMPLETE PAGE 2 ot the white form.
F.22.2
(Satisfactory; photo reproduction)
2) Ha. the patient been hospitalized at any time since his last quarterly exammation’ . . . t^)
THIS SURVEY IS AUTHORIZED UNDER THE DOE ORGANIZATION ACT Pt 95-91
F.21.2
F.21.2.1
OM0 NO 1901-0250 EXP DATE 9/30784
If YES, give the date, duration and reason
Boxes
Boxed instructions for entire form (Photo reproduction)
for each hospitalization in the space below.
REASON
DURATION (Days)
DATE
Complete thia form for all candidates who remain eligible after BL 1.
Proceed immediately to Form 04CP once this form is completed.
Answers to questions with boxed | | item numbers must be reviewed by an
HPT physician. Questions preceded by"TAAW) are to be asked as written; those
preceded by (SAW) are to be shown as written using HPT flashcards. Questions
preceded by (DA) are to be answered using information from previous items.
F.22.3 (Satisfactory; facsimile)
F.21.2.2
Boxed instructions for specific item (Photo reproduction)
14. BL 2 Appointment:
AM
PM
a. Time of day:
b. Day of week:
hr.
“ffo
7. Did the participant die
hospital?
(
(
Yes
No >2
in a
(
)
Unk
If Yes, give the name and address
c. Date:
“D’ay
NOTE: Date in item 14 must be at
least 7 days after date in item
20b on Form 01CP
of tFe hospital and attending
physici an.
a. Name and address of hospital.
15. Candidate given:
(
(
(
(
(
)1
)1
)’
)1
)1
Appointment card
Participant reminder form
Informed consent material
Material introducing HPT
Other (specify):
F.21.3 Comment
Instructional information contained on a form should be identified. This is done via reverse image
of a- different print font in F.21.1. Boxes are used to set off instructional
printing, pointers, and use
t-------material in F.21.2.
__
APPROXIMATE HOUR OF ONSET OF STROKE
b. Name and address of attending
physician.
(ju>
JV j.xrthltfiiiilii
F.24 Layout for check positions 411
F. Data items and forms illustrations
410
F.22.4 Comment
There is not enough vertical space between lines in F.22.1. There should be at least
of separation
between lines, as in F.22.2, or the form should provide a boxed space of adequate size, as in F.22.3,
when handwritten responses are required.
F.23
F.24
LAYOUT FOR CHECK POSITIONS
Varied order for yes-no responses (Unsatisfactory; photo reproduction)
F.24.1
At lose ona a
No
Ya
FORMATTED RESPONSES
F.23.1
Boxes (Photo reproduction)
INTERVIEWER ASK:
3.
Do you wish to have the results of your tests sent to your
physician?
1 [J YES i
33
20W
GO TO QUESTION 12
33
5.
4.
Tit la
i’• i N i i First
i i Nama
i i u ^~ri
Actrnty
How many dm«a p«r wtk?.
SOCIAL AND DIETARY HABITS
No
Ya How many CIGARETTES par day?
♦. Do you imoke CIGAR11 I ES now?
No
Ya How many CUPS per day?
Do.you drink COFFEE now?
No
Ya How many CUPS par day?
Do you drink TEA now?
No
Ya How many GLASSES par day?
Do you drink MILK now?
IX Whan aunt mau do you avoid atin< the fax? Ya Q No
IX Do you often add salt to your food? Ya No Q Ptppor? Ya No Q
Caiauy. award, or ipica? Ya No Q Mayonnam or >alad oil? Ya O No
I I I I I I I Last
I I Nona
I I I I I I 1 I I I I I
6.
do you enpft in iny rtfuiar ictivity akin to bn* wrtuif. joumf. biodinf. etc. lon( tnoufh co work
si
i i i i i i r i i i Clinia/Building
i Ji, i. i i i i I i i i i i i
7»
F.24.2
3.
Illi rT"l I I IStraat
I I Addrasa
I I I I I I I I I I I I I I
as
8.
Scattered check positions (Unsatisfactory; photo reproduction)
I I I I I I I I I I Czty
,1 I I I I I I I I I I 10.1*0|.
n. I I I Zip
1,1.1,
I T~i~l
Coda
9.
Stata
in
1*2
F.23.2
Hatched lines (Photo reproduction)
Social Security Number
L
c 62
i
I
63
M
Long leg sitting (knees <90° flex) with arm support R/L
(I) without arms support
time
weight distribution —symmetrical —more on L/R
hips abd —neutral; pel vis: thigh
pelvis: thigh
hip rotation
neutral
'
| w/knees pass. ext. to
trunk
erect
kyphosis
lateral flexion R/L
shoulders
elevated
retracted
protracted
head
erect
cradTeS’
lateraTTlexion R/L
LE’s
relaxed
fixing in flexion
tonic extension knees
UE support
fonlard R/L
lateral ITT
posterior R/L
shifts weigKt”to
L
recovers
R
recovers
ant ” recovers
post
recovers
diag’Tack L
recovers
diag back R
recovers
Social Security Number unavailable
L
c 55
!
66
C
67
IC32-70 all
PUNCH j O'sif SS
I
|
68
69
70
unavail.
NOW I WOULD LIKE TO ASK YOU A FEW QUESTIONS ABOUT WHERE YOU HAVE LIVED
FOR THE PAST FEW YEARS
F.24.3
Scattered check positions (Photo reproduction)
/
25.
When was the last time you smoked that much?
26.
About how many cigarettes per day did you smoke 2 years ago?
27. Did you smoke cigarettes a year ago?
YES |
| ASK Q. 28
IF YES, ASK:
?; -
28. About how many cigarettes did you smoke a year ago? -----
See Section 12.6.8.
29. Was there ever any period of time when you stopped smoking?
YES [71 *SI( Q- 30‘3’
UNKNOWN
NO
SKIP TO Q. 32
UNKNOWN O SK,P T0 Q- 29
UNKNOWN
Comment
F.23.3
UNKNOWN
MONTH YtAft
NO
SKIP TO Q. 34
UNKNOWN [31 SKIP TO Q. 34
412
F.24 Layout for check positions 413
F. Data items and forms illustrations
Inadequate check space (Photo reproduction)
F.24.4
SOOLAJU.T OUALITY OF MOOIUM FACULTY
) OISriHGUlSHCD
strong
( ) GOOD
( ) AOCOUATe
( ) MARGINAL
( )) NOT SUfRICICNT fOR DOCTORAL EDUCATION
) DON'T KNOV MILL INOUGN TO tVALUATe
F.24.9
EFFECTIVENESS OF PROGRAM IN EDUCATING RESEARCH
SCHOLARS/SCIENTISTS
f
9
6
1.
2.
2. { )
3. ( )
R. ( )
eXTRlMlLY IFFICTIVI
RCASONABLY IFFCCTIve
MINIMALLY IFFeCTIVB
NOT eFFCCTIve
0. ( )
DON'T KNOtt NBLL INOUGH TO IVALUATC
r. ( )
Eye lines and blocking (Contrived from F.24.8)
Hepatobiliary Pathology on Autopsy (check one or more)
3.
4.
5.
6.
CMAMOe IN PWOOIUM QUALITY IN LAST Five YEARS
FAMUMITY «TW WOW OF FWOORAM FACULTY
t. ( )
2. ( )
3. ( )
F.24.5
t. ( )
2. ( )
3. ( )
coNsroeRAtLe rartlimttv
SOMf gAMlLlARITY
LITTLl OR NO FAMILIARITY
Unsatisfactory list layout (Photo reproduction)
20. An you uactor tTMOMut for
blood prim?
21. How imsy boun bao it bna sdcr your Iwt ybrnII
Qj
Qi
SEPARATED Qi
YESO1
......................................
Fatty metamorphosis
Cirrhosis
Congestion
Blocked list (Facsimile)
10. »reo'nui^g;
(AAW) Have you taken any
™-pfresemptTion
(Use HPT
drugs in
i.. the
—■ last month?
Chart 13 as flash card if you wish)
(JS)
noQi
bourt
(
(
(
(39)
Satisfactory list layout (Contrived from F.24.5)
Please
(
(
(
(
(
F.24.7
F.24.10
MARRIED
WIDOWED
)2
)5
Never married
Married
Separated
Divorced
)5
Widowed
[h”i
I
I
pi
3.
4.
5.
6.
7.
8.
9.
)’
)i
)1
)«
(
)’
)’
)i
No autopsy
Normal liver
Hepatitis
Fatty metamorphosis
Cirrhosis
Congestion
Nonspecific changes
Gall stones
Other findings (please specify):
Coryban D
Dexatrim
Dristan
Excedrin
Midol
Nodoz
Permathene - 12
!! Prolamine
Excessive white space (Contrived from F.24.8)
(
(
Hepatobiliary Pathology on Autopsy (check one or more)
1.
2.
Anacin
Appedrine
Bromoquinine
)i
)i
check marital status:
)i
)
)
)
7. Nonspecific changes
8. Gall stones
9. Other findings (please specify)
DON'T KNOM RILL INOUGN TO IVALUATI
NKVER
MARRIED LJ1
18. Rmm d»efc Marital Statna:
DIVORCED Q*
IS. How masy GRADES o< xhooi. indndinE Colton, dM yoa axnptow? ..........
F.24.6
BITTIR THAN FIVE YtARS AGO
LITTLl OR NO CHANGt IN LAST FIVl YIAR
POORIR THAN Five YCARS AGO
(
(
(
No autopsy ..
Normal liver
Hepatitis ..
(
(
(
(
(
(
(
(
(
)
)
)
)
)
)
)
)
)
F.24.11
Comment
)’
)1
Tri aminicin
Vanquish
of the variation in the position of the Yes-No check spaces.
12Hems F 24 2 and F.24.3 are both examples of layouts with scattered locations for check positions.
F.24.8 Eye lines (Photo reproduction)
Hepatobiliary Pathology on Autopsy (check one or more)
No autopsy .
Normal liver
Hepatitis ..
( )
( )
( )
6.
Fatty metamorphosis
Cirrhosis
Congestion
)
( )
( )
7.
8.
9.
Nonspecific changes
Gall stones
Other findings (please specify):
( )
( )
( )
1.
2.
3.
4.
5.
shown in F.24.6, or with better spacing or .^ffere^
layout ^n F.24.7 requires the
(
F.24.6 through F.24.10 involve lists with verucal
layo^. TheJayout, i^
XKiXS i« --™
To the'left of the item, as shown in F.24.10. avo.ds the need for eye hues.
F.25 Field designations and precoded responses 415
414
F. Data items and forms illustrations
F.25
FIELD DESIGNATIONS AND PRECODED RESPONSES
F.25.1
Precoded responses
F.25.2
F.25.2.1
Field designations (Satisfactory; photo reproduction)
io.
III.
FAMILY HISTORY
Age if
Living
Age at
Death
(Satisfactory; photo reproduction)
Which choice below would help you lose the most weight?
(Circle one.)
Cauae of*
Death
20. Father
145-46
11.
21. Mother
TTC-sT
22. Brother(»)
01
.02
.03
.04
Jogging 1 mile
Walking 3 miles
Refusing one piece of cherry pie.
Refusing one 12-ounce soda . .
Reed each statement below and then circle the number to the right of the statement that best describes how
YOU feel. If you STRONGLY AGREE with a statement, circle number 1. If you STRONGLY DISAGREE,
circle number 5. Try not to circle 3 unle» you are really unsure of how you feel.
(Circle one number on each line.)
A
”137
133-53
Strongly
Agree
Agree
Unsure
Disagree
Strongly
Disagree
B
160-161
137-5T
”131
135-63
”167-68
“137
a.
My health is directly affected
by the actions I take every
day
01
02
03
04
05
b.
I am doing all I can to
keep myself healthy
01
02
03
04
05
c.
People can't do anything to
reduce their chances of
having a heart attack
01
.02
03
04
05
d.
I am not concerned about
preventive health measures ... .01
03
02
04
05
e.
I just go on doing the things
that make me less healthy even
when I know I shouldn't
03
02
04
05
C
D
E
“177
23. Sleter(e)
A
”135-31
“137-31
"T73
“137-B3-
“137
01
B
185-86
F.25.2.2 (Photo reproduction)
C
“173-71
“177
“175-^75”
“177
“FJ3-3I
“737
8) In your best judgment (based1 on a capsule
count and/or any other information or impressions obtained from the patient at this
visit), what percentage of the capsules of AL
LOCATED medication (i.c., from bottles 1-30)
prescribed since Initial Visit 3 has die patient
actually taken?
D
E
At least 80%---------------------------------------------- ( 1 )”
At least 60%but less than 80%----------------- ( * )
At least 40%but less than 60%------------------- ( 3 )
At least 20%but less dian 40%------------------- ( 4 )
Less than 20%------------------------------------------ ( 5 )
24. CoeMOts:
•Caueee of Death:
1
2
.... I
i'!
3
4
5
6
7
8
9
Coronary Heart Diaeaae (Heart Attack)
Stroke
Hypertenalve Heart Dlaease
Cancer
Other Heart Diaeaae (apecify)
Other Diaeaae (epeclfy)
Accident or trauma
Unknown
Old Age
F.25.2.3
(Facsimile)
29. (AAW) Who usually prepares the
111 at home?? (Use
meals you eat
list below to categorize response)
(
(
(
(
(
(
alone
Spouse or partner, alone
J2 Candidate
Someone else, alone
and spouse or
r Candidate
partner
I5
Candidate and someone else
Other (specify)
416
F. Data items and forms illustrations
F.25.3 Comment
The numbers appearing in items 20 through 23 in F.25.1 correspond to field designations for the data
entry process. The numbers appearing in F.25.2.1, in the parentheses in F.25.2.2, and next to the
parentheses in F.25.2.3, correspond to defined codes for designated fields.
G. Sample manual of operations, handbook,
and monitoring report
Chapter 4. Policy Matters (19 pages)
4.1 Introduction
4.2 Informed Consent
4.3 Blinding and Breaking of the Code
4.4 Ancillary Studies and the Disclosure
of Ancillary Study Results
4.5 Blinding Ancillary Study Investiga
G.l Introduction
G.2 Table of contents of the National Coopera
tive Gallstone Study Clinic Manual of
Operations (July 1975 version)
G.3 Listing of pages in the Hypertension Pre
vention Trial Handbook (April 7, 1983
version)
G.4 Sample tables from Macular Photocoagu
lation Study Treatment Monitoring Re
port (January 31, 1982)
G.5 Listing of tables in the final Treatment Ef
fects Monitoring Report of the Persantine Aspirin Reinfarction Study (Octo
ber 15, 1979 Database)
G.l
INTRODUCTION
The sample documents contained herein were
provided by the Hypertension Prevention Trial
(HPT), Macular Photocoagulation Study
(MPS), National Cooperative Gallstone Study
(NCOS), and Persantine Aspirin Reinfarction
Trial (PARIS). All four trials are sketched in
Appendix B.
G.2 TABLE OF CONTENTS OF THE
NATIONAL COOPERATIVE GALLSTONE
STUDY CLINIC MANUAL OF
OPERATIONS (July 1975 Version)
I
tors
Protocol Changes During the Trial
Evaluation of Treatment Center Per
formance
4.8 Phase III
4.9 Sharing Costs
Chapter 5. Patient Recruitment (10 pages)
5.1 Introduction
5.2 Mechanism of Referral
5.3 Patient Recruitment Sources
5.4 Editorials and Newsletters
4.6
4.7
PART I Introduction
Chapter 1. Background and Development (2
pages)
Chapter 2. Objectives and Study Design (3
pages)
2.1 Study Objectives
2.2 Study Design
2.3 Treatment Schedule
2.4 Interim Monitoring
Chapter 3. Organization (7 pages)
3.1 Introduction
3.2 The Coordinating Center and
Its Units
3.3 The Steering Committee and Sub
committees
3.4 The Advisory Board
PART II Evaluation and Follow-Up
Chapter 6. Patient Eligibility and Exclusion
Criteria (10 pages)
6.1 Introduction
6.2 Eligibility Criteria
Chapter 7. Randomization and Drug Pack
aging (10 pages)
7.1 Introduction
7.2 Master Randomization List
7.3 Drug Packaging and Distribution
7.4 Randomization
7.5 Parallel Ancillary Studies
7.6 Local Access to the Treatment Code
Chapter 8. Drug Administration (13 pages)
8.1 Introduction
8.2 Project Physician
8.3 Treatment Center Pharmacist
Chapter 9. Dosage Level Adjustment (5 pages)
9.1 Introduction
9.2 Initial Stabilization
9.3 Dosage Level Re-Adjustment
9.4 Assignment to Known Placebo
9.5 Reinstatement of Assigned Therapy
Chapter 10. Schedule and Description of Pa
tient Visits (18 pages)
10.1
417
Introduction
G.3 Listing of pages in the Hypertension Prevention Trial handbook
418
II
I
I
419
G. Sample manual of operations, handbook, and monitoring report
10.2 Evaluation Visit
10.3 Randomization Visit
10.4 Follow-up Visits
10.5 Interim Visits
Chapter 11. Dropouts and Withdrawals (3
pages)
11.1 Introduction
11.2 Dropouts and Withdrawals
11.3 Missed Visit Form
11.4 Reinstatement of Dropouts
11.5 Transfers
11.6 Tracking Procedures
Chapter 12. Management of Illness (15 pages)
12.1 Introduction
12.2 Medical Staff and Responsibilities
12.3 Summary of Clinical Management
Procedures
12.4 Management of Gastrointestinal dis
orders
12.5 Management of Hepato-Biliary Dis
orders
12.6 Management of Any Other Illness
Possibly Related to Bile Acid Feed
ing
Chapter 13. Permanent Cessation of Therapy
(3 pages)
13.1 Introduction
13.2 Permanent Cessation of Therapy
13.3 Death of a Patient
13.4 Biological Specimens
PART III Procedures
Chapter 14. Radiological Procedures (5 pages)
14.1 Introduction
14.2 Patient’s Preparation
14.3 Obtaining the X-Ray
14.4 Development of Film
14.5 Non-Visualizing Gallbladder
14.6 Reading the X-Ray
14.7 Dissolution of Gallstones
14.8 Handling and Storage
Chapter 15. Determination of Gallstone Vol
ume and Number (4 pages)
15.1 Introduction
15.2 Visual Reading
15.3 Computer Assisted Reading
Chapter 16. Procedures for Duodenal Intu
bation (6 pages)
16.1 Introduction
16.2 Preparation
16.3 Duodenal Intubation
16.4 Shipment of Specimens
16.5 Processing of Specimens
16.6 Supplies
Chapter 17. Procedures for Liver Biopsies (5
pages)
17.1 Introduction
17.2 Preparation
17.3 Performing the Biopsy
17.4 Treatment Center Pathology Labora
tory
17.5 Central Morphology Laboratory
Chapter 18. Procedures for Electrocardio
gram (1 page)
18.1 Introduction
18.2 Performing the ECG
Chapter 19. Collection and Handling of Lab
oratory Specimens (16 pages)
19.1 Introduction
19.2 Preparation
19.3 Urine
19.4 Blood and Serum
19.5 Shipment of Specimens
19.6 Processing of Specimens
19.7 Supplies
Chapter 20. Laboratory Evaluations and
Tests (6 pages)
20.1 Introduction
20.2 Required Tests
20.3 Test Result Ranges
Chapter 21. Medication Procurement and
Handling (8 pages)
21.1 Introduction
21.2 Capsule Formulation
21.3 Capsule Requirements
21.4 Ordering
21.5 Manufacture
21.6 Shipping
Chapter 22. Treatment Center Procedures (6
pages)
22.1 Introduction
22.2 Scheduling Visits
22.3 Visit Procedures
22.4 Preparing for a Patient Visit
22.5 Checking Forms
22.6 Medical Records
22.7 Other Responsibilities
PART IV Data Handling and Analysis
Chapter 23. Completion and Mailing
Forms (11 pages)
23.1 Introduction
23.2 Forms Supplies
23.3 Patient Identification
of
23.4 Completing Forms
23.5 Batching and Mailing
23.6 Receipt of Forms
23.7 Treatment Center Enquiry
Chapter 24. Data Management Procedures
(9 pages)
24.1 Introduction
24.2 Eligibility Evaluation
24.3 Data Entry
24.4 Randomization Processing
24.5 Editing
24.6 Update Procedures
24.7 Form Revision
24.8 Back-Up and Security
Chapter 25. Endpoints and Statistical Anal
yses (9 pages)
25.1 Introduction
25.2 Evaluation of Treatment
25.3 Final Statistical Analyses
25.4 Intermediary States
25.5 Interim Statistical Analyses
Chapter 26. Publicity and Publications (4
pages)
26.1 Introduction
26.2 Press Releases and Interviews
26.3 Presentations and Publications^
Review of
and Publi
26.4 Rw..."
-- Presentations
-------cations
G.3 LISTING OF PAGES IN THE HYPER
TENSION PREVENTION TRIAL HAND
BOOK (April 7, 1983 Version)1
1. Background
• Abbreviations
• Definitions (3 pages)
• Proposed time schedule
• HPT objectives
• Specific aims and operational tasks of
Stage 1
• HPT landmark dates
• Design and operating synopsis for Stage 1
• Projected distribution of participants for
Cohorts 1, 2, and 3 combined
• Minimal detectable treatment-control dif
ference
• Test cohort distribution of participants by
treatment assignment
• Anticipated distribution of participants by
treatment group for Cohorts 1, 2, and 3
combined (2 pages)
I. Each line in the listing represents a page in the Handbook,
except where indicated.
• Identification numbers
• Document numbering scheme (3 pages)
• HPT numbered documents (2 pages)
2. Organization
• HPT centers
• HPT bodies
• Investigative Group specifications (1G)
• Steering Committee specifications (SC)
• Executive Committee specifications (EC)
• Quality Assurance Committee specifica
tions (QAC)
• Intervention Methods Committee specifi
cations (IMC)
• Laboratory Committee specifications
(LC)
A ,
• Treatment Effects Monitoring and Analy
sis Committee specifications (TEMAC)
• Policy Advisory Board specifications
(PAB)
• HPT participating centers
• HPT functions by centers
• Steering Committee and Executive Com
mittee composition
• Policy Advisory Board composition
• HPT meetings
• Steering Committee rules
• HPT Committee structure
3. Enrollment
• Major HPT inclusion and exclusion cri
teria
• HPT Chart 05: Exclusion conditions by
form and item number (2 pages)
• HPT Chart 06: Baseline Visit 1 exclusion
drugs (2 pages)
• HPT recruitment approaches
• Number of persons to be covered in re
cruitment mailing
• Randomization features
• HPT consent and enrollment procedures (2
pages)
4. Weight strata
• HPT weight strata
• HPT Chart 07: Minimal BL 1 body weight
required for classification in the high (H)
weight stratum
• HPT Chart 15: BL 1 body weight exclusion
limit for a Quetelet’s Index > 0.0500
• HPT Chart 11: Quetelet’s index by height
and weight
G.4 Sample tables from MPS treatment monitoring report
420
421
G. Sample manual of operations, handbook, and monitoring report
5. Intervention
• HPT Stage 1 study treatments
• HPT Chart 13: Time schedule for interven
tion contacts
• Required forms and procedures for inter
vention contacts
• HPT Chart 16: Intervention documents list
6. Data schedule
• Forms list
• Charts list
• Flashcards list
• Required forms and related documents for
data collection visits (2 pages)
• HPT Chart 18: HPT data collection exami
nation schedule (2 pages)
• HPT procedures by data collection visit
• Blood pressure measurement
• Height measurement
• Weight measurement
• Blood collection
• Blood processing for serum collection
• Upper right arm girth measurement
• Skinfold thickness measurement
• Electrocardiogram (ECG)
• 24-hour food record collection
• 24-hour food record processing
• 24-hour food record review and mailing
• Urine collection
• Urine processing
7. Time constraints
• Cohort specifications
• Time window specifications for baseline
visits
• Time window specifications for follow-up
visits
• HPT Chart 13: Time schedule for interven
tion contacts
8. Assurance procedures
• Quality assurance procedures
• Certification and recertification schedule
• Equipment inspection and maintenance
schedule
• Clinic certification and recertification
• Personnel certification and recertification
• BP observer certification and recertifica
tion specifications
• Center director certification and recertifi
cation specifications
• Clinic coordinator certification and recerti
fication specifications
• Data entry operator certification and recer
tification specifications
• Dietary interventionist certification and
recertification specifications
• Food record documentor certification and
recertification specifications
• Laboratory technician certification and
recertification specifications
• ECG technician certification and recertifi
cation specifications
• Skinfold observer certification and recerti
fication specifications
• Study physician certification and recertifi
cation specifications
• Wt-Ht observer certification and recertifi
cation specifications
• General inspection and maintenance proce
dures
• Standard BP manometer inspection proce
dure
• Standard BP manometer maintenance
procedure
• Random zero BP manometer inspection
procedure
• Random zero blood pressure manometer
maintenance procedures (2 pages)
• Scale inspection procedure
• Scale maintenance procedure
• Skinfold caliper inspection procedure
• Room thermometer inspection procedure
10. Miscellaneous
• Medical management principles
• Clinic supply procedures
• Document storage and distribution
• Center to center communications ground
rules (2 pages)
G 4 SAMPLE TABLES FROM MACULAR PHOTOCOAGULATION STUDY
TREATMENT MONITORING REPORT (January 31, 1982 RePort>
Full report available via the Nat.onal Technical Information Service, Springfield, VA. accession no.
PB83-168-179.
Available information* by assignment, SMD study, 01-31-82
• Design and data separation principles
• HPT policy prohibitions
• HPT document access policy
• HPT publicity policy
• HPT authorship policy
• HPT presentation and publication policy
• HPT ancillary study policy
• HPT abstract submission clearance policy
(2 pages)
• HPT external presentation clearance proce
dure
• HPT manuscript clearance procedure
• HPT distributed data analysis policy
Treated
Untreated
Visit
Initial visit
Treatment visit
Completed
Missed
Delinquent]
Completed
Missed
Delinquent^
113
113
111
Post-treatment
(6-weeks)
106
I
3
100
0
1
Follow-up 01
( 3 months)
97
81
3
0
81
1
2
Follow-up 02
( 6 months)
52
0
1
53
0
0
Follow-up 03
(12 months)
23
0
0
18
2
0
Follow-up 04
(18 months)
10
0
1
10
0
I
Follow-up 05
(24 months)
0
0
0
2
0
0
Follow-up 06
(30 months)
263
4
5
264
3
4
5
596
5
5
All follow-up visits
9. Policy matters
• Exercise and diet preparation for BL 1, 2,
3, IE and all FUs
• Data rounding rules
• Data entry and correction procedures
• Data documentation assurance
All visits
374
4
at the coordinating center.
-Tl.v
•The numbers in this table renect clinic forms received and entered no form has been received in the coordinating
tDelinquent visits (visits for which the time window has expired, but
center).
422
G.5 Table listing from PARIS treatment effects monitoring report
G. Sample manual of operations, handbook, and monitoring report
G.5 LISTING OF TABLES IN THE
FINAL TREATMENT EFFECTS
MONITORING REPORT2 OF THE
PERSANTINE ASPIRIN REINFARCTION
STUDY (October 15, 1979 Database)
Change in visual acuity from initial visit (IV) to specified follow-up visits (FV), SMD study
Untreated
Treated
4 ( 4.1)
45 (46.4)
23 (23.7)
15 (15.5)
8 ( 8.2)
2 ( 2.1)
97
16 (16.0)
57 (57.0)
9 ( 9.0)
4 ( 4.0)
6 ( 6.0)
8 ( 8.0)
100
4 ( 4.9)
22 (27.2)
18 (22.2)
13 (16.0)
8 ( 9.9)
16 (19.8)
16 (19.8)
38 (46.9)
13 (16.0)
1 ( L2)
6 ( 7.4)
7 ( 8.6)
81
81
Distribution of change from IV to FV01:
> 1.5 lines better
< 1.5 lines change
1.5 3.5 lines worse
3.5 5.5 lines worse
5.5 7.5 lines worse
> 7.5 lines worse
Total eyes
Section 1. Patient follow-up
• Status of patients by treatment group
• Number and percentage missed visits by
treatment group and follow-up visit
Distribution of change from IV to FV02:
> 1.5 lines better
< 1.5 lines change
1.5- 3.5 lines worse
3.5- 5.5 lines worse
5.5- 7.5 lines worse
> 7.5 lines worse
Total eyes
Section 2. Fatal and nonfatal events
p = 0.06
Distribution of change from IV to FV04:
> 1.5 lines better
< 1.5 lines change
1.5- 3.5 lines worse
3.5- 5.5 lines worse
5.5- 7.5 lines worse
> 7.5 lines worse
Total eyes
2 ( 8.7)
5 (21.7)
2 ( 8.7)
1 ( 4.3)
5 (21.7)
8 (34.6)
0 ( 0.0)
II (61.1)
1 ( 5.6)
2 (11.1)
0 ( 0.0)
4 (22.2)
23
18
7 ( 8.3)
16 (19.0)
14 (16.7)
9 (10.7)
13 (15.5)
25 (29.8)
14 (17.1)
33 (40.2)
8 ( 9.8)
15 (18.3)
3 ( 3.7)
9 (11.0)
84
82
p = 0.06
Distribution of change from IV to LAST FV after FV01:
> 1.5 lines better
< 1.5 lines change
1.5- 3.5 lines worse
3.5- 5.5 lines worse
5.5- 7.5 lines worse
> 7.5 lines worse
Total eyes
P
0.00003
Number of patients with selected baseline characteristics, SMD study
?
<
Untreated
Treated
105 ( 94.6)
6 ( 5.4)
109 ( 96.5)
4 ( 3.5)
23 ( 20.7)
88 ( 79.3)
16 ( 14.2)
95 ( 84.1)
Cigarette use
Little or none
> 10 per day
84 ( 75.7)
27 ( 24.3)
86 ( 76.1)
27 ( 23.9)
Aspirin use
< 4 per week
4-28 per week
> 28 per week
87 ( 78.4)
22 ( 19.8)
2 ( 1.8)
79 ( 69.9)
27 ( 23.9)
7 ( 6.2)
Untreated
Treated
Study Eye
Right
Left
66 ( 59.5)
45 ( 40.5)
60 ( 53.1)
53 ( 46.9)
Sex
Male
Female
63 ( 56.8)
48 ( 43.2)
46 ( 40.7)
67 ( 59.3)
Race
Caucasian
Black
Other
Hypertension
None
BP > 140/80 or on
medication
III (100.0)
0 ( 0.0)
0 ( 0.0)
113 (100.0)
0 ( 0.0)
0 ( 0.0)
Histo belt resident
Never
Ever
Presently
54 ( 48.6)
57 ( 51.4)
38 ( 34.2)
57 ( 50.4)
56 ( 49.6)
38 ( 33.6)
Diabetes
No
Yes or ?
• Number and percentage dead by treatment
group and cause of death (15 pages)
• Percentage dead by selected baseline char
acteristics and treatment group (4 pages)
• Percentage of patients with specified non
fatal events as reported on follow-up
visit forms by treatment group (3 pages)
• Number and percentage of patients with
specified nonfatal events as classified by
the Mortality and Morbidity Committee
by treatment group
• Percentage of patients with de novo and
recurrent angina by specified baseline
characteristics and treatment group (2
pages)
• Percentage of fatal and nonfatal events by
treatment group (6 pages)
• Number and percentage fatal and nonfatal
events as classified by the Mortality and
Morbidity Committee by treatment
group
• Plot of Z-values for observed differences in
proportion of deaths and critical bound
aries for a = 0.05 (2 pages)
• Plot of Z-values for observed differences in
proportion of coronary deaths or defi
nite Mis and critical boundaries
• Plot of Z-values for observed differences in
proportion of deaths or definite Mis and
critical boundaries for a = 0.05 (2 pages)
• Percentage of selected fatal and nonfatal
events by clinic and treatment group (7
pages)
• Cumulative fatal and nonfatal event rates
and number of patient intervals by treat
ment group and months of follow-up (5
pages)
• Lifetable plots by treatment groups of:
• Death—all causes
• Coronary death
2. Supplied by Persantine Aspirin Reinfarction Study Coordinating
Center. April 1983.
423
• Sudden coronary death
• Definite nonfatal Mis
• Fatal or nonfatal pulmonary coronary
death or definite MI
• Fatal or nonfatal pulmonary embo
lism or thrombophlebitis
• Death—all causes or definite MI
• Cardiovascular (CV) death or any CV
event, new angina excluded
• Death—all causes or any CV event,
new angina excluded
• New angina
• CV death or any CV event
• Death—all causes or any CV event
• Percentage of patients hospitalized by rea
son and treatment group, all follow-up
visits combined
• Percentage of patients hospitalized for spec
ified symptoms by treatment group, all
follow-up visits combined
Section 3. Adherence and side effects
• Percentage distribution of patients by ad
herence level and follow-up visits (2
pages)
• Percentage distribution of patients by
number of tablets prescribed and follow
up visits
• Percentage of patients with non-com
pliance as defined by the urine salicylate
and urine dipyridamole tests by follow
up visit and treatment group (3 pages)
• Percentage of patients with non-com
pliance as defined by the urine salicylate
and urine dipyridamole tests by follow
up visit and treatment group, excluding
patients on zero prescription (3 pages)
• Percentage of patients with non-com
pliance as defined by the urine salicylate
and urine dipyridamole tests by clinic,
follow-up visit and treatment group (10
pages)
• Percentage of all possible tablets taken, by
clinic and treatment group, all follow-up
visits combined
• Percentage of all possible tablets taken, by
follow-up visit and treatment group, all
clinics combined
• Percentage of patients with reduced level of
adherence by reason and treatment
group, all follow-up visits combined (2
pages)
1
424
G. Sample manual of operations, handbook, and monitoring report
• Percentage of patients complaining of spe
cific types of problems by treatment
group, all follow-up visits combined
• Percentage of patients with reduced pre
scription by reason and treatment group,
all follow-up visits combined
• Number of side effects forms submitted, by
clinic and treatment group
• Percentage of patients with side effects
forms by reason and by treatment group,
all follow-up visits combined (2 pages)
• Mean urine salicylate and percent of pa
tients with urine salicylate level above 10
mg/100 ml by visit and treatment group
• Percentage of patients with positive dipy
ridamole test for specified visits by treat
ment group
• Percentage of patients using or prescribed
specific types of medication since entry,
by follow-up visit and treatment group
(2 pages)
Section 4. Laboratory and clinical findings
• Percentage of patients reporting specified
symptoms by treatment group, all fol
low-up visits combined
• Percentage of patients with one oi more
laboratory or clinical measurements out
side given limits by treatment group, all
follow-up visits combined (2 pages)
• Percentage of patients with one or more
laboratory or clinical measurements out
side given limits by treatment group, all
follow-up visits combined excluding pa
tients outside the limits at baseline (2
pages)
• Distribution of changes from baseline to
first annual visit of laboratory measure
ments by treatment group (5 pages)
• Distribution of changes from baseline to
second annual visit of laboratory mea
surements by treatment group (5 pages)
• Distribution of changes from baseline to
third annual visit of laboratory measure
ments by treatment group (5 pages)
• Distribution of changes from baseline to
fourth annual visit of laboratory mea
surements by treatment group (5 pages)
• Percentage of patients with changes from
baseline to third anual visit in systolic
blood pressure and fasting glucose by
selected baseline characteristics and treat
ment group (4 pages)
H. Budget summary for Hypertension Prevention Trial Data
Coordinating Center
Table H-l DCC ceiling support levels as speci
fied in NHLBI Notice of Grant
Award
Table H-2 Projected allocation of funds by
budget category and year of study
Table H-3 Projected staffing pattern by year of
study, in full-time equivalents
(FTEs)
Table H-4 Projected travel expenses by year of
study
Table H-5 Other DCC expenses by year of
study
Table H -6 DCC percent allocation of funds, ex
cluding nonDCC related costs
Table H-7 Cost of DCC, relative to total pro
jected HPT cost
Table H-l
Award
Section 5. Non-study drug usage
• Percentage of patients using non-study
medications by treatment group, all fol
low-up visits combined (2 pages)
• Percentage of patients using acetamino
phen or dextropropoxyhene hydrochlo
ride, by follow-up visit
The tables in this Appendix are from documents
prepared by the Hypertension Prevention Trial
(HPT) Data Coordinating Center (DCC). See
Sketch 13, Appendix B, for design details. The
ceiling support levels reported in Table H-l cor
respond to actual values recorded in the Notice
of Grant Award from the National Heart, Lung,
and Blood Institute (August 17, 1981). The total
direct cost figures in Table H-2 are from tables
prepared by the DCC in a rebudgeting process,
prior to the award. Final negotiations led to a
modest reduction in the funds awarded for each
of the five years, thereby accounting for small
discrepancies in the total direct costs in Ta
ble H-2 compared to Table H-l.
DCC ceiling support levels as specified in NHLBI Notice of Grant
Time period covered
Study year
Start
End
Ceiling DCC
support level
01
02
03
04
05
Sept. 1, 1981
Sept. I, 1982
Sept. 1. 1983
Sept. I. 1984
Sept. 1, 1985
Aug. 31. 1982
Aug. 31, 1983
Aug. 31, 1984
Aug. 31, 1985
Aug. 31. 1986
$ 297,120
$ 524,963
$ 590,080
$ 636.524
$ 551.670
2,600.357
Total
!
Table H-2
Projected allocation of funds by budget category and year of study
Year of study
Category
' S'
1. Personnel*
2. Consultant
3. Equipment
01
02
03
04
05
$136,224
5,750
8,660
$198,441
5.750
500
$224,148
5,750
550
5.450
41.140
16,940
24,200
317.920
98.003
219.917
594.958
$237,596
5.750
605
6.200
45.258
18,618
26.640
346,029
118.884
227,145
641.438
$251,851
5.750
731
4. Supplies
5. Travel (Table H-4)
a. DCC-related
b. NonDCC-related
4,250
34,000
17,000
17,000
4,800
35,750
17,050
18.700
6. Other expenses (Table H 5)
a. DCC-related
b. NonDCC-related
112,610
49,666
62,944
284,136
72.182
211,954
7. Total direct costs
301,494
529.377
•Includes salaries as well as cost of fringe benefits.
425
6,750
32.252
11.728
20.524
258.648
121.775
136,873
555.982
H. Budget summary for Hypertension Prevention Trial Data Coordinating Center 427
426
H. Budget summary for Hypertension Prevention Trial Data Coordinating Center
Table H-4
Table H-3
Projected travel expenses by year of study*
Projected staffing patterns by year of study, in fulltime
Study year
equivalents (FTEs)*_________________________________
Year of study
Type of travel
01
02
03
04
05
DCC director
•/ DCC deputy director
v Senior statistician
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
Junior statistician
✓ Physician coinvestigator
■J Nutritionist
0.15
0.05
0.10
0.40
0.10
0.10
0.40
0.10
0.10
0.40
0.10
0.10
0.20
0.10
0.10
Research associate
Research associate
v' DCC coordinator
0.50
0.10
1.00
0.50
0.20
1.00
0.50
0.20
1.00
0.50
0.20
1.00
0.50
0.20
1.00
1.00
1.00
0.50
0.75
1.00
0.50
0.75
1.00
0.50
0.75
Study position^
A. Professional
I
Senior programmer
Junior programmer
Assistant programmer
0.20
1.00
0.50
0.50
Total professional PTEs
3.70
4.90
5.15
5.15
4.95
Secretary
Secretary
Clerk-typist
0.50
0.20
0.65
1.00
0.50
0.65
1.00
0.75
0.65
1.00
0.75
0.65
1.00
0.75
0.65
Administrator
Administrator
0.15
0.10
0.15
0.10
0.15
0.10
0.15
0.10
0.15
0.10
1.60
2.40
2.65
2.65
2.65
5.30
7.30
7.80
7.80
7.60
B. Support
J
Total support FTEs
C. Total FTEs
r
•The corresponding table submitted to the NHLB1 contained actual projected
salaries for each person or position listed.
tPositions preceded by the symbol had a named individual listed in the applica
tion. Those not so identified were designated in the proposal as TBA (to be
appointed).
ri
I
1..
fl
,j
A. Travel for DCC staff
1. Advisory-Review Committee:
Person meetings/year
Cost
2. Steering Committee:
Person meetings/year
Cost
3. Executive Committee:
Person meetings/year
Cost
4. Standing subcommittees:
Person meetings/year
Cost
5. Treatment Effects Monitoring Committee:
Person meetings/year
Cost
6. Training sessions for clinic personnel.
Person sessions/year
Cost
7. Clinic site visits:
Person site visits/year
Cost
8. Professional meetings:
Person meetings/year
Cost
Total DCC travel
1. Executive Committee:
Person meetings/year
Cost
2. Standing subcommittees:
Person meetings/year
Cost
3. Treatment Effects Monitoring Committee:
Person meetings/year
Cost
4. Food Record Coding Center personnel:
Person trips/year
Cost
Total other travel
C. Total travel (Sum of parts A and B)
:
■
■
02
03
04
05
2
$ 1,000
2
$ 1,100
2
$ 1,210
2
$ 1,332
2
$ 1,466
6
$ 3.000
6
$ 3,300
6
$ 2,420
4
$ 2,664
4
$ 1,466
2
$ 1,000
2
$ 1,100
2
$ 1,210
2
$ 1,332
2
$ 1,466
2
$ 1,000
2
$ 1,100
2
$ 1,210
2
$ 1,332
2
$ 1,466
3
$ 1,500
3
$ 1,650
6
$ 3,630
6
$ 3,996
3
$ 2,199
3
$ 1,500
3
$ 1,650
3
$ 1,815
3
$ 1,998
0
0
12
$ 6,000
8
$ 4,400
4
$ 2,420
4
$ 2,664
0
0
4
$ 2,000
5
$ 2,750
5
$ 3,025
5
$ 3.300
5
$ 3,665
$17,000
$17,050
$16,940
$18,618
SI 1,728
4
$ 2,000
6
$ 3,300
6
$ 3,630
6
$ 3,996
6
$ 4,398
6
$ 3,000
6
$ 3,300
6
$ 3,630
6
$ 3.996
6
$ 4,398
6
$ 3,000
6
$ 3,300
12
$ 7.260
12
$ 7,992
6
$ 4,398
4
$ 2,000
4
$ 2.200
4
$ 2,420
4
$ 2,664
2
$ 1,466
14
$ 7,000
12
$ 6,600
12
$ 7,260
12
$ 7,992
8
$ 5,864
$17,000
$18,700
$24,200
$26,640
$20,524
$41,140
$45,258
$32,252
B. Travel for non-DCC staff
5. Study consultants:
Person trips/year
Cost
il
01
$34,000
•Costs calculated assuming $500 per person trip in year 01
$35,750
and increased by 10% each year thereafter.
H. Budget summary for Hypertension Prevention Trial Data Coordinating Center 429
428
H. Budget summary for Hypertension Prevention Trial Data Coordinating Center
Table H-6
Table H-5
DCC percent allocation of funds, excluding nonDCC related costs*
Other DCC expenses by year of study
Study year
Study year
01
02
03
04
$ 11.000
$ 19,500
$ 30,000
$ 41,000
$ 41,000
7.212
27,072
3.000
7,200
840
1.000
3,800
1,500
3,500
0
750
0
300
150
400
3,869
2,340
19,490
7,933
59,560
3,300
7,920
924
1,200
4,400
2,500
2,000
0
900
250
400
200
500
48,265
63,577
21,439
8,726
65,520
3,630
8,712
1,016
1,400
4,820
4,500
2,000
1,000
1,100
400
475
250
550
48,463
54,789
23,583
9,599
72,072
3,992
9,588
1,118
1,550
5,162
4,500
1,000
4,000
1,200
550
550
300
600
49.958
49,342
25,941
10,559
44,592
4,392
10,548
1,230
1,000
5,636
4,000
0
6,000
600
700
625
300
650
13,739
13,563
28,536
93,423
44,768
260,934
282,022
187,670
5,000
3,960
600
1,000
300
0
100
200
1,650
26,558
10,000
6,534
800
1,200
400
100
150
250
1,800
35,752
12,500
7,191
900
15.000
7,911
800
1,400
500
200
1,600
600
300
25. Bank vault rental
26. Books and journals
27. Space rental
28. MMRI overhead
1.500
1,080
400
800
200
0
0
150
1.500
13,557
175
300
2,000
38,841
200
350
2,200
42,017
Total MMRI
19,187
39,368
56,986
64.007
70,978
346,029
258,648
hem
A. Johns Hopkins (JHU)
1. Computing time
2. Intelligent terminals
a. JHU central terminal
b. Satellite terminals (8)
3. CRT workstation terminals (2)
4. Word processor
5. Telecopier
6. Telephone line charges
7. Photocopying
8. Data entry services
9. Forms printing
10. Manuscript page charges
11. Postage
12. Microfilming
13. Bank vault rental
14. Books and journals
15. Equipment maintenance
16. Central Laboratory
17. Food Record Coding Center
18. Study insurance
Total JHU
B. Maryland Medical Research Institute (MMRI)
19. Computing time
20. Word processor
21. Telephone line charges
22. Photocopying
23. Postage
24. Microfilming
p'lix
C. Total (Sum of parts A and B)
112,610
284,136
317.920
Category
01
02
03
04
05
Personnel
Equipment
Supplies
Travel
Other expenses
63.1
4.0
2.0
7.9
23.0
67.7
0.2
1.6
5.8
24.6
65.0
0.2
1.6
4.9
28.4
62.2
0.2
1.6
4.9
31.1
100.0
$381,903
64.1
0.2
1.7
3.0
31.0
05
Total
Total cost after exclusions
100.0
100.0
100.0
$215,800
$292,973
$345,091
100.0
$392,835
‘Excluding costs for consultants (line 2, Table H-2), nonDCC travel (Part B of Table H-4). Central Laboratory. Food
Record Coding Center, and study insurance (lines 16-18, Table H 5). Also excluded are costs for seven of the eight
terminals (line 2b, Table H-5) and overhead payments to MMRI (line 28, Table H 5).
Table H-7
Cost of the DCC relative to total projected HPT cost
Study year
01
02
03
04
05
Total
5-year
cost
a. Projected dollar cost, all HPT
centers combined*
879,966
2,054,880
1,597,509
1,671,896
1,541,713
7,745,964
b. DCC dollar cost (from Table
H-6)
215.800
292.973
345.091
381,903
392,835
1.628,602
c. DCC cost as percentage of line a
24.5%
14.3%
21.6%
22.8%
25.5%
21.0%
‘Based on totals derived from summary prepared prior to actual award. Funded totals differ only slightly froi>m those cited
above.
B 431
1. Combined bibliography
This appendix contains a listing of all citations appearing in this book, except those in Appendixes B
which the citations appear.
A
I. Abt K: Problems of repeated significance testing. Controlled Clin Trials 1:377-381, 1981.
2 Alamercery Y, Lo Presti F, Karisson T, Bowden G, and IMPACT Research Group: Functional
equality of coordinating centers in a multicenter clinical trial: Experience of the International
Mexiletine and Placebo Antiarrhythmic Coronary Trial: IMPACT (abstract). Controlled Clin
Trials 3:149, 1982.
3. Amberson JB Jr, McMahon BT, Pinner M: A clinical trial of sanocrysin in pulmonary
tuberculosis. Am Rev Tuberc 24:401-435, 1931.
4. American Bible Society: The Holy Bible: Old and New Testaments. King James Version
(1611), New York, 1816.
5. Anderson LK, Hendershot RA, Schoolmaker RC: Self-checking digit concepts. J Systems
Management 25:36-42, 1974.
6. Anscombe FJ: Sequential estimation. JRSS 40(series B):l-21, 1953.
7. Anscombe FJ: Fixed-sample size analysis of sequential observations. Biometrics 10:89-100,
1954.
8. Anscombe FJ: Sequential medical trials. JASA 59:365-383, 1963.
9. Anturane Reinfarction Trial Research Group: Sulfinpyrazone in [he Prevention of^cardiac
death after myocardial infarction: The Anturane Reinfarction Trial. N Engl J Med 298.289
295, 1978.
10. Anturane Reinfarction Trial Research Group: Sulfinpyrazone in the prevention of sudden
death after myocardial infarction. N Engl J Med 302:250-256, 1980.
II. Armitage P: Restricted sequential procedures. Biometrika 44:9-26, 1957.
FJ Anscombe’s paper. JA SA
12. Armitage P: Sequential medical trials: Some comments on F.
58:384-387, 1963.
13. Armitage P, McPherson CK, Rowe BC: Repeated significance tests on accumulating data.
JRSS I32(series A):235-244, 1969.
14. Armitage P: Statistical Methods in Medical Research. John Wiley and Sons, New York, 1971.
15. Armitage P: Sequential Medical Trials (2nd ed.). John Wiley and Sons, New York, 1975.
16 Aspirin Myocardial Infarction Study Research Group: Aspirin Myocardial Infarction Study:
' Design. Methods, and Baseline Results. Publ. no. 80-2106, National Heart, Lung, and Blood
Institute, Bethesda, MD, 1980a.
17. Aspirin Myocardial Infarction Study Research Group: A randomized controlled trial of
aspirin in persons recovered from myocardial infarction. Z4AM 243:661-669, 198Ub.
B
CH, Hursh-Cesar G: Survey Research (2nd ed). John Wiley and Sons, New York,
18. Backstrom
1981.
19. Bailey NTJ: The Mathematical Approach to Biology and Medicine. John Wiley and Sons,
New York, 1967.
430
20
5
I
I
A
20
20
9
9, 18
9, 22
9
9
9, 20
18
9
6
16, 18
12
20
20. Barker KN: Data collection techniques: Observation. Am J Hosp Pharm 37:1235-1243, 1980.
21. Barnard PJ, Wright P, Wilcox P: Effects of response instructions and question style on the
ease of completing forms. J Occup Psychol 52:209-226, 1979.
22. Baron J: The life of Edward Jenner: With illustrations of his doctrines and selections from his
correspondence (book review). Br Foreign Med Rev 6:477-497, 1838.
23. Bayes T: An essay towards solving a problem in the doctrine of chances (reproduced in
Biometrika 45:296-315, 1958). Philos Trans Roy Soc bond 53:370-418, 1763.
24. Beatty WK: Searching the literature and computerized services in medicine: Guides and
methods for the clinician. Ann Intern Med 91:326-332, 1979.
25. Beeson PB, McDermott W, Wyndgaarden JB (editors): Cecil's Texthoqk of Medicine (15th
ed). WB Saunders Co, Philadelphia, 1979.
26. Begg CB, Iglewicz B: A treatment allocation procedure for sequential clinical trials. Biometrics
36:81-90, 1980.
27. Begin G, Boivin M, Bellerose J: Sensitive data collection through the random response
technique: Some improvements. J Psychol 101:53-65, 1979.
28. Begin G, Boivin M: Comparison of data gathered on sensitive questions via direct question
naire, randomized response technique, and a project method. Psychol Rep 47:743-750, 1980.
29. Begun JM, Gabriel KR: Closure of the Newman-Keuls multiple comparisons procedure.
JASA 76:241-245, 1981.
30. Beta Blocker Heart Attack Trial Research Group: Beta Blocker Heart Attack Trial. Design
features. Controlled Clin Trials 2:275-285, 1981.
31. Bishop GF, Oldendick RW, Tuchfarber AJ: Effects of presenting one versus two sides of an
issue in survey questions. Public Opinion Quart 46:69-85, 1982.
32. Boston Globe (The): Data changed to put patients in cancer test. June 30, 1980a.
33. Boston Globe (The): Research staff feared for patient’s welfare. July I, 1980b.
34. Boston Globe (The): Passing the buck. July 2, 1980c.
35. Boston Globe (The): MD under fire gets new grant. July 3, 1980d.
36. Boston Sunday Globe: Cancer research data falsified; Boston project collapses. June 29, 1980.
37. Box JF: R. A. Fisher and the design of experiments, 1922-1926. Am Statistician 34:1-7,1980.
38. Brandt AM: Racism and research: The case of the Tuskegee Syphilis Study. Hastings Center
Report 8:21-29, Dec, 1978.
39. Braunwald E: Coronary-artery surgery at the crossroads. N Engl J Med 297:661-663, 1977.
40. Breslow NE, Day NE: Statistical Methods in Cancer Research (vol 1, pg 93). International
Agency for Research on Cancer, Lyon, 1980.
41. Bross I: Sequential medical plans. Biometrics 8:188-205, 1952.
42. Bross IDJ: Laetrile. N Engl J Med 307:118, 1982.
43. Brown BW Jr, Hollander M: Statistics: A Biomedical Introduction. John Wiley and Sons,
New York, 1977.
44. Brown BW Jr: Designing for cancer clinical trials: Selection of prognostic factors. Cancer
TYeat Rep 64:499-502, 1980.
45. Brown BW Jr: Comments on the Dupont manuscript. Controlled Clin Trials 4:13-17, 1983.
46. Bryan LS Jr: Blood-letting in American medicine, 1830-1892. Bull Hist Med 38:516-529,1964.
47. Bull JP: The historical development of clinical therapeutic trials. J Chronic Dis 10:218-248,
1959.
48. Bulpitt CJ: Randomised Controlled Clinical Trials. Martinus Nijhoff Publishers, London,
1983.
49. Burton BT, Hirschman GH: Demographic analysis: End-stage renal disease and its treatment
in the United States. Clin Nephrol 11:47-51, 1979a.
50. Burton BT, Hirschman GH: Treatment trends of end stage renal disease in the United States.
Artif Organs 3(suppl):29-33, 1979b.
12
12
1
i
2
7
10
12
12
20
8
12
19
19
19
19
19
1
14
2
A
9
2
18
10
20
2
1
18
7
7
C 433
I. Combined bibliography
432
51. Buyse ME, Staquet MJ, Sylvester RJ (editors): Catmcer Clinical Dials: Methods and Practices.
Oxford University Press, Oxford, 1984.
52.
Canner PL: Monitoring clinical trials data for evidence of adverse or beneficial treatment
effects (in Essais Controles Multicentres: Principles et P/obkmes ed‘ted by JP Boissel CR
Klimt). Institut National de la Same et de la Recherche Medicale (INSERM) 76.131-149,
1977a.
53. Canner PL: Monitoring treatment differences in long-term clinical trials. Biometrics 33:603615, 1977b.
54. Canner PL: Comment on “Statistical inference from clinical trials: Choosing the right p
value." Controlled Clin Trials 4:13-17, 1983a.
55. Canner PL: Further comment on “Statistical inference from clinical trials: Choosing the right
p value.” Controlled Clin Trials 4:35, 1983b.
56. Canner PL, Krol WF, Forman SA: External quality control programs. Controlled Clin Trials
4:441-466, 1983c.
57. Carleton RA, Sanders CA, Burack WR: Heparin administration after acute myocardial
infarction. N Engl J Med 263:1002-1005, 1960.
58. Casagrande JT, Pike MC, Smith PG: The power function of the “exact” test for comparing
two binomial distributions. Appl Statist 27:176-180, 1978.
Cascinelli N, Davis HL Jr, Flamant R, Kenis Y, Lalanne CM, Muggia FM, Rozencweig M,
59.
Staquet MJ, Veronesi U: Methods and Impact of Controlled Therapeutic Trials tn Cancer
(part II). Union Internationale Centre le Cancer Report Series (vol 59), Geneva, 1981.
60. Cassel GH, Ferris FL III: Site visits in a multi-center ophthalmic clinical trial. Controlled Clin
I
illlKI
' • .. I
c
'IS
75. Cochran WG: Some methods of strengthening the common X2 tests. Biometrics 10:417-451,
9, 18
6
1957.
77. Collen MF, Cutler JL, Siegelaub AB, Celia RL: Reliability of a self-administered medical
questionnaire. Arch Intern Med 123:664-681, 1969.
78 Collins JF, Bin8h.m SF, Weiss DG, Williford WO, Kuhn RM: Some adaptive strategies for
inadequate sample acquisition in Veterans Administration Cooperative Clinical Trials. Con
20
20
20
16
8
9
7
14, 16
4, 7
287:75-78, 1972.
62. Chalmers TC: The impact of controlled trials on the practice of medicine. Mt Sinai J Med
7
41:753-759, 1974.
Chalmers TC: Randomization of the first patient. Med Clin North Am 59:1035-1038, 1975.
63.
64. Chalmers TC, Matta RJ, Smith H Jr, Kunzler A: Evidence favoring; the use of anticoagulants
in the hospital phase of acute myocardial infarction. N Engl J Med 297.1091 1096. 1977.
65 Chalmers TC, Silverman B, Shareck EP, Ambroz A, Schroeder B, Smith H Jr: Randomized
control trials in gastroenterology with particular attention to duodenal ulcer (in Report to the
Congress of the United States of the National Commission on Digestive Diseases, Vol IV:
Reports of the Workgroups: Part 2B: Subcommittee on Research-Targeted and NonDirected, pp. 223-255). Publ no. NIH 79-1885, National Institute of Arthritis, Metabolism,
and Digestive Diseases, Bethesda, Md, 1978.
66. Chalmers TC, Smith H Jr, Blackburn B, Silverman B, Schroeder B, Reitman D, Ambroz A: A
method for assessing the quality of a randomized control trial. Controlled Clin Dials 2.31-49,
1981
*il
8
76. Cochran WG, Cox GM: Experimental Designs (2nd ed). John Wiley and Sons, Inc, New Yor ,
20
THals 5:251-262, 1984.
61. Chalmers TC, Block JB, Lee S: Controlled studies in clinical cancer research. N Engl J Med
67. Chalmers TC: Informed consent, clinical research and the practice of medicine. Dans Am Clin
:;z
73 Cobb LA, Thomas GI, Dillard DH, Merendino KA, Bruce RA: An evaluation of internal
mammary-artery ligation by a double-blind technic. N Engl J Med 260:1115-1118, 1959.
74. Cocco AE, Cocco DV: A survey of cimetidine prescribing. N Engl J Med 304:1281, 1981.
1954
C
t
18
trolled Clin Trials 1:227-248, 1980.
79. Colsky J: Clinical investigator: The clinical investigator and evaluaion of new drugs. Am J
Hosp Pharm 20:517-519, 1963.
80. Colton T: A model for selecting one of two medical treatments. JASA 58:388-400, 1963.
81. Commission on Professional and Hospital Activities: Diabetes mellitus: Use of oral antidiabet
ics. PAS Reporter 10:263, Nov 24, 1972.
82. Commission on Professional and Hospital Activities: Hospital Record Study: 1969-1975.
Joint publications of CPHA and IMS America, Ltd, Ann Arbor, Mich, 1976.
83. Committee for the Assessment of Biometric Aspects of Controlled Trials of
-™1'
Agents: Report of the Committee for the Assessment of Biometric Aspects of Controlled Trials
of Hypoglycemic Agents. JAMA 231:583-608, 1975.
84. Committee on Drugs: Unapproved use of approved drugs: The physician, the package insert,
and the FDA. Pediatrics 62:262-264, 1978.
85. Committee on Lipoproteins and Atherosclerosis Technical Group and the Co"’™««e on
Lipoproteins and Atherosclerosis of the National Advisory Heart Council: Evalua ion of
serum lipoprotein and cholesterol measurements as predictors of climca complications of
atherosclerosis: Report of a cooperative study of lipoproteins and atherosclerosis. Circulation
14:691-741, 1956.
86 Coordinating Center Models Project Research Group: Coordinating Center Models Project:
A Study of Coordinating Centers in Multicenter Clinical Trials: I. Design and
(l" lJO
parts). Division of Heart and Vascular Diseases, National Heart, Lung, and Blood Institute,
8
7
2
A
7, 14
Climatol Assoc 94:204-212, 1982a.
68. Chalmers TC: A potpourri of RCT
Rd topics.
topics, Controlled
tontrottea Clin
Mtn Dials
mats 3:285-298, 1982b.
8
69. Chandor A: The Facts on File Dictionary of Microcomputers. Facts on File, Inc, New York, A
70. Charen T: MEDLARS Indexing Manual. No. PB 271-306, National Technical Information
2
Service, Springfield, Va. 1977.
71 Chlebowski RT, Weiner JM, Ryden VMJ, Bateman JR: Factors influencing the interim
interpretation of a breast cancer trial: Danger of achieving the “expected result. Controlled
19, 20
Clin Trials 2:123-132. 1981.
72. Christiansen I, Iverson K, Skouby AP: Benefits obtained by the introduction of a coronary
care unit: A comparative study. Acta Med Scand 189:285-291, 1971.
2
Bethesda, Md, March, 1979a.
87 Coordinating Center Models Project Research Group: Coordinating Center Models Project:
A Study of^Coordinating Centers in Multicenter Clinical Trials: IL RFPsfor Coordtmtmg
Centers: A Content Evaluation. Division of Heart and Vascular Diseases, National Heart,
Lung, and Blood Institute, Bethesda, Md, March, 1979b.
88 Coordinating Center Models Project Research Group: Coordinating Center Models Project.
A Smc^ of Coordinating Centers in Multicenter Clinical Trials. IV. Termino ogy^ Division of
Heart and Vascular Diseases, National Heart, Lung, and Blood Institute, Bethesda, Md, Aug,
1979C
9, 10
12
6, 14
1
9
7
7
7, 10,
16, 19,
24
6
5, A
3,21
A
j
89 Coordinating Center Models Project Research Group: Coordinating Center Models Project:
A Study of Coordinating Centers in Multicenter Clinical Dials: VI. Phases of a M^ncen e
Trial. Division of Heart and Vascular Diseases, National Heart, Lung, and Blood Institute,
Bethesda, Md, Sept, 1979d.
90 Coordinating Center Models Project Research Group: Coordinating Center Models Project:
Centers in Multicenter Clinical Trials: XIV. Enhancement of Meth
odological Research in the Field of Clinical Dials. Division of Heart and Vascular Diseases,
National Heart, Lung, and Blood Institute. Bethesda, Md. Sept. 1979e.
91 Cornfield J: Principles of research. Am J Ment Defic 64:240-252, 1959.
92. Cornfield J: A Bayesian test of some classical hypotheses-With applications to sequent.al
8
9, 20
clinical trials. JASA 61:577-594, 1966a.
93. Cornfield J: Sequential trials, sequential analysis and the likelihood principle. Am Statistician
20, A
20:18-23, 1966b.
1
435
D
434
1, 20, A
116 Cuppies LA, Heeren T, Sch.tzkin A, Colton T: Multiple testing of hypotheses in comparing
two groups (review). Ann Intern Med 100:122-129, 1984.
117. Curran WJ, Shapiro ED: Law. Medicine, and Forensic Science (2nd ed). Little, Brown and
7, 19
Co. Boston, 1970.
.
118. Cutler SJ, Ederer F: Maximum utilization of the life table method in analyzing survival.
I. Combined bibliography
95. Cornfield J: The University Group Diabetes Program: A further statistical analysis o t e
mortality findings. JAMA 217:1676-1687. 1971.
Communicative Disorders and Stroke, Bethesda. Md. 197..
97. Cornfield J: Recent methodological contributions to clinical trials. Am J Epidemiol 104:408-
98 Coronary Artery Surgery Study Research Group: National Heart, Lung and Blood ’nst’^
r nnarv Arterv Surgery Study: A multicenter comparison of the effects of randomized
medical^nd surgical treatment of mildly symptomatic patterns with coronary artery^disease,
and a registry of consecutive patients undergoing coronary angiography (edited by T K p,
LD Fisher, MB Mock). Circulation 63(suppl I):I-I—1-81. 1W1
2, 6, 10,
11, 12,
19, 24
•assssxssssssaa:
2
14
18
J Chronic Dis 8:699-712, 1958.
20
20
20
D
119. Dawkins HC: Multiple comparisons misused: Why so
studies (commentary)? Biometrics 39:789-790. 1983.
frequently used in response-curve
20
120 Day HW: Effectiveness of an intensive coronary care area. Am J Cardiol 15:51-54, 1965.
2
8
A multicenter trial. Practitioner 224.105-107, 1980.
122. DeBakey ME, Lawrie GM: Aortocoronary-artery bypass: Assessment after 13 years.
2
239:837-859, 1978.
123. DeMets DL, Ware JH: Group sequential methods for clinical trials with a one-s.ded hypothe
A
sis. Biometrika 67:651-660, 1980.
.
survival in randomized patients and nonrandomized patients meeting randomization criteria. J
Am Coll Cardiol 3:114-128, 1984.
101 Coronary Drug Project Research Group: Coronary Drug Project ^ata and Safety
’ Report ^L Conner, University of Maryland School of Medicine, Baltimore, Feb 1. 19 0a.
0
1972
104. Coronary Drug Project Research Group: The Coronary Drug Project: Design, methods, and
baseline results. Circulation 47(suppl I):l-1—1-50, 1973a.
105. Coronary Drug Project Research Group: The Coronary»
107. C^X’orug Project Research Group: Clofibrate and niacin in coronary heart disease.
1
! "i'!1'
11*
•••t, I
II
8
t25 Department of Health. Eduction .nd Wdf.re: The Negotiated Contracting Process:A Guide
20, 24
for Project Officers. Washington, Oct, 1977.
126. Department of Health and Human Services: Administration of Grants: Federal Regulations,
4, 18,
19, 20,
24
1, 3, 6,
8-12,
14-16,
19, 21,
23-25
18, 20,
24
10, 18,
24
9,11,15,
18, 24
,08. ctr».D Xprojeri7Rese.rch Group: Aspirin in coronary heart disease. J Chronic Dis 3, 10,
19, 24
19, 20,
24
18
lj n. Cox DR: XXZdstd Hfe table?'(Zdingdi'-sion). JRSS 34(series B):187-202,
18, A
"
ssrssSST
'•<1
114. ccnna: Scientist and Philosopher, edited by GM Wickens). Luzac and Co, Ltd. London 952.
Title 45, Part 74. Washington, June 9, 1981.
. .
127 Department of Health and Human Services: Health Care Financing Program Statistics: The
Medicare and Medicaid Data Book, 1981. Publ no. 03128, Hea th Care Fmanctng Admmistration Office of Research and Demonstrations, Baltimore, April, IMa.
f 28. Department of Health and Human Services: Public Heahh ^iee Grants Policy Statement.
Publ no. (OASH) 82-50,000 (rev), Washington, Dec 1, 1982b.
129. Devan WJ, Brown MB (editors): Biomedical Computer Programs. P-Senes. University of
California Press, Los Angeles, Calif, 1979.
130. Diabetic Retinopathy Study Research Group: Preliminary report on effects of pholocoagula-
tion therapy. Am J Ophthalmol 81:383-396, 1976.
131 ^r£S
85:82-106, 1978.
n
_ .
,32. Diabetic Retinopathy Study Research Group: Diabetic ^’"OP"^
R'Port DeS,g"’
methods, and baseline results. Invest Ophthalmol Vis Sci 21.149-209, 1981.
133 Diehl HS, Baker AB, Cowan DW: Cold vaccines: An evaluation based on a controlled study.
111:1168-1173, 1938.
134. Dimond EG, Kittle CF, Crockett JE: Comparison of internal mammary artery ligation and
sham operation for angina pectoris. Am J Cardiol 5.483-486, 1960.
135. Dixon WJ (editor): BMDP Statistical Software (\98\ ed). University of California Press, Los
17, A
17
19, 20
19
13
1
8
17
136 Donaldson
Donaldson B: How Did They Die? St. Martins Press, Inc New York, 1980.
137. Draper N, Smith H: Applied Regression Analysis (2nd ed). John Wiley and Sons, New York,
2
18
138. Drummond JC, Wilbraham A: The Englishman's Food: A History of Five Centuries of
1
English Diet. Jonathan Cape, London, 1940.
139 Duncan DB: Multiple range and multiple F tests. Biometrics 11:1-42, 1955.
140. Duncan DB: t tests and intervals for comparisons suggested by the data. Biometrics 31.33 -
20
20
359, 1975.
Annual Meeting, Pittsburgh, May 3, 1982.
A
F 437
436
I. Combined bibliography
for differences between unequally
141. Duncan DB, Godbold JH: Approximate Ik-ratio t tests
replicated treatments. Biometrics 35:749-756. 1979.
39-787-794
,42 Dunean DB, Brandt L: Adaptive t tests for mult.ple compansons. B.ometncs
20
20
20
„ xri?™•'
““ ■*
Years 1846 1900. Oxford University Press, London. 1947.
for comparing several treatments with a
146 Dunnett CW: A multiple comparison procedure f_
’ control. J/l 5/1 50:1096-1121, 1955.
with a control. Biometrics 20:482-491,
147. Dunnett CW: New tables for multiple compansons
X- ™. «-■««“■
■*-'D” “
12
1
9, 20
require the
Xf ™■— - - - -- »*■ -
164. Fibiger J: Om serumbehandlung af difteri. Hospitalstidende 6.309-325, 1898.
165 Finkel MJ: The development of orphan products. N Engl J Med 307:963-964, 1982.
166. Finney HC: Improving the reliability of retrospective survey measures: Results of a longitudi
1
6
12
nal field survey. Evaluation Rev 5:207-229, 1981.
167. Fisher RA, Mackenzie WA: Studies in crop variation: II. The manurial response of different
1
potato varieties. J Agric Sci 13:311-320, 1923.
168. Fisher RA: The arrangements of field experiments. J Ministry Agric Great Bntatn 33:503-513,
1
169. Fisher RA: Statistical Methods for Research Workers (10th ed). Oliver and Boyd, Edinburgh,
18
170 Hsher R A, Yates F: Statistical Tobies for Biological. Agricultural and Medical Research (6th
ed). Hafner Publishing Company, Inc, New York, 1963.
171. Fisher RA: Statistical Methods and Scientific Inference (3rd ed). Hafner Press, Macmillan
10
20
Publishing Co, Inc, New York, 1973.
172. Fleiss JL: Statistical Methods for Rates and Proportions (2nd ed). John Wiley and Sons, New
9, 18
9. 20, A
York, 1981.
,
.
173. Fletcher RH, Fletcher SW, Wagner EH: Clinical Epidemiology: The Essenltals. Williams and
7
20
Wilkins, Baltimore, 1982.
174. Food and Drug Administration: Procedural and interpretative regulations: Invest.gational
.
Controlled Clin Trials 4:27-33, 1983b.
use. Federal Register 28:179-182, Jan 8, 1963.
175 Food and Drug Administration: Hearing procedure for refusal or ^jWrawal of aPPr°va’£
3
National Eye Institute, Bethesda, Md. 1982.
n„Ktion„aire Percept Mol Skills
151. Edv.rdsson B: Effect of reversal of response scales tn a questtonnatre.
P
Register 34:14596-14598, Sept 19, 1969a.
176 Food and Drug Administration: Novobiocin-tetracycline combination drugs: Calcium novobioci^sXmethizole tablets. Final order repealing regulations and working certificates. Federal
12
Register 34:14598-14599, Sept 19, 1969b.
177 Food and Drug Administration: Hearing requests on refusal or withdrawal of New Drug
Applications and issuance, amendment or repeal of Antibiotic DruJ
ing scientific content of adequate and well-controlled clinical investigations. Federal Register
10
18
of drugs for unlabeled
35:3073-3074, Feb 17, 1970a.
2
6
. _
and regulations describing scientific
1
May 8, 1970b.
,v.
179. Food .nd Drug Administration: Oral hypoglycemic agents; Diabetes prescnbmg mformatton.
7
FDA Current Drug Information, Oct, 1970c.
180 Food and Drug Administration: Oral hypoglycemic agents. FDA Drug Bull, June 23, 19
24
24
2, 24
market. FDA Drug Bull 7:6-7, May-July, 1977a.
185. Food and Drug Administration: HEW Secretary suspends general marketing of phenformin.
""rge°y in «abie'angina pectoris: A progress report on survival. Circulation 65(suppl
I
' tr i
ifel
7
7
.
181 Food and Drug Administration: Oral hypoglycemic drug labeling. FDA Drug Bull. May.
1972a.
182 Food and Drug Administration: DESI Who? FDA Consumer, Oct. 1972b.
183. Food and Drug Administration: Drug labeling: Failure to reveal material facts; labeling of oral
hypoglycemic drugs. Federal Register 40:28582 28595, July ,
184 Food and Drug Administration: Phenformin: New labeling and possible removal from e
24
I
1
178
24
15
1
|60-
FDA Drug Bull 7:14-16, Aug, 1977b.
12
- “-
A Jo. r„
,63 Fener W: An m.roduc.ion ,0 ProbMi.y Theory and Us Appbeanons
Wiley and Sons, New York, 1968.
Group
3rd ed). John
7, 19, 26
2
7
7
7
7, A
186. Food .nd Drug Administration:
Publ no. HEW (FDA) 77-3040, Rockville, Md, Sept. 1977c.
187. Food .nd Drug Administration: Status of withdrawal of phenformm. FDA Drug Bull 7.19- 7
20. Sept-Oct. 1977d.
A
20, A
52734, Nov 14, 1978.
.
,
7, 16, 24
H
438
189. Food and Drug Administration: Medical devices: Procedures for Investigational Device
Exemptions. Federal Register 45:3732-3759, Jan 18, 1980.
190. Food and Drug Administration: Clinical Testing for Safe and Effective Drugs. Investigational
Drug Procedures. HHS publ no. (FDA) 74-3015, Rockville. MD, 1981.
191. Food and Drug Administration: Regulatory Requirements for Medical Devices: A Workshop
Manual. HHS publ no. (FDA) 83-4165, Rockville, Md, June, 1983.
192. Food and Drug Administration: Guideline Labeling for Oral Hypoglycemic Drugs of the
Sulfonylurea Class. Docket no. 83D-0304, Rockville, Md, March 16, 1984a.
193. Food and Drug Administration: Labeling of oral hypoglycemic drugs of the sulfonylurea class.
Federal Register 49:14303-14331, April II, 1984b.
194. Food and Drug Administration: Oral hypoglycemic drug products: Availability of guideline
labeling. Federal Register 49:14441-14442, April II, 1984c.
195. Food and Nutrition Board: Toward Healthful Diets. National Academy of Science, Washing
ton, May 28, 1980.
196. Francis T, Korns RF, Voight RB, Boisen M, Hemphill FM, Napier JA, Tolchinsky E: An
evaluation of the 1954 poliomyelitis vaccine trial: Summary report. Am J Public Health
45(part II, May suppl): 1-51, 1955.
197. Fredrickson DS: The field trial: Some thoughts on the indispensable ordeal. Bull NY Acad
Med 44:985-993, 1968.
198. Freedman LS, White SJ: On the use of Pocock and Simon’s method for balancing treatment
numbers and prognostic factors in the controlled clinical trial. Biometrics 32:691-694, 1976.
199. Freiman J A. Chalmers TC, Smith H Jr, Kuebler RR: The importance of beta, the type II error
and sample size in the design and interpretation of the randomized control trial: Survey of 71
“negative” trials. N Engl J Med 299:690-694, 1978.
200. Frenette C, Begin G: Sensitive data gathering through the random response technique: A
validity study. Psychol Rep 45:1001-1002, 1979.
201. Friedman LM, Furberg CD, DeMets DL: Fundamentals of Clinical Thais. John Wright, PSG
Inc, Boston, 1982.
G
202. Gail M, Gart JJ: The determination of sample sizes for use with the exact conditional test in
2x2 comparative trials. Biometrics 29:441-448, 1973.
203. Gail M: Power computations for designing comparative Poisson trials. Biometrics 30:231-237,
1
'
439
/. Combined bibliography
1974.
204. Gail M: Applicability of sample size calculations based on a comparison of proportions for use
with the logrank test. Controlled Clin Trials 6:112-119, 1985.
205. Galton F: Regression towards mediocrity in hereditary stature. Anthropological Institute
15:246-263, 1886.
206. Garrett HE, Dennis EW, DeBakey ME: Aortocoronary bypass with saphenous vein graft:
Seven-year follow-up. JAMA 223:792-794, 1973.
207. Gart JJ: The comparison of proportions: A review of significance tests, confidence intervals
and adjustments for stratification. Review Internal Stat Inst 39:148-169, 1971.
208. Gordis L, Naggan L, Tonascia J: Pitfalls in evaluating the impact of coronary care units on
mortality from myocardial infarctions. Johns Hopkins Med J 141:287-295, 1977.
209. Grant AP: Sequential trial of isocarboxazid in angina pectoris. Br Med J 1:513-515, 1962.
210. Greenberg BG (chairman): A Report from the Heart Special Project Committee to the
National Advisory Heart Council: Organization. Reviews and Administration of Cooperative
Studies (unpublished). National Heart Institute, Bethesda, Md, May, 1967.
211. Greenhouse SW, Halperin M: Jerome Cornfield: 1912-1979. Controlled Clin Trials 1:75 80,
1980
212. Greenhouse SW: Discussion on early stopping. Controlled Clin Thais 4:23-25, 1983.
213. Grizzle JE: A note on stratifying versus complete random assignment in clinical trials.
Controlled Clin Trials 3:365-368, 1982.
H
13
13
13, A
7
7
7
22
I
2,3
10
2,9
12
10
9
9
9
A
2
9
2
9
1
1
20
10
214. Hagans J: The design and methodology of cooperative drug trials. Drug Intell Clin Pharm
8:531-534, 1974.
215. Haggard HW: The Lame, the Halt, and the Blind: The Vital Role of Medicine in the History of
Civilization. Harper and Brothers, New York, 1932.
216. Hainline A Jr, Miller DT, Mather A: Role and methods of the central laboratory. Controlled
Clin Trials 4:377-387, 1983.
217. Halperin M, Rogot E, Gurian J, Ederer F: Sample sizes for medical trials with special reference
to long-term therapy. J Chronic Dis 21:13-24, 1968.
218. Hamilton MP: Role of an ethicist in the conduct of clinical trials in the United States.
Controlled Clin Trials 1:411-420, 1981.
219. Harville DA: Experimental randomization: Who needs it? Am Statistician 29:27-31, 1975.
220. Haseman JK: Exact sample sizes for use with the Fisher-Irwin test for 2x2 tables. Biometrics
34:106-109, 1978.
221. Hauck WW: Letter to the editor concerning random number generators. Controlled Clin
Trials 3:73, 1982.
222. Haupt BJ: Deliveries in short-stay hospitals: United States, 1980. Vital and Health Statistics of
the National Center for Health Statistics, no. 83, Hyattsville, Md, Oct 8, 1982.
223. Haverkamp AD, Thompson HE, McFee JG, Cetrulo C: The evaluation of continuous fetal
heart rate monitoring in high-risk pregnancy. Am J Obstet Gynecol 125:310-320, 1976.
224. Haverkamp AD, Orleans M, Langendoerfer S, McFee J, Murphy J, Thompson HE: A
controlled trial of the differential effects of intrapartum fetal monitoring. Am J Obstet
Gynecol 134:399^12, 1979.
225. Hawkins BS, Canner PL: Impact of closeout on operations at two coordinating centers. The
Johns Hopkins University, Baltimore. Presented at the 4th Annual Symposium on Coordinat
ing Clinical Trials, Arlington, Va, May 25-26, 1978.
226. Hawkins BS: Staffing the coordinating center: Who is minding the data? (in Coordinating
Center Models Project: A Study of Coordinating Centers in Multicenter Clinical Trials: XVI.
CCMP Manuscripts Presented at the Annual Symposia on Coordinating Clinical Trials, pp.
155-187). Division of Heart and Vascular Diseases, National Heart, Lung, and Blood Institute,
Bethesda, Md, June, 1979.
227. Hawkins BS: Perusing the literature. Controlled Clin TYials 1:71-74, 181-185, 269-274, 1980.
228. Hawkins BS: Perusing the literature. Controlled Clin Trials 2:51-57, 257-265, 327-333, 1981.
229. Hawkins BS: Perusing the literature. Controlled Clin Thais 3:371-383, 1982.
230. Hawkins BS: Perusing the literature. Controlled Clin Thais 4:75-86, 239-253, 1983.
231. Hay garth J: Of the Imagination, as a Cause and as a Cure of Disorders of the Body:
Exemplified by Fictitious Tractors, and Epidemical Convulsions. R Cruttwell, Bath, 1800.
232. Helsing KJ, Comstock GW: Response variation and location of questions within a question
naire. Internal J Epidemiol 5:125-130, 1976.
233. Hill AB: Statistical Methods in Clinical and Preventive Medicine. Oxford University Press,
New York, 1962.
234. Hill AB: Principles of Medical Statistics (9th ed). Oxford University Press, New York, 1971.
235. Hill JD, Holdstock G, Hampton JR: Comparison of mortality of patients with heart attacks
.
. .
_ .. j ________ i:____________ i:__ 1
DQI
lu//
admitted
to a coronary care unit and
an ordinary medical ward. Br
Med J7 0.
2:81-83,
19//.
236. Hill JD, Hampton JR, Mitchell JRA: A randomised trial of home-versus-hospital manage
ment for patients with suspected myocardial infarction. Lancet 1:837-841, 1978.
237. Himmelfarb S, Edgell SE: Additive constants model: A randomized response technique for
eliminating evasiveness to quantitative response questions. Psychol Bull 87:525-530, 1980.
238. Hochstim JR, Renne KS: Reliability of response in a sociomedical population study. Public
Opinion Quart 35:69-79, 1971.
239. Holland WW, Ashford JR, Colley JRT, Morgan DC, Pearson NJ: A comparison of two
respiratory symptoms questionnaires: 1. Methodology and observer variation. Br J Prev Soc
Med 20:76-96, 1966.
5, 10
2
16
9
23
19
9
10
2
2
2
15
5
26
26
26
26
I, 2
12
1
5
2
2
12
12
12
1
262. King LS: The blood-letting controversy: A study in the scientific method. Bull Hist Med 35.1
240. Holman DV: Venesection before Harvey and after. Bull NY Acad Med 31:661-670, 1955.
241. Howard JM, DeMets D, and the BHAT Research Group: How informed is informed consent?
The BHAT experience. Controlled Clin Trials 2:287-303, 1981.
242. Hypertension Detection and Follow-Up Program Cooperative Group: Five-year findings of
the Hypertension Detection and Follow-Up Program: I. Reduction in mortality of persons
with high blood pressure, including mild hypertension. JAMA 242:2562-2571, 1979a.
243. Hypertension Detection and Follow-Up Program Cooperative Group: Therapeutic control of
blood pressure in the Hypertension Detection and Follow-Up Program. Prev Med 8:2-13,
1979b.
I
244. IMS America, Ltd (Ambler, Pa): Diabetic therapy. National Disease Therapeutic Review 8:7,
June, 1977.
245. Ingelfinger JA, Mosteller F, Thibodeau LA, Ware JH: Biostatistics in Clinical Medicine.
Macmillan Publishing Co, New York, 1983.
246. International Committee of Medical Journal Editors: Uniform requirements for manuscripts
submitted to biomedical journals. Br Med J 284:1766-1770, 1982.
J
247. James G, James RC (editors): Mathematics Dictionary (multilingual ed). D Van Nostrand
Company, Inc, Princeton, NJ, 1959.
E: An Inquiry into the Causes and Effects of the Variolae Vaccinae. Sampson Low,
248. Jenner
.
London, 1798.
I
249. Jones JH: Bad Blood: The Tuskegee Syphilis Experiment. The Free Press, New York, 1981.
250. ___
JuhlE, Christensen E, Tygstrup N: The epidemiology of the gastriontestinal randomized
clinical trial. N Engl J Med 296:20-22, 1977.
K
251. Kalbfieisch JD, Prentice RL: The Statistical Analysis of Failure Thime Data. John Wiley and
Sons, New York, 1980.
252. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J/1S/4 53.457
2
,3’ 1961
8, 14
1, 11, 14
8
II'i'isl
I i!
253. Keefer CS, Blake FG, Marshall EK Jr, Lockwood JS, Wood WB Jr: Penicillin in the treatment
of infections: A report of 500 cases. JAMA 122:1217 1224, 1943.
254. Keen H, Jarrett RJ: The effect of carbohydrate tolerance on plasma lipids and atherosclerosis
in man (pp 435-444, 446-449 in Atherosclerosis, RJ Jones, editor). Springer-Verlay, New
York, 1970.
255. Keen H: Factors influencing the progress of atherosclerosis in the diabetic. Acta Diubet Lot
8(suppl l):444-456, 1971.
Kelsey FO: Government: The Kefauver-Harris Amendments and investigational drugs. Am J
256. I
Hosp Pharm 20:515-517, 1963.
257. Kelso IM, Parsons RJ, Lawrence GF, Arora SS, Edmonds DK, Cooke ID: An assessment of
continuous fetal heart rate monitoring in labor: A randomized trial. Am J Obstet Gynecol
131:526-532, 1978.
Kendall MG, Buckland WR: A Dictionary of Statistical Terms (2nd ed). Hafner Publishing
258.
Company, New York, 1960.
259. Kenton C, Scott YB: MEDLINE searching and retrieval. Med Inf 3:225-235, 1978.
260. Kidder LH: Selltiz, Wrightsman and Cook's Research Methods in Social Relations (4th ed).
Holt, Rinehart and Winston, New York, 1981.
261. Kilo C, Miller JP, Williamson JR: The achilles heel of the University Group Diabetes
Program. JAMA 243:450-457, 1980.
<-.
•
2
c •
263. Klein E: A Comprehensive Etymological Dictionary of the English Language. Elsevier cientific Publishing Co, Amsterdam, 1971.
264. Kleinbaum DG, Kupper LL, Chambless LE: Logistic regression^analysis of epidemiologic
data: Theory and practice. Commun Statist-Theor Meth 11.485-547, 1982.
265. Klimt CR, Canner PL: Terminating a long term clinical trial. Clin Pharmacol Ther 25:641646, 1979.
266. Klimt CR: Terminating a long-term clinical trial. Controlled Clin Thais 1:319-325, 1981.
267. Knatterud GL: Lessons learned in monitoring for treatment effects in the Diabetic Retino
pathy Study (abstract). Am J Epidemiol 106:247, 1977.
268. Knatterud GL: Methods of quality control and of continuous audit procedures for controlled
7
clinical trials. Controlled Clin Trials 1:327-332, 1981.
269. Knatterud GL, Forman SA, Canner PL: Design of data forms. Controlled Clin Trials 4.429-
18
25
A
1
14
4
18
440, 1983.
■
.
270.
270. Knox
Knox JHM
JHM Jr:
Jr: The
The medical
medical history of George Washington, his physicians, friends and
advisers. Bull Inst Hist Med 1:174-191, 1933.
271. Knox RA: The pill for diabetics. Boston Sunday Globe, Nov 7, 1971.
272. Knuth DE: Seminumerical Algorithms (vol 2 of The Art of Computer Programming). Addi
I, A
18
15
15
8
3
12
2
7
A
son-Wesley Publishing Co, Reading, Mass, 1969.
273. Kolata GB: Controversy over study of diabetes drugs continues for nearly a decade. Science
7
203:986-990, 1979.
274. Kotz S, Johnson NL, Read CB (editors): Encyclopedia of Statistical Sciences. John Wiley and
A
Sons, New York, 1982.
Krol WF: Closing down the study. Controlled Clin Thais 4:505-412, 1983.
275.
Kruskal WH, Tanur JM (editors): International Encyclopedia of Statistics (m 2 vols). The Free
15
A
276.
Press, Macmillan Publishing Co, Inc, New York, 1978.
18, A
L
277. Lachin JM: Introduction to sample size determination and power analysis for clinical trials.
Controlled Clin Trials 2:93-113, 1981.
278. fst JM (editor); A Dictionary of Epidemiology. Oxford Univent.ty Press. New York 1983.
I
279. Layne BH, Thompson DN: Questionnaire page length and return rate. J Soc Psychol 113.291-
A
12
7
292 1981
280. Ledger RR: Safety of Upjohn’s oral antidiabetic drug doubted in study: Firn, disputes findings.
7
Dala Analysis. Lifetime Learning Publicattons,
18
481. 1958.
P'«ii
441
L
440 I. Combined bibliography
281.
7
Wadsworth Inc, Belmont, Calif, 1980.
,oA8 iQ7?6n
9
2
I
2
edited by FG McMahon). Futura Publishing Co, Mt. Kisco, NY, iv/4.
283 Levine MM: Informed consent (abstract). Am J Epidemiol 104:350. 1976.
284. Levine RJ, Lebacqz K: Some ethical considerations in clinical Inals. Chn Pharmacol The
14
14
A
285. Lev^RJ: X and Regulations of Clinical Research. Urban and Schwarzenberg, Baht-
1, 14
2
12
286. Lkberm^n GJ. Owen DB: Tables of the Hypergeometric Probability Distribution. Stanford
18
University Press, Stanford, Calif, 1961.
A ca.iir
287. Lilienfeld AM: Ceteris paribus: The evolution of the clinical tnal. Bull Hist Med 56.1-18,
1
7, 26
1982.
M 443
442
/. Combined bibliography
19
Medical Research Council: Clinical trials of new remedies (annotations). Lancet 2:304. 1931
310. Medical Research Council: Srieptomycin treatment of pu'monary tuberculosis:
Med.ca
311. Research Council investigation. Br Med J 2.769-782 1948.
312 Meier P: Stratification in the design of a clinical Inal. Controlled Cltn Trta s L355-361. 19«1.
313. Meinert CL: Qua.ity assurance in c.inical trials. Ins.itu. Ntt.tona, de la Sante e. de la
14
Recherche Medicale (INSERM) 76.123 130,197
3!4. Meinert CL: Cost profiles of data coordinatingcenter (in Coordinattn, Center, Modes
I. 8
288.
burgh, 1753.
289. Lindley DV: The Role of Randomization in Inference. Report no. M 627. Department of
Statistics, Florida State University, Tallahassee, 1982.
290. Lipid Research Clinics Program: Recruitment for clinical trials:
Coronary Primary Prevention Trial experience: Its> imphca ton for
(WS Agras,
RH Bradford, GD Marshall, editors). Circulation 66 (suppl IV).IV 1 IV 78. 198
pinlu,
1$
•i'
7, 18
20, 24
292. Macular Photocoagulation Study Group: Argon laser photocoagulation for ocular histoplas 18, 20,
24
mosis: Results of a randomized clinical trial. Arch Ophthalmol 101:1347-1357. 1983a.
18, 24
293. Macular Photocoagulation Study Group: Argon laser Pho‘ocoa,^,a;ion. [ap i'icq PpA 1°
neovascularization. Results of a randomized clinical trial. Arch Ophthalmol 101.1358 1361,
315. Meinert CL: Recommendation5 and.
1983b.
Macular Photocoagulation Study Group: Changing the protocol: A case report from the
294.
Macular Photocoagulation Study. Controlled Clin Trials 5:203-216, 1984.
295. Management Committee of the Australian Therapeutic Trial in Mild Hypertension: The
Australian Therapeutic Trial in Mild Hypertension. Lancet 1:1261-1267. 1980.
296. Mantel N, Haenszel W: Statistical aspects of the analysis of data from retrospective studies of
317. Meinert CL: Clinical trials and data integrity (editorial). Controlled Clin THufe 1:189-192,
1980
306. McDill M: Activity analysis of data coordinating centers (in Coordinating Center M°dels
Project: A Study of Coordinating Centers in Multicenter Clinical Rials: XVI. CCMP^a^'
cripts Presented at the Annual Symposia on Coordinating Clinical Rials, pp 105 125).
Division of Heart and Vascular Diseases, National Heart, Lung, and Blood Institute, Bethesda,
5,6
^coordinating Cemers^Muhicenter
23
Coordinating Centers Models Project.
? Annual Symposia on Coordinating
£c“J
W 257 C2ML D=Mean and VascuUr Diseases, Nationa! Heart, Lung.
“ rstlndardization (editorial). Controlled Clin
316.
1:97-
1, A
16
7
24
18, A
MeTert CL: Organization of multicenter clinical trials. Controlled Clin THo/s 1:305-312,1981
318.
Meinert CL: Funding for clinical trials. Controlled Clin TYials 3:165-171, 1982.
319. Meinert CL, Heinz EC, Forman SA: Role and methods of the coord.nat.ng center. Controlled
320.
321. Mrin^CL, Ton.sci. S, Higgins KH: Content of reports on clinical trials: A critical revtew.
1,4, 23
6
5, 17
2, 25
18
322. MUIrtRG j”'Simutat’us Statistical Inference. McGraw-Hill Book Company. New York.
20
323 Miller RG Jr: Developments in multiple comparisons: 1966-1976. JASA 72-779-’88;
Mi man N, Seheibe! J Jessen 0: Lysine prophylaxis in recurrent herpes srmplex iab.ahs: A
324. double blind, controlled crossover study. Ada Derm Venereal 60.85-87, 1980.
325 Milne JS. Williamson J: Comparison of leaching^—d'UC"0"
angma pectoris by questionnaire. Br J Prey Soc Med 25:105-108, 1971.
20
8
7
8
12
12
327. Mintz M: Di"o“ d^dZ^n^
<"«
leS,S'
2
328' J?„« s/, D.vigntnS1 P^ cfinicaf ’trial°of amigdafin (tae^)
2
th^e'aymenyTh’uman
329 MontgomeZB.Zusesrf Freedom of Information Act. JAMA 242:1007-1009 1979
12
7
7
2
24
A
14
2
3,5
331. Morris RA^slles BD, Berman JJ: Research and the Freedom of Information Act. American
7
332 MortW (Zto’ry
A
'American Heritage Dictionary of the English Language. American
332' hXPublishing Co. inc and Houghton Miffi.n Co. New, York, 19«
333. Moses LE, Oakford RY: Tables of Random Permutations. Stanford Un.vers.ty Press,
334.
Md, June, 1979.
307. McFarland SG: Effects of question order on survey responses. Public Opinion Quart 45:208-
12
215, 1981.
308. McLaughlin GH: SMOG grading—a new readability formula. J Reading 12:639-646, 1969.
309. Meadows AJ: Dictionary of New Information Technology. Nichols Publishing Co, New York,
E
A
1982.
10
6
Project: A study of Coordinating Centers in
■
Ciinicai 7Yials, pp. 127-153).
D^ir^ H^lnd^ZDiSX^a. Heart, Lung, and Blood Institute. Bethesda.
M
disease. J Natl Cancer Inst 22:719-748, 1959.
297. Mantel N: Evaluation of survival data and two new rank order statistics arising in its
consideration. Cancer Chemother Reports 50:163-170, 1966.
298. Market Facts, Inc: The Impact of T\vo Clinical -Rials on Physician Knowledge and Practice^
Final Report. Division of Heart and Vascular Diseases, National Heart, Lung, and Blood
Institute, Bethesda, Md, March, 1982.
299. Marks IN, Wright JP. Denyer M, Garisch JAM, Lucke W: Comparison of sucralfate with
cimetidine in the short-term treatment of chronic peptic ulcers. 5 Afr Med J
198U.
300. Marks RG: Designing a Research Project: The Basics of Biomedical Research Methodology.
Lifetime Learning Publications, Wadsworth Inc, Belmont, Calif, 1982.
301. Martin GL, Newman IM: Randomized response: A technique for improving the validity of
self-reported health behaviors. J Sch Health 52:222-226, 1982.
302 Mather HG, Pearson NG, Read KLQ, Steed GR, Thorne MG, Jones S, Guerrier CJ, Eraut
CD, McHugh PM, Chowdhury NR, Jafary MH, Wallace TJ: Acute myocardial infarction:
Home and hospital treatment. Br Med J 3:334-338, 1971.
303. Mnther HG, Margin DC, Pearson NG, Rend KLQ, Shaw DB, Steed GR, Thome MG,
Lawrence CJ, Riley IS: Myocardial infarction: A comparison between home and hospital care
for patients. Br Med J 1:925-929, 1976.
304. Matts JP, Long JM, and POSCH Group: Modeling recruitment from the population to the
304. Matts
randomized patient (abstract). Controlled Clin Rials 1:169, 1980.
305. McCarn DB: MEDLINE: An introduction to on-line searching. J Am Soc InfSci 31:181-192,
1
335.
JP, McPeek B: Reporting standards and research strategies for controhed
isoniazid, and streptomycin-para-aminosahcyl^
Report on twelve-week observations on 526 patients.
66:632-635, 1952.
10
2,9
P 445
444 I. Combined bibliography
336. Mount FW. Ferehee SH: Con.ro.
21
'
6
DiXion of Research Grants, Bethesda, Md,
33,.
of Program Planning
'
Publ no.
22
33’- SS™"“S«14
«. .9.
ne
34°-
ssarxe?= «-* “*
2
25
267, Bethesda. Md, 1983
Committee on Dimethyl Sulfoxide. Division of
360. National Research Counclit R^^A{ Sciences, Washington, 1973.
», ”
2
9
« - C—■
2
20
A
20
13, 14
w N.tion.l Cancer Institute: Compilo.ion of Experimental Cancer Therapy Protocol Summaries
SI”■
,«ss,: xx™ z c—
clinical
trial. Controlled Clin Trials
The National Cooperative Gallstone Study: A mult.center c------------2:177-299, 1981a.
I’ali
'.,'1
5
iE:
N
2
7
6, 15, A
“ S-XS'SS'S'™ton, Aug, 1983.
5, 11,
12-14,
16,18,19,
20, 23, 24
II, 13,
22, 24
13
analysis obstet Gynecol 58:691-695, 1981.
7
»,
L..,.™ — »—
1510-1590. Paul B Hoeber, New York, 1921.
1
372. WtuHn cXl Trials Committee (of the Medical Research Counc,:il): Clinical trial of Patulin
1
suppl 507:1-82, 1970.
349. Naina. Diet-Heart Stud, Research Group: The National Diet-Heart Stud,: Fina! report.
Circulation 37(suppl 1):I-1—1-428, 1968.
35°-
Pressure
19
7
^for Research 21
in the common cold. iMncet 2:373-375, W
. Press princeton, NJ, 1951.
351. National Institute of Arthritis,Trial on
■■
21
i
2
23
352. Nation.. Institute of Arthritis. Me—;Vaseuiar Compsons of msuhn. Dependent Diahe.es
...... ..........
2, 4,6
12
8
5, 6, 10,
16,2123, 25, E
9,11,18,
24, 25
1
446
377. Peto R, Pike MC, Armitage P, Breslow NE, Cox DR, Howard SV, Mantel N, McPherson K,
Peto J. Smith PG: Design and analysis of randomized clinical trials requiring prolonged
observation of each patient: I. Introduction and design. Br J Cancer 34:585-612, 1976.
378. Peto R, Pike MC, Armitage P, Breslow NE, Cox DR, Howard SV, Mantel N, McPherson K,
Peto J, Smith PG: Design and analysis of randomized clinical trials requiring prolonged
observation of each patient: 11. Analysis and examples. Br J Cancer 35:1-39, 1977.
379. Pines WL: A Primer on New Drug Development. Publ no. 1980-311-254/83, Food and Drug
Administration. Rockville, Md, 1980.
380. Pitman EJG: Significance tests which may be applied to samples from any populations: HI.
The analysis of variance test. Biometrika 29:322-335, 1937.
381. Plackett RL: Current trends in statistical inference. JRSS 129 (series A):249-267, 1966.
382. Pocock SJ, Simon R: Sequential treatment assignment with balancing for prognostic factors in
the controlled clinical trial. Biometrics 31:103-115, 1975.
383. Pocock SJ: Group sequential methods in the design and analysis of clinical trials. Biometrika
64:191-199, 1977.
384. Pocock SJ: Clinical Trials: A Practical Approach. John Wiley and Sons, Chichester, 1983.
385. Presberg J, Timnick L: Federal investigation into heart research project here. St. Louis GlobeDemocrat, May 3. 1976.
386. Prout TE, Knatterud GL, Meinert CL: Diabetes drugs: Clinical trial (letter). Science 204:362363, 1979.
p'Un
|i!P
S
/. Combined bibliography
R
387. Rand Corporation (The): A Million Random Digits with 100,000 Normal Deviates. The Free
Press, Glencoe, Ill, 1955.
388. Rand LI, Knatterud GL: Certification of clinics and staff in a multicenter clinical trial.
University of Maryland, Baltimore. Presented at the Combined Annual Scientific Sessions of
the Society for Clinical Trials and the 7th Annual Symposium for Coordinating Clinical Trials,
Philadelphia, May 6-8, 1980.
389. Ray AA (editor): SAS User’s Guide: Statistics (1982 ed). SAS Institute Inc, Cary, NC, 1982.
390. Reiman AS: Publications and promotions for the clinical investigator. Clin Pharmacol Ther
25:673-676, 1979.
391. Reiman AS: Closing the books on Laetrile. N Engl J Med 306:236, 1982.
392. Remington RD: Problems of university-based scientists associated with clinical trials. Clin
Pharmacol Ther 25:662-665, 1979.
393. Renou P, Chang A, Anderson I, Wood C: Controlled trial of fetal intensive care. Am J Obstet
Gynecol 126:470-476, 1976.
394. Research Development Committee, Society for Research and Education in Primaryf Care
Internal Medicine^RH Fletcher, chairman): Clinical research methods: An annotated bibliog
raphy. Ann Intern Med 99:419-424, 1983.
395. Rheumatic Fever Working Party of the Medical Research Council of Great Britain and the
Subcommittee of Principal Investigators of the American Council of Rheumatic Fever and
Congenital Heart Disease, American Heart Association: The evolution of rheumatic heart
disease in children: Five year report of a cooperative clinical trial of ACTH, cortisone and
aspirin. Circulation 22:503-515, 1960.
396. Robbins H: Some aspects of the sequential design of experiments. Bull Am Mathematics Soc
58:527-535, 1952.
397. Robbins H: A sequential decision problem with finite memory. Proc Natl Acad Set 42:920923, 1956.
398. Romm FJ, Hulka BS: Developing criteria for quality of care assessment: Effect of the Delphi
technique. Health Serv Res 14:309-312, 1979.
399. Roper FW, Boorkman J: Introduction to Reference Sources in the Health Sciences. Medical
Library Association, Inc, Chicago, Ill, 1980.
10, 18
18
400. Roth A, Klassen D, Lubin B: Effects of follow-up procedures on survey results. Psychol Rep
47:275-278, 1980.
401. Rothman DJ: Were Tuskegee and Willowbrook “Studies in Nature”? Hastings Center Report
12:5-7, April 1982.
402. Royall RM: Discussion of Dupont’s “Statistical inference from clinical trials: Choosing the
right p value.” Controlled Clin Trials 4:23-25, 1983.
447
12
14
20
A
S
8
9
10
A
10, 18
19
7
10
3
17
24
2
4, 24
2
26
I
A
A
12
26
403. Sackett DL, Gent M: Controversy in counting and attributing events in clinical trials. N Engl J
Med 301:1410-1412, 1979.
404. Sackett DL: The competing objectives of clinical trials. N Engl J Med 303:1059-1060, 1980.
405. Schade RR, Donaldson RM: How physicians use cimetidine: A survey of hospitalized patients
and published cases. N Engl J Med 304:1281-1284, 1981.
406. Scheffe H: A method for judging all contrasts in the analysis of variance. Biometrika 40:87104, 1953.
407. Schlesselman JJ: Case-Control Studies: Design, Conduct, Analysis. Oxford University Press,
New York, 1982.
408. Schmeck HM Jr: Scientists wary of diabetic pill: FDA study indicates oral drug may be
ineffective. New York Times, May 22, 1970.
409. Schor S: The University Group Diabetes Program: A statistician looks at the mortality results.
JAMA 217:1671-1675, 1971.
410. Schriesheim CA: The effect of grouping or randomizing items on leniency response bias.
Educat Psychol Measurement 41.401 -411, 1981.
411. Schuman SH, Olansky S, Rivers E, Smith CA, Rambo DS: Untreated syphilis in the male
Negro: Background and current status of patients in the Tuskegee Study. J Chronic Dis 2:543558, 1955.
412. Schwartz D, Lellouch J: Explanatory and pragmatic attitudes in therapeutic trials. J Chronic
Dis 20:637-648, 1967.
413. Schwartz TB: The tolbutamide controversy: A personal perspective. Ann Intern Med 75:303306, 1971.
414. Sciotti H, Sciotti S, Zogbi M (editors): Ubrich’s International Periodicals Directory (21st ed,
in 2 vols). RR Bowker Co, New York, 1982.
415. Scott DHT, Arthur GR, Scott DB: Haemodynamic changes following buprenorphine, and
morphine. Anaesthesia 35:957-961, 1980.
416. Seigel D (chairman, editorial committee): A Memorial Symposium in Honor of Jerome
Cornfield. Biometrics 38(suppl): 1-165, 1982.
417. Seigel D, Milton RC: Further results on a multiple-testing procedure for clinical trials.
Biometrics 39:921-928, 1983.
418. Selmer ES: Registration numbers in Norway: Some applied number theory and psychology.
JRSS 130(series A):225-231, 1967.
419. Seltzer HS: A summary of critcisms of the findings and conclusions of the University Group
Diabetes Program (UGDP). Diabetes 21:976-979, 1972.
420. Shapiro SH, Louis TA (editors): Clinical THals: Issues and Approaches. Marcel Dekker, Inc,
New York, 1983.
421. Sherwin R, Kaelber CT, Kezdi P, Kjelsberg MO, Thomas HE Jr: The Multiple Risk Factor
Intervention Trial (MRFIT): II. The development of the protocol. Prev Med 10:402-425,1981.
422. Simon R: Adaptive treatment assignment methods and clinical trials. Biometrics 33:743-749,
1977.
423. Smith CV, Pyke R: The Robbins-Isbell two-armed-bandit problem with finite memory. Ann
Math Statistics 36:1375-1386, 1965.
424. Smith TW: Qualifications to generalized absolutes: “Approval of hitting” questions on the
GSS. Public Opinion Quart 45:224-230, 1981.
A
A
6
20
9, A
7
7, 19, 26
12
14
A
7
26
8
1
20
A
7, 19, 26
18
3, 13
10, A
A
12
448
U 449
I. Combined bibliography
425. Smith WM: Problems in long-term trials. Merck, Sharp and Dohme, West Point, Pa. Pre
sented at the Joint WHO/ISH Meeting on Mild Hypertension, Susona, Japan, Sept, 1978.
426. Smythe M: Record numbering (pp 179-187 in Record Linkage in Medicine, HD AchjSon,
editor). The Williams and Wilkins Co, Baltimore, 1968.
427. Snedecor GW, Cochran WG: Statistical Methods (6th ed). The Iowa State University Press,
Ames, Iowa, 1967.
428. Snell, ES, Armitage P: Clinical comparisons of diamorphine and pholcodine as cough sup
pressants: By a new method of sequential analysis. Lancet 1:860-862, 1957.
429. Society for Clinical IVials, Inc: By Laws. Controlled Clin Trials 1:83-89, 1980.
430. Stallones RA: The effects of the Freedom of Information Act on research. Am J Public Health
72:335-337, 1982.
431. Stokes J, Noren J, Shindell S: Definition of terms and concepts applicable to clinical preven
tive medicine. J Community Health 8:33-41, 1982.
432. Strauss MB (editor): Familiar Medical Quotations. Little, Brown and Company, Boston, 1968.
433. Stress JK, Harlan WR: The dissemination of new medical information. JAMA 241:26222624, 1979.
434. Stross JK, Harlan WR: Dissemination of relevant information on hypertention. JAMA
246:360-362, 1981.
435. Sudman S, Bradburn NM: Asking Questions: A Practical Guide to Questionnaire Design.
Jossey-Bass Inc, Publishers, San Francisco, 1983.
436. Sutton HG: Cases of rheumatic fever. Guy’s Hosp Rep 11:392-428, 1865.
437. Szklo M: The epidemiologic basis for hypertension control. J Pub Hlth Policy 1:312-327,
1980.
3
A
9
9
1
7
A
2
7
7
12
I
7
T
438. Taylor SH, Silke B, Ebbutt A, Sutton GC, Prout BJ, Burley DM: A long-term prevention
study with oxprenolol in coronary heart disease. N Engl J Med 307:1293-1301, 1982.
439. Temple R, Pledger GW: Special report: The FDA’s critique of the Anturane Reinfarction
Trial. N Engl J Med 302-1488-1492, 1980.
440. Thorndike £l' Lorge CThe Teacher’s Word Book of30,000 Words. Columbia University, New
11 IK
iS
I
I
I
f
■
>
York, 1944.
441. Tonascia J: Lecture notes from Epidemiology 2. The Johns Hopkins School of Hygiene and
Public Health, Baltimore, 1983.
442. Tucker WB: The evolution of the cooperative studies in the chemotherapy of tuberculosis of
the Veterans Administration and Armed Forces of the USA: An account of the evolving
education of the physician in clinical pharmacology. Adv Tuberc Res 10:1-68, I960.
443. Tukey JW: “Quick and dirty methods in statistics.” Part II: A simple analysis for standard
designs. Proceedings of the Sth Annual Convention of the American Society for Quality
Control, pp 189-197, 1951.
444. Tukey JW: Some thoughts on clinical trials, especially problems of multiplicity. Science
198:679-684, 1977.
445. Tuskegee Syphilis Study Ad Hoc Advisory Panel: Final Report of the Tuskegee Syphilis Study
Ad Hoc Advisory Panel (47 pages). United States Department of Health, Education, and
Welfare, Washington, 1973.
446. Tygstrup N, Lachin JM, Juhl E (editors): The Randomized Clinical Dial and Therapeutic
Decisions. Marcel Dekker, Inc, New York. 1982.
U
447. United States Congress (87th): Drug Amendments of 1962, Public Law 87-781, S 1522.
Washington, Oct 10, 1962.
448. United States Congress (94th): Medical Device Amendments of 1976, Public Law 94-295.
S 510. Washington, May 28, 1976.
A
9, 18
12
18
1
20
20
14
18
1
1
5
449. United States Congress (97th): Small Business Innovation Research Act of 1981 (S 881).
Washington, Dec 8, 1981.
450. United States Court of Appeals for the District of Columbia Circuit: Peter H Forsham et al, 7
versus Joseph A Califano, Jr, et al. No. 76-1308, July 11, 1978.
451. United States Court of Appeals for the First Circuit: Robert F Bradley, et al, versus Caspar W 7
Weinberger, Secretary of Health, Education, and Welfare et al. No. 73-1014, July 31, 1973.
452. United States District Court for the District of Columbia: Forsham et al, versus Mathews, 7
et al. No. 75-1608, Sept 30. 1975.
453. United States District Court for the District of Columbia: Peter H Forsham et al, versus David 7
Mathews, et al. No. 75-1608, Feb 5, 1976.
454. United States District Court for the District of Delaware: Pharmaceutical Manufacturers 1
Association versus Robert H Finch, Secretary of Health, Education, and Welfare and Herbert
L Ley, Jr, Commissioner of Food and Drugs. Affidavit of William Thomas Beaver, MD. Civil
action no. 3797, 1969.
455. United States District Court for the District of Delaware: Pharmaceutical Manufacturers 1
Association versus Elliot Richardson, Secretary of Health, Education, and Welfare and
Charles C Edwards, Commissioner of Food and Drugs. Order denying preliminary injunction
and granting summary judgment of dismissal. Civil action no. 3946, Oct 20, 1970.
456. United States District Court for the District of Maryland: Shirley Bailey et al. versus Robert J 14
Lally, etc., et al. Civ nos. K-74-II02, K-75-1370, K.-75-1371 and K-75-1372, July 21, 1979.
457. United States District Court for the Southern District of New York: Ciba-Geigy Corporation 7
versus David Mathews, et al. No. 75 Civ 5049, Oct 14, 1975.
458. United States District Court for the Southern District of New York: Ciba-Geigy Corporation 7
versus David Mathews, et al. No. 75 Civ 5049 (CHT), March 8, 1977.
7
459. United States Senate Select Committee on Small Business: Oral Hypoglycemic Drugs: Hearing
before the Subcommittee on Monopoly, Sept 18, 19, and 20, 1974. Part 25, U.S. Government
Printing Office, Washington, 1974.
460. United States Senate Select Committee on Small Business: Oral Hypoglycemic Drugs: Hearing 7
before the Subcommittee on Monopoly, Jan 31, July 9-10, 1975. Part 28, U.S. Government
Printing Office, Washington, 1975.
461. United States Supreme Court: Peter H Forsham, et al, versus Joseph A Califano, et al. 7
Petition for a writ of certiorari to the United States Court of Appeals for the District of
Columbia Circuit, Oct term, 1978.
462. United States Supreme Court: Forsham, et al, versus Harris, Secretary of Health, Education 7, 24
and Welfare, et al. Certiorari to the United States Court of Appeals for the District of
Columbia Circuit. No. 78-1118, argued Oct 31, 1979, decided March 3, 1980.
25
University of Chicago Press: The Chicago Manual of Style (13th ed). Chicago, 1982.
463.
,
________________
g The effects of hypoglycemic
_
Group ___________
Diabetes Program
Research Group:
agents on
464. University
vascular complications in patients with adult-onset diabetes: 1. Design and methods (abstract).
Diabetes 19 (suppl l):387, 1970a.
465. University Group Diabetes Program Research Group: The effects of hypoglycemic agents on
vascular complications in patients with adult-onset diabetes: 2. Findings at baseline (abstract).
Diabetes 19 (suppl I):374, 1970b.
466. University Group Diabetes Program Research Group: The effects of hypoglycemic agents on
vascular complications in patients with adult-onset diabetes: 3. Course and mortality (ab
stract). Diabetes 19 (suppl l):375, 1970c.
467. University Group Diabetes Program Research Group: A study of the effects of hypoglycemic
agents on vascular complications in patients with adult-onset diabetes: 1. Design, methods, and
baseline characteristics. Diabetes 19 (suppl 2):747-783, 1970d.
468. University Group Diabetes Program Research Group: A study of the effects of hypoglycemic
agents on vascular complications in patients with adult-onset diabetes: II. Mortality results.
Diabetes 19 (suppl 2):785-83O, 1970e.
7
------
7
7
3, 6. 7,
8.10.11,
19, 24,
25
1,4, 7,
8.10.11,
15,1720,24,25
◄
Z 451
/. Combined bibliography
450
489 Wei LJ, Durham S: The randomized play-the-winner rule in
.ii Research Group: Effects of hypoglycemic agents on
469. University Group Diabetes Program
adult-onset diabetes: 111. Clinical implications ot
vascular complications in patients with
v
UGDP results. JAMA 218:1400-1410, 1971a.
470 llniversitv Group Diabetes Program Research Group: Effects of hypoglycemic agents on
4
“ compTahons tn patients with adult-onset dtabetes: IV. A prel.mtnary report on
10
6
490.
Z MH, S—M
7, 8, 20,
24
phenformin results. JA MA 217:777-784, 1971b.
A
, ./ C—™
Company, Inc, New York, 1970.
Diahetes: Modern
492-
472.
phenformin therapy. Diabetes 24(suppl I). 65-184, 1975.
473 University Group Diabetes Program Research Group: Effects of hypoglycemic agents on
471 vascular^contphcations in patten'ts w.th adult-onset dtabetes: VL Supplementary report on
nonfatal events in patients treated with tolbutamide. Diabetes 25.1129-1153, 1976.
University Group Diabetes Program Research Group: University Group Diabetes Program
474.
releases data on 1,027 patients. Diabetes 26:1195, 1977.
7, 17,
19, 20,
24, 25
7
nonfatal events with insulin treatment. JA MA 240:37-42, 197».
™«==
476.
medical trials. JASA 73:840-843,
7
therapy: Final report. Diabetes 31 (suppl 5): 1-81, 1982.
477. University Group Diabetes Program Research Group: Paper listing of^aselme and fofew-up
data on UGDP patients (one volume per treatment group). No. PB 83-136 325, Nation
Technical Information Service (NTIS), Springfield, VA, 1983a.
New
Bases
14
494.
G and C Merriam Co,
1, 7, 8
8
Treatment with Anticoagulants: A Study of 1.03
P
" - <;>: “/xx ',x
498.
499.
500.
7
j
7:265-300, 1981.
64:131-146. 1969.
50L
V
5, 23
479. Veterans Administration Cooperative Studies Program: Guidelines for VA Cooperative
Studies (Sth ed). Washington, April, 1982.
7, 16
480. Veterans Administration Cooperative Study Group on Antihypertensive
treatment on morbidity in hypertension: Results in patients with diastolic blood pressure
averaging 115 through 129 mm Hg. JA MA 202:116-122, 1967.
7, 16
481. Veterans Administration Cooperative Study Group on Antihypertensive^
treatment on morbidity in hypertension: II. Results in patients with diastolic blood pressure
averaging 90 through 114 mm Hg. JAMA 213:1143-1152, 1970.
482. Vonderlehr RA, Clark T, Wenger OC, Heller JR Jr: Untreated syphilis in the male Negro. A 14
comparative study of treated and untreated cases. J Vener Dis Inform 17:260-265, 1936.
I800
. „
487. Waterhouse B: A Prospect of Exterminating- the Small Pox. (part II). University Press,
Cambridge, 1802.
.
488. Watsop BL: Disclosure of computerized health care information. Am J Law Medicine
12
12
’2
12
(NTIS), Springfield, VA, 1983b.
W
’ ’ Waue
483.
Wade N;
N: Tvvd
Food Board’s fat report hits fire: Academy discovers Cassandra’s problem: What
good is the truth if its not spreadable? Science 209:248-250, 1980.
484. Wald A: Sequential Analysis. John Wiley and Sons, New York, 1947.
to Longer Life. Public
485. Warner R, Wolfe SM, Rich R: Off Diabetes Pills: A Diabetic’s Guide
Citizen’s Health Research Group, Washington, 1978.
486. Waterhouse B: A Prospect of Exterminating the Small Pox. Cambridge Press, Cambridge,
A
7
7, 17, 24
478.
'> h<
1, 2
washi"g'on’
4W-
495 Woolf HB (editor in chief): Webster's New Collegtale Dtcttonary.
7, 8, 15,
17, 23,
24, 25
7
7
22
9
7
,
504. Zelnio RN: Data collection techniques: Ma.l quest.onnatres. An. Hosp
Zukel WJ: Evolution and funding of the Coronary Drug Project. Con.roUed Clin THaisA:^-
505.
312, 1983.
10, A
14
12
13,21
Index
This index eons.s.s of 6.3 au.hor head.ngs and
505 references listed in the Combined mbhography < Append
the 130 references in Table B-3 of Appendix B or
The index contains an entry for each person orc-
in Appendix C.
represented in a bibliographic citation, even if
of the Terence is cited or the
that author’s name does not appear on the Pa8e
( -gauthor is not the first author on papers jnvoMng th e or more^t O
cited in the index is used to denote citations of the latter4ype. All a
superscript asterisk next to the page
P
a
of twf) page
of
Combined Bibliography (i.e., last
listings, at least one of which refers to a page i
first (or second) name is used if the heading contains citations reiated to
"s^of^XX^cjXteemitm
'Zt^tXi“■" C°roT DrU8hePhratCt <CDP) <ReSearCh Gr0UP)-,nClUde
references to the study itself as well as to papers Pr®du“
g
d eflected in comments throughout
««■ “■ ss- “ ""
appea^in iuiics the majority of which refer to pages in the Giossary (pages 281 to 308).
A
I
I
Ih..
1 '.i.1
Abt, K. 212,214, 430
acts: see legislative acts and regulations
adherence: see treatment, adherence
adverse side effect, adverse drug reaction: see side effect.
adverse
Alamercery, Y: 31, 430
Albert Einstein College of Medicine: 354
Albrink, Margaret J: 349
allocated (assigned) treatment: 199-200, 282
allocation: 6, 12, 16, 282, 305, see also randomization
adaptive: 281
baseline: 282
number: 295
outcome: 296
play the winner: 298
two-armed bandit: 307
design: 281, 290, 305
dynamic: see allocation, adaptive
equal: see allocation, uniform
fixed: 289-90
methods: 12. 16, 25-26, 67-68
303. 305
restricted: 301
schedule: 282, 290, 301, 305, 307
strata: 93, 305
stratified: 303
systematic: 6, 68, 198
uniform: 78, 92, 289, 295, 307
unit: 10, 305
unrestricted: 307
Amberson, JB Jr: 4, 6, 430
Ambroz, A: 15*. 300*, 432
American Bible Society: 3, 430
American Diabetes Association: 51, 53, 263
American Heart Association: 7, 102, 123
American Medical Association (Journal of): 59, 257
analysis: 8, 72-74, 80, 96, 173, 175-95, 201-2, 204-6, 20816, 269, 275, 282-83, 286, 288-92, 294, 301-2, 304,
363-64, 367
adjustment procedures: 193-95, 286, 294, 304
multiple regression: 194-95, 286, 294
subgrouping: 19,3-94, 304, see also
also stratification
stratification
Bayesian: 8, 283
(by) intention to treat: 186, 204, 282
(by) treatment administered: 186, 282
cohort analysis: 192-93
database: 179-84, 282
fixed sample size: 72-74, 185-95
groundrules: 80, 185-87, 201-2, 204-5
group sequential: 290
...
..
interim: 96. 173, 206, 208-16, 291, see also interim result
lifetable: 188-92, 292
procedures: 185-95. 204-6. 208-15. 269. 275, 363-64,
367, see also analysis, adjustment procedures; analy
sis, lifetable; chi-square test (of significance);
Fisher’s exact test (of significance); Mantel-Haenszel
test statistic; inverse sine transform test (of signifi
cance); log rank test statistic; mutiple, comparisons;
multiple, looks; Poisson test (of significance); /-test
(of significance)
regression procedures: see analysis, adjustment proce
dures
risk factor: 301
sequential: 72-73, 302
subgroup analysis: 204, 213-15, see also analysis, adjustment procedures
453
!
Index 455
454 Index
ancillary study(ies): 46-48, 160, 238, 260, 282
Anderson, I: 17*, 446
Anderson, LK: 284, 430
Anscombe, FJ: 73, 212, 430
Anturane Reinfarction Trial (ART) (Research Group): 80,
186, 236. 353, 430
archives: see repositories
Armitage, P: 72-73, 94*. 187, 190*. 212, 430, 446, 448
Arora, SS: 17*, 440
Arthur, GR: 68*. 447
Ashford, JR: 120*. 439
Aspirin Myocardial Infarction Study (AMIS) (Research
Group): 35-36, 44, 52, 117, 142, 150, 176, 182, 189,
195, 243-44, 313, 316, 321, 327-48, 353, 430
assigned (allocated) treatment: 199-200, 282
audit(s): 68, 100-101, 105, 170-71, 176, 198, 263, 286
Avicenna: 3
award: 287, 290-91, see also contract(s); funding; grant(s)
B
•5
■>
••Il
s
1
Backstrom, CH: 120, 430
backup: see data, backup procedures
Bacon, Sir Francis: 141
Bailey, NTJ: 212, 430
Baker, AB: 6*. 435
Barker, KN: 120, 431
Barnard, PJ: 120, 431
Baron, J: 5, 431
Barth, SN:9*, 13*. 451
baseline: 282
comparability of treatment groups: 6, 56-57, 67, 91, 93
95, 116-18, 199-201,204,206
via adjustment: 201
via baseline adaptive allocation: 91
via blocking: 95
via matching: 67, 93
via patient selection: 67, 116-18
via post-stratification: 93
via stratification: 67, 93-95, 200
data: 282
examination: see examination, baseline
observation: 283
visit (prerandomization and randomization): 283, 299300
Bateman, JR: 206*. 209*, 432
Bayes, Thomas: 9, 73, 431
Beatty, WK: 13,431
Beaver, William: 8
Beck, DF: 68*, 451
Beeson, PB: 58, 431
Begg, CB:9I,431
Begin, G: 125, 431,438
Begun, JM: 213, 431
Bellerose, J: 125*. 431
Bennett, PH: 260
Bergman, JJ: 44*, 436
Berman, JJ: 58*, 443
Beta Blocker Heart Attack Trial (BHAT) (Research
Group); 69, 156-57*, 353,431,440
bias: 283
control: 68-70, 90, 118
laboratory: 172
treatment related: 67-68, 118, 122, 172, 186, 198-200,
203-4
bibliography(ies):
(of) clinical trials published in 1980: 355-62
(of) clinical trials sketched in Apendix B: 319-26
construction: 268-69, 271-72
Bingham, Stephen F: 46*. 149*. 354, 433
Biological Abstracts (B1OS1S): 272, 283
Biometric Laboratory Information Processing System
(BLIPS): 11
Bishop, GF: 120, 431
Blackburn, B: 300*, 432
Blackburn, Henry: 349
Blake, FG: 5*. 440
blind(ed): 283, see mask(ed) (ing)
Block. JB: 24*, 50*, 432
block(ing): 92, 95-96, 198-200, 283, 305, 335, 350
Blue Cross, Blue Shield: 43
boards: see committee(s) (of clinical trials)
Boisen, M: 7*. 438
Boissel, Jean-Pierre: 353
Boivin, M: 125, 431
Bonferroni’s inequality: 214
Boorkman, J: 271, 446
Boston (Sunday) Globe (The): 204, 431
Bowden, G: 31*, 430
Box, JF: 6, 431
Bradburn, NM: 120, 448
Bradley, H: 281*. 444
Bradley, Robert F: 58
Brandt, AM: 153, 431
Brandt, L: 213, 436
Braunwald, E: 17, 431
Breslow, NE: 94*, 187*. 190*, 285, 431, 446
Bross, IDJ: 17, 73,431
Brown, BW Jr: 94, 187, 212, 215, 431
Brown, J Howard: 40
Brown, MB: 179, 435
Bruce, RA: 66*, 433
Bryan, LS Jr: 15, 431
Buckland, WR: 281,440
Bull, JP: 3, 431
Bullowa, JGM: 6*, 445
Bulpitt, CJ: 187, 431
Burack, WR: 68*, 432
Burley, DM: 281-82*. 448
Bursic, ES: 281*. 301, 443
Burton, BT: 50, 431
business office: 227, 283
Buyse, ME: 187, 432
c
Cameron, WR: 68*, 435
Canadian Arthritis and Rheumatism Society: 7
Cancer Literature (CANCERLIT): 273, 283
Cancer Therapy Abstracts: 273
Charen, T: 13, 432
check digit: 171, 284, 291, see also identification number
(patient)
Chemical Abstracts: 272
Gherkin, DC: 44*. 436
Chicago Manual of Style (The): 268
chi-square test (of significance): 81-88, 187-88
Chlebowski, RT: 206, 209, 432
Chowdhury, NR: 17*, 442
Christensen, E: 24*, 440
Christiansen, I: 17,432
Ciba-Geigy Corporation: 353
centers) (in clinical trials): 16, 18-19, 23, 31-33, 283, 329Clark, T: 153*. 450
31, 349
Cleary, PA: 11*. 441
central laboratory: 23, 30, 36-38, 284, 330, 349
clinic(s): 94-95, 164-65, 197, 203, 284, 292, 296, 301, 303
clinical: 23, 284, 329, see also clinics)
coordinator: see coordinators)
clinical coordinating; see centers), treatment coordinat
director: see directors), center
ing
lead: 292
coordinating: 18, 21, 23, 30-31, 32-37, 164,t, 175-76, 181,
monitor: 284
225, 286, 330, 349, 353-54, 425-29
parent: 296
activities: 31-33
satellite: 301
budget: 225, 425-29
staffing: see centers) (in clinical trials), staffing
cost: 34-37, 425-29
study: 303
equipment: 34-35
-- ----------termination: 164-65,
197, 203,see also stage(s) (of a climperiod of support: 164
cal trial), termination
review: 175-76, see also site visit(s)
variability: 94-95
staffing: 33-34, 181, 426
visit: 284
coordinator: see coordinators)
clinical event(s): 10, 284, see also outcome (measure)
data 286, see centers) (in clinical trials), coordinating
Clinical Protocols (CL1NPROT): 272
data coordinating: 287, see centers) (in clinical trials),
clinical trial(s): 3, 11, 13, 17, 23-24, 284, 293
coordinating
activities: 18-22, 32, 363-73
director: see directors), center
comparative: 285
lead: 292
controlled: 8-9, 286
liason office: see centers) (in clinical trials), project
cooperative: 286, see clinical trial(s), multicenter
office
cost: 28-29, 40-48, 425-29
parent: 296
counts of: 9, 11-16,24.41-43,69, 78,91-92,94, 96, 117,
procurement and distribution: 23, 30, 39, 29 ,
167, 209, 220
project office: 23, 30, 38-39, 299, 330, 349
crossover: 14, 286
quality control: 23, 30, 176, 300, 330
designs: 10, 65-70, see also sample size, design; treat
reading: 23, 30, 38, 300-301, 330, 349
ment, design
resource: 19, 23, 30, 301, 330
diagnostic: 11-12, 16, 25-26, 287
satellite: 23, 301
double-mask(ed): 6, 9, 14, 16, 68-70, 288
staffing: 19-21, 33-34. 36-37, 226-28, 233, 235-36, 426,
drug: 14, 288
see also funding, budget; organization and manage
ethical issues in: 49-50, 65-66, 73-74, see also informed
ment practices
consent (process and statement)
support: see center(s) (in clinical trials), resource
explanatory: 289
treatment coordinating: 30-31, 39, 306, 330
funding: see funding
types: 283-84
history: 3-8
central laboratory: see centerfs) (in clinical trials), central
identification of: see publication(s)
laboratory
impact on practice of medicine: 5, 15, 17, 49-62
certification: see also training (staff)
incentives for doing: 27-28, 42-45, 238
clinic: 21, 365, 367
. .
initiation: 19-21, 27-28, 31, 219-31
personnel: 21, 147-48, 365, 367, 369, see also training
intervention: 292
(staff)
inventories: 11-13, 24-25. 27-29, 40^3, 45, 220
Cetrulo, C: 17*. 439
investigator initiated: 19, 27, 219-20, 224-25, 230, 328,
chairman of the study: see study, chairman
349
Chalmers, Thomas C: 4, 8, 15, 24, 49-50, 52, 65,
management: 293
153, 300, 351,432, 438
multicenter: 7, 10, 14-16, 23-24, 24-29, 237, 293, 313,
Chambless, LE: 194*. 441
353-54
Chandor, A: 281*, 432
need for: 15
Chang, A: 17*. 446
Canner, Paul L: 120*, 163, 172, 212. 215, 353, 432, 439,
441
Carcinogenesis Abstracts: 273
Carleton, RA: 68, 432
Casagrande, JT: 82, 84, 87, 432
Cascinelli, N: 49, 432
case-control study: 283
Case-Western Reserve University: 351
Cassel, GH: 152, 175,432
Catalog on Line (CATLINE): 272, 283
i
..
456
I
_ /l
t.3
Index 457
Index
clinical trial(s) (continued)
nonmasked: 16, 295
open (label): 296
organization: see organization and management practices
origin of term: 8
partially masked: 297
phase I: 13, 297
phase II: 11, 13, 297
phase III: II, 298
phase IV: 298
placebo-controlled, 6, 298
prevention: see clinical trial(s), prophylactic
primary prevention: 13, 299
prophylactic: 11-13, 16, 25-26, 299, 327
publication: see publication(s)
quality assurance (control): see quality assurance, quality
control
randomized (controlled): 12-13, 17, 25, 295, 300, see also
randomization
secondary prevention: 13, 302
sequential: 15, 285, 296, 302
single-center: 14, 16, 23, 24-29
single-mask(ed): 69, 302
sponsor initiated: 19, 27, 219-20, 223-24, 230, 328
staffing for: see centerfs) (in clinical trials), staffing
stage(s): see stage(s) (of a clinical trial)
surgical: 14, 304
therapeutic: 11-12, 14, 16, 25-26, 305, 327, 349
timetable: see timetablefs)
treatment: see clinical trial(s), therapeutic
triple-mask(ed): 307
uncontrolled: 307
unmasked: see clinical trial(s), nonmasked
vaccine: 4, 7
close of trial: 284
close-out: 22, 118, 159, 163-65, 200, 202, 216, 284-85, 297,
337, 351, 370-72
clinical trial: 164-65
data collection: 164-65
examination: see examination, close-out
follow-up visit: see follow-up, close-out follow-up visit
methods: 22, 118, 163-64, 285, 337, 351
common calendar date: 22, 118, 285, 337, 351
common period of follow-up: 22, 118, 285, 337
patient follow-up: 22, 118, 159, 163-64, 200, 202, 216,
284, 297, 370-72
Cobb, LA: 66, 433
Cocco, AE: 44, 433
Cocco, DV: 44, 433
Cochran, WG: 74, 82-84, 90, 96, 105-6, 188, 433, 448
cohort: 20, 192-93, 285
Collen, MF: 120, 433
Colley, JRT: 120*, 439
Collins, JF: 46, 149, 433
Colsky, J: 8, 433
Colton, T: 73», 212% 433, 435
Commerce Business Daily: 223
Commission on Professional and Hospital Activities
(CPHA): 52, 60, 61*, 433
Committee for the Assessment of Biometric Aspects of
Controlled Trials of Hypoglycemic Agents: 52-55*,
57-58, 100, 171, 204, 263, 433
Committee on Drugs: 44, 433
Committee on Lipoproteins and Atherosclerosis Technical
Group and the Committee on Lipoproteins and
Atherosclerosis of the National Advisory Heart
Council: 8, 433
committee(s) (of clinical trials): 4, 7, 70, 176, 234, 240-51,
282, 287-89, 292, 296, 303, 306, 339, 342, 351, see
also organization and management practices
advisory-review and treatment effects monitoring
(ARTEMC): 240-41 (functions), 282, 339
advisory-review (ARC): 234, 240-41 (functions), 282, 351
chairman’s: see committee(s) (of clinical trials), executive
data and safety monitoring: see committee(s) (of clinical
trials), treatment effects monitoring
data monitoring: 240, 287, see also committee(s) (of clini
cal trials), treatment effects monitoring
director’s: see committee(s) (of clinical trials), steering
editorial review: 288
ethics: see institutional review board (1RB); committee(s)
(of clinical trials), treatment effects monitoring
ethics review: see institutional review board (IRB);
committee(s), treatment effects monitoring
executive (EC): 240-41 (functions), 245-46 (member
ship), 289, 339, 351
human experimentation (HEX): see institutional review
board (IRB)
human volunteers: see institutional review board (IRB)
key: 292, 351
operations: 240, 296, see committee(s) (of clinical trials),
advisory-review; committee(s), treatment effects
monitoring
other: 242, 246, 351
policy (policy advisory board): see committee(s) (of clini
cal trials), advisory-review
quality control: 176
review: see committee(s) (of clinical trials), advisory
review
safety (monitoring): see committee(s) (of clinical trials),
treatment effects monitoring
steering (SC): 240-41 (functions), 244-46 (Membership,
operations), 303, 339, 351
Therapeutic Trials (of Great Britain): 4, 7, see also Medi
cal Research Council (of Great Britain)
treatment effects monitoring (TEMC): 70 (masking of),
240-41 (functions), 245-48 (membership), 306, 339,
342
communications: 248-50, 255, 261, 263, see also publica
tions)
center-to-center: 248-50
committee-sponsor: 248-50
constraints: 255, 261, 263
models: 249
comparison group: 4, 11, 13, 285, see also control(s), treat
ment
compliance: see treatment, adherence (compliance)
Computer Retrieval of Information on Scientific Projects
(CRISP): 273, 286
computing facilities: 179-81
Comstock, GW: 120,439
Condie, R: 68% 435
confidentiality: see data, confidentiality and security
conflict(s) of interest: 62, 237, 247-48, 274
confounding variable: see variable, confounding
consent: see informed consent (process and statement)
consortium agreement, consortium award: 228, 230-31,
237, 285, 297, see also award
contract(s): 8, 27, 47, 142, 220, 223-30, 285-86, 289, 301,
328
administration: 230
application: 223-24, 301
budget: 47, 225-29
contents: 224
considerations when applying: 223-24
deadline: 223
cost-reimbursement: 47, 286
fixed-cost: 47, 289
negotiations: 230
office(r): 285
review: 142, 223
control(s): 3, 5, 8, 10-11, 65-66, 75, 281, 285, 291, 332,
350
concurrent: 3, 5, 10, 285
group: 285, see also control(s), treatment
historical control(s) (group): 3, 5, II, 291
patient: 285
treatment: 3, 8, 10, 65-66 (choice), 75, 281, 285, 291,
332, 350, see also test treatment(s)
Controlled Clinical Trials (journal of): 4, 7, 272
controlled, uncontrolled (observation): 5, 8-9, 11, 13, 93,
286, 307
Cooke, ID: 17% 440
cooperative agreement: 286
Coordinating Center Models Project (CCMP) (Research
Group): 9, 18, 20, 31, 33-34, 223, 281, 433
coordinating center: see centers) (in clinical trials),
coordinating
coordinators) (clinic, data): 227. 284, 286. 287, 341, 344
Cornfield, Jerome: 8, 52*. 53, 56-57, 65, 73, 204, 208,
212, 283, 293, 301,433, 434
Coronary Artery Surgery Study (CASS) (Research
Group): 17, 36, 47, 102, 104-5, 117, 120, 149-50,
152, 167, 181, 198, 256, 313, 317, 323-25, 327-48.
353, 434
Coronary Drug Project (CDP) (Research Group): 4, 8,
18-21, 27, 33-37, 44, 46, 50, 52, 66-67. 69, 77-79,
86-87 (sample size illustration), 89, 92-94, 96, 9899, 101-3, 112, 114, 116-18, 120-23, 142, 145, 148,
150, 152, 157, 160 (access to data), 161, 162% 16365, 172, 175-76, 181-82, 184, 189-90, 192% 19597, 202, 206, 209, 212, 214-16, 219, 229, 242, 244,
248, 256, 258, 259% 260, 263, 269*, 286, 313, 316,
320-21, 327-48. 353, 434
Coronary Primary Prevention Trial (CPPT): see Lipid
Research Clinics (LRC)—Coronary Primary Pre
vention Trial (LRC-CPPT)
cost(s): 28-29, 40-48, 50, 224-25, 228-29, 288, 291
clinical trials: 28-29, 40-48
control measures: 46-48
direct: 225, 229, 288
factors in clinical trials: 45-46, 50
indirect: 224, 229, 291
patient care: 228
Cowan, DW: 6*, 435
Cox. DR: 74, 94*, 187*, 190*, 194-95, 286, 434, 446
Cox, GM: 74, 83-84, 90, 96, 105-6, 433
Cox proportional hazards regression model: 195, see also
analysis, adjustment procedures, multiple
regression
Craigie, WA: 281•, 444
Creighton, C: 5, 434
Crockett. JE: 66’, 435
Crombie, AC: 3, 434
cross contamination: see treatment, cross contamination
crossed treatments: see treatment, design, crossover; or
treatment, design, factorial
Crout, JR: 8, 434
Cuppies, LA: 212, 435
Curran, WJ: 4, 435
Current Contents: 271-72
Currie, VE: 17*, 443
Cutler, JL: 120*, 433
Cutler, SJ: 187, 435
D
data: 286, 290, 300
access: 160, 261-62
analysis: see analysis
audit(s): see audit(s)
backup procedures: 182-84
center: 286, see centerfs) (in clinical trials),
coordinating
collection: 20, 46, 48, 119-37, 148, 236, 287, 336, 350,
363
considerations (requirements): 122-26
forms: see forms; data, item(s)
schedule: 20, 120-22, 148, 336, 350
visit: 287
confidentiality and security: 137, 155, 164-65, 182-83,
261
coordinating center: see centers) (in clinical trials),
coordinating
coordinator: see coordinators)
dredging: 214-15, 287
edit(ing): 168-71, 182, 287-88
check: 168-71, 288
examples: 169-70
methods: 168-71
query: 168-71, 182, 288
entry: 57-58, 70, 167-71, 284, 287-88, 337, 351
falsification and purges: 203-4
flow and intake: 136-37, 166-68, 206, 209
integrity: 173, 182-83
item(s): 124-32 (construction), 379-416 (examples)
management: 147, 179-84
monitoring committee: see committee(s) (of clinical
trials), treatment effects monitoring
458
data (continued)
processing: 236, 364-65. 367-69, 371-72
purges: see data, falsification and purges
safety monitoring committee: see committee(s),
treatment effects monitoring
security: see data, confidentiality and security
storage: 136-37, 184, 270, 370, see also repositories
system: 287-88
database: 167, 284, 287-88, 303
Davignon, JP: 17*, 443
Davis, HL Jr: 49*. 432
Dawkins, HC: 213, 435
Day, HW: 17,435
Day, NE: 285, 431
deAlmeida, DJW: 68, 435
DeBakey, ME: 17, 435, 438
Deering, RB: 68*, 435
DeMets, DL: 69*. 91*, 156-57*, 290, 435, 438, 440
Dennis, EW: 17*, 438
Denyer, M: 68*. 442
si
J
Index 459
Index
Department of Health. Education, and Welfare (DHEW):
59, 184, 281*. 287, 435
Department of Health and Human Services (DHHS): 44,
184, 221, 281*, 287, 435
design: see sample size, design; study, design; treatment,
design
Detre, K: 17*. 444
Devan, WJ: 179, 435
Diabetes Control and Complications Trial (DCCTj. 2?0,
353
Diabetes: 257
Diabetic Retinopathy Study (DRS) (Research Group)
52, 70, 148-50, 196, 216, 219, 353, 435
Diehl, HS: 6, 435
Dillard, DH: 66*. 433
Dimond, EG: 66, 435
directors) (of centerfs]): 226-27, 284, 286, 299, 331, 341,
344, see also principal investigator
Disraeli, Benjamin, 196
Division of Research Grants (of the National Institutes of
Health): 11. 220, 223
Dixon, DO: 213, 436
Dixon, WJ: 179,435
Donaldson, B: 15, 435
Donaldson, N: 15, 435
Donaldson, RM: 44, 447
double-blind(ed): see double-mask(ed)
double-mask(ed): 288, see also mask(ed)
Dow Jones News Service: 55
Draper, N: 194, 435
drop-in(s): see treatment, cross contamination
dropout(s): 79, 89, 118, 157, 160, 161, 162-63, 197, 202,
288
follow-up: 157, 160-63, 202
how to handle in analysis and sample size calculation:
79, 89, 197, 202
reinstatement: 161, 202
tracing: 161-62
Drummond, JC: 4-5, 435
Duncan, DB: 213, 435-36
Duncan, WJ: 120, 436
Duncum, BM: 5, 436
Dunnett, CW: 78, 213, 436
Dunphy, JV: 66*. 445
Dupont, WD: 80, 212, 215, 293, 436
Durham, S: 92, 451
E
Early Treatment of Diabetic Retinopathy Study
(ETDRS) (Research Group): 21, 152, 353, 436
East India Shipping Company: 4
Eastern Cooperative Oncology Group (ECOG): 204, 353
Ebbutt, A: 281-82*, 448
Ederer, F: 80*. 187, 435, 439
Edgell, SE: 125, 439
edit(ing): see data, edit(ing)
Edmonds, DK: 17*. 440
Edvardsson, B: 120, 436
Edward, Tyron: 255
Efron, B: 91, 436
Elandt-Johnson RC: 187, 189, 436
Elliott, J: 15. 436
endpoint: 288-90, 303, see outcome (measure)
enrollment: see also recruitment (patient)
clinic: 20, 196
patient: 10, 20-21, 73, 102-3, 118, 157, 297, 305
entry criteria: see patient, entry and eligibility criteria
Equifax: 162
Eraut, CD: 17*. 442
Erickson, SH: 44, 436
Ethics Advisory Board of the Department of Health and
Human Services: 261, 436
European Coronary Surgery Study Group: 17, 256, 436
evaluable patient(s): see patient(s), evaluable
event: 3, 10, 66-67, 188-92, 197, 284-85, 289, 299, 302,
see also outcome (measure)
rate: 188-92, 197, 289
types: 284-85, 299, 302
clinical: 284
composite: 285
primary: 299
secondary: 302
examination: 120-24, 161, 283-84, 289-90, 297-300, 350
baseline: 122-23, 283, 350, see also baseline
close-out: 284
final: 289
follow-up: 121-22, 124, 290, 350, see also follow-up
missed: 161
post-randomization: 298
prerandomization: 299
pretreatment: 299
randomization: 300
Excerpta Medica (EMED): 272-73, 288
executive committee (EC): see committee(s) (of clinical
trials), executive
F
factorial design (structure): see treatment, design, facto
rial
feasibility study: see study, feasibility
Federal Register: 8
Feinstein, AR: 52-53*. 56-58, 204, 277, 436
Fellegi, IP: 284, 436
Feller, W: 214, 283,436
Ferebee, SH: 7, 443-44
Ferris, FL III: 152, 175, 432
Fibiger, J: 6, 437
Finch, Robert H: 8
Finkel, MJ: 44, 437
Finney, HC: 120,437
Fisher, Lloyd D: 353
Fisher, Sir Ronald A: 4, 6, 96, 188, 437
Fisher’s exact test (of significance): 81-84, 87, 188
Fitzgerald, RB: 68*. 435
Flamant, R: 49*, 432
Fleckenstein, L: 44*, 444
Heiss, JL: 74, 187, 437
Fleming, TR: 17*, 212, 443. 445
Fletcher, RH: 52*. 437
Hetcher, SW: 52*, 437
follow-up: 3, 10, 12, 14, 16, 25-26, 73, 76, 89, 120-24, 159,
160-63, 197, 202-3, 205, 216, 284, 290, 292-93, 295,
297-98, 301, 305, 336, 350, 373
close-out follow-up visit: 284, see also close-out
cohort: 290
data collection visit: 290
examination: see examination, follow-up
interim follow-up visit: 121, 205 (how to handle in analy
sis), 292
length, period: 12, 14, 16, 25-26, 73, 76, 89, 197, 336,
350
loss(es) to: 78, 160-63, 205 (how to handle), 293
mortality: 162-63
nonrequired follow-up visit: 121, 295
observation: 290
patient: 297
post-close-out follow-up visit: 298
post-randomization follow-up visit: 120, 298
post-treatment: 298
post-trial: 159, 216, 298, 350, 373
post-trial follow-up visit: 298
regular follow-up visit: 301
required follow-up visit: 121, 301
routine follow-up visit: 301
schedule: 121-22
scheduled follow-up visit: 121
study: see study, prospective follow-up
treatment application and adjustment follow-up visit:
305
visit: 290
Food and Drug Administration (FDA): 8, 17, 44, 51, 5255*, 57-59, 65, 142-45, 171, 201, 263, 289, 292-93,
297-99, 366, 368, 370, 372, 437-38
Food, Drug and Cosmetic Act of 1938 (and amendments):
4, 7, 143
Food and Nutrition Board: 237, 438
Forman, SA: 33-34*, 36-37*, 120*, 172*, 181-82*, 432,
441, 443
forms: 20, 119-20, 122-37, 144 47, 287, 364, 379-416
clearance by Office of Management and Budget: 144-45
construction: 122-36, 379-416
development: 119-20, 122-37, 364
examples: 379-416
format: 126, 132-36,401-16
review and test: 20, 145-47
Forsham, PH: 58
Francis, T: 7, 438
Fredrickson, Donald S: 17-18, 438
Freedman, LS: 91, 438
Freedom of Information Act (FOIA): 58, 165, 261
Freiman, J A: 15, 74, 89, 438
Frenette, C: 125, 438
Friedman, LM: 91, 438
Fruin, RC: 66‘, 445
funding: 8. 12-13, 15-16, 19-25, 27, 40-48, 219-31, 233,
284, 287, 290, 297, 328-29, 349, 425-29, see also
award; contract(s); grant(s)
budget: 224-29, 425-29 (example)
methods: 8, 16, 19-20, 27, 42-45, 47, 219-31, 284, 287,
297, 328-29, 349
negotiations: 230
office(r): 290
payline: 221, 297
period: 25, 329, 349
Furberg, CD: 91*, 438
G
Gabriel, KR: 213,431
Gail, M: 81-84, 438
Gallon, F: 301, 438
Garisch, JAM: 68*, 442
Garrett, HE: 17, 438
Gart, JJ: 82, 438
General Services Administration: 225
generalizability and validity issues: 196-207
Gent, M: 289, 293, 447
George Washington University (The): 353-54
Gilbert, JP: 15*. 74*. 89*. 443
Godbold, JH: 213, 436
Good Manufacturing Practices Regulation: 144
Gordis, L: 17, 438
Grant, AP: 73, 438
grant(s): 4, 8, 27-28, 47, 142, 220-30, 282, 290, 301, 349,
388, 425-29
administration: 230
application(s): 47, 220-29, 282, 301, 425-29
budget: 47. 224-29, 425-29 (example)
contents: 221-23
deadline (for National Institutes of Health): 220
management office(r): 290
negotiations: 230
planning grant: 28
review: 142, 220-21
Greenberg, BG: 8, 438
Greenhouse, SW: 8, 215, 438
Gregg, Alan: 271
Grizzle, JE: 94, 438
group sequential analysis: see analysis, group sequential
Grundy, SM: 260
Guerrier, CJ: 17*, 442
Gull, Sir William: 4, 6
Index 461
460 Index
Gurian, J: 80*. 439
Guy, W: II*. 441
H
1
Haenszel, W: 187, 190, 293.442
Hagans, J: 39, 99, 439
Haggard, HW: 15, 439
Hainlinc, A Jr: 172, 439
Halperin, M: 8, 80, 438-39
Hamilton, MP: 247,439
Hampton, JR: 17*. 439
handbook: 145-46 (development and maintenance), 304,
419-21 (example), see also manual of operations
haphazard: 6, 68, 290
Hardison, WGM: 260
Harlan, WR: 52, 448
Harris, Fred R: 4
Harvard University: 353-54
Harville, DA: 198, 439
Haseman, JK: 82, 87, 439
Hauck, WW: 96, 439
Haupt, BJ: 17, 439
Haverkamp, AD: 17, 439
Hawkins, Barbara S: 33, 163, 271, 354, 439
Hawkins, C Morton: 353
Haybittie, J: 120, 127, 133, 451
Haygarth, John: 4, 6, 9, 15, 439
Health, Education and Welfare (HEW): see Department of
Health, Education, and Welfare (DHEW)
Health and Human Services (HHS): see Department of
Health and Human Services
Hceren, T: 212*, 435
Heinz, EC: 33-34*, 36-37*. 181-82*, 443
Heller, JR Jr: 153*. 450
Helsing, KJ: 120, 439
Hemphill, FM: 7*. 438
Hendershot, RA: 284*, 430
Hennekens, Charles: 354
Higgins, KH: 14-16*, 268*, 443
Hill, Sir Austin Bradford: 4, 6-7, 30, 439
Hill, JD: 17, 439
Himmelfarb, S: 125, 439
Hinkley, DV: 74, 434
Hirschman, GH: 50, 431
historical control(s) (group): see control(s), historical
Hochstim, JR: 120, 439
Holdstock, G: 17*, 439
Holland, WW: 120,439
Hollander, M: 187, 431
Holman, DV: 15, 440
Hospital Cardiologique (Lyon): 353
Hospital Index: 272
Hospital Tribune: 52
Howard, JM: 69, 156-57, 440
Howard, SV: 94*, 187*, 190*, 446
Hulka, BS: 120, 446
Hultgren, HN: 17*, 444
Hursh-Cfcsar, G: 120, 430
Hypertension Detection and Follow-up Program (HDFP)
(Cooperative Group): 7, 52, 66, 117-18, 150, 152,
189, 229, 248, 313, 317, 322-23, 327-48, 440
Hypertension Prevention Trial (HPT): 21, 35-36, 39, 66,
104, 117, 125, 129, 145, 150, 152, 157, 167, 176, 189,
225, 235, 248. 250, 309, 313, 318, 326-48. 353,
374-75. 377-78, 417. 419-21, 425-29
hypothesis, null and alternative: 76, 212, 214, 282, 295-96,
307
Johnston, EM: 68*, 435
Jones, JH: 153, 440
Jones, S: 17*, 442
Jones, SE: 17*, 443
Joslin Clinic: 58
Juhl, E: 24, 187*, 440, 446
K
I
identification number (patient): 102, 131, 171, 291, 297
Iglewicz, B: 91, 431
IMS America, Ltd: 52, 59-61, 440
Index to Dental Literature: 272
Index Medicus: 13, 257, 259-60, 265, 269, 271-72
informed consent (process and statement): 4, 141-42, 144,
153-58, 164, 216, 230, 291, 364, 374-78
guidelines: 153-57
institutional reveiw board approval: 141-42, 144, 156-57,
230
sample statements: 374-78
updates: 157, 164, 216
Zelen procedure: 157-58
Ingelfinger, J A: 187*, 440
institutional review board (IRB): 4, 141-42, 144, 153, 15657, 230, 291, 364, 372-73
composition: 141
function: 141-42, 144, 156-57, 230
history: 141, 153
interaction (effect): 291, see also treatment, interaction
interim result: 256, 292, see also analysis, interim
International Cancer Research Data Bank: 11
International Committee of Medical Journal Editors: 268,
440
International Mexiletine Placebo Antiarrhythmic Coronary
Trial (IMPACT) (Research Group): 31*. 353, 430
International Nursing Index: 272
International Reflux Study in Children (IRSC): 31, 117,
309, 313, 318, 326-48, 354
inventories of clinical trials: see clinical trial(s), inventories
inverse sine transform test (of significance): 81, 83, 85-86,
88
Investigational Device Exemption (IDE) (Application):
142-45, 292
Investigational New Drug Application (INDA): 8, 142-45,
292, 366, 368, 370, 372
investigative group: 159-60, 292
Iverson, K: 17*. 432
J
Jafary, MH: 17*. 442
James, 0: 281,291,440
James, RC: 281,291,440
Jarrett, RJ: 51, 440
Jenner, Edward: 5, 9, 440
Jessen, O: 68*. 443
Joannou, P: 68*, 435
Johns Hopkins University (The): 353-54
Johnson, NL: 187, 189, 281*. 436, 441
Johnson, Samuel: 179
Kaelber, CT: 20*, 148*, 447
Kalbfleisch, JD: 187, 189, 440
Kane, Sidney H: 353
Kaplan, EL: 187,292,440
Kaplan-Meier product limit: 292, see also analysis, lifetable
Karisson, T: 31*. 430
Keefer, CS: 5, 440
Keen, H: 51, 440
Kefauver, Estes: 4
Kelsey, FO: 8, 440
Kelso, IM: 17, 440
Kendall, MG: 281, 440
Kenis, Y: 49*. 432
Kenton, C: 13, 440
Kezdi, P: 20*. 148*, 447
Kidder, LH: 120, 440
Kilkenny, MJ: 125*, 451
Kilo, C: 52*. 56, 274, 440
Kilo Research Foundation: 274
King George I: 4
King, LS: 15, 441
Kipling, Rudyard: 159
Kittle, CF: 66*. 435
Kjelsberg, Marcus O: 20*, 148*, 354, 447
Klassen, D: 120*, 447
Klein, E: 9, 281*, 441
Kieinbaum, DG: 194, 441
Klimt, Christian R: 163, 349, 353-54, 441
Knatterud, Genell L: 21, 52*. 58*. 70. 120. 349, 352-54,
441, 446
Knox, JHM Jr: 15,441
Knox, RA: 51, 441
Knuth, DE: 281*. 300. 441
Koch, R: 17*. 443
Kolata, GB: 58, 441
Korns, RF: 7*. 438
Kotz, S: 281, 441
Krol, William F: 164, 172*, 353, 432, 441
Krueger, Keatha K: 349
Kruskal, WH: 281, 441
Kuebler, RR: 15*, 74*, 89*. 438
Kuhn, RM: 46*, 149*. 433
Kunzler, A: 50*, 432
Kupper, LL: 194*. 441
Kvols, LK: 17*. 443
L
label insert (for licensed drugs): see package insert (for
licensed drugs)
Lachin. John M: 74. 84. 187*, 260, 353-54, 441
Lalanne, CM: 49*, 432
Lancaster, General James: 4
Langendoerfer, S: 17*. 439
Lannon, L: 9*. 13*. 451
Lasagna, L: 44*, 444
Last, JM: 281, 283, 285,441
Lawrence, CJ: 17*, 442
Lawrence, GF: 17*, 440
Lawrie, GM: 17, 435
Layne, BH: 120, 441
Lebacqz, K: 153-54,441
Ledger, RR: 53*. 441
Lee, ET: 187, 189, 441
Lee, S: 24*. 50*, 432
legislative acts and regulations: see Food, Drug and
Cosmetic Act of 1938; Freedom of Information
Act; Investigational Device Exemption;
Investigational New Drug Application; Medical
Device Act (Amendment) of 1976
Lellouch, J: 289, 293, 447
Levine, J: 11, 441
Levine, MM: 157, 441
Levine, RJ: 4, 153-54, 441
Ley, Herbert L: 8
liaison office: see centerfs) (in clinical trials), project office
Lieberman, GJ: 188, 441
lifetable: 292, see also analysis, lifetable
likelihood principle: 8, 73, 212, 292, see also analysis, Baye
sian
Lilienfeld, AM: 3, 441
Lind, James: 4, 5, 67. 442
Lindley. DV: 198, 442
Lipid Research Clinics (LRC) (-Coronary Primary Preven
tion Trial) (LRC-CPPT): 35-36, 134, 150, 152, 354,
442
Littman, A: 66*, 445
Lo Presti, F: 31*. 430
Lockwood. JS: 5*. 440
log rank test statistic: 81, 190, 192, 293, see also MantelHaenszel test statistic
Long, Crawford: 5
Long, John M: 152*, 354, 442
Lorge, I: 120, 448
Louis. TA: 187,447
Lowrie, Edmund G: 354
Lubin, B: 120*. 447
Lucke, W: 68*. 442
M
Mackenzie, WA: 4, 6, 437
Macular Photocoagulation Study (MPS) (Research
Group): 51, 110 (randomization example), 117, 150,
168-70, 173, 182, 188, 209, 216, 235, 256, 311, 313,
315, 319, 327-48, 354, 374-76, 417, 421-22, 442
Maitland, Charles: 4
Management Committee of the Australian Therapeutic
Trial in Mild Hypertension: 259. 442
management practices: see organization and management
practices
Mantel, N: 94*, 187, 190, 293, 442. 446
462
I
I
1
■
ij
I
j
Index 463
Index
Milman, N: 68, 443
Mantel-Haenszel test statistic: 190, 192, 293, see also log
Milne, JS: 120, 443
rank test statistic
Milton, RC: 212, 447
manual of operations: 145-46 (development and mainte
Mintz, M: 51, 53*. 443
nance), 262 (public access), 304, 417-19 (examples)
Mitchell, JR A: 17*. 439
manuscript preparation: see publication(s)
Moertel, CG: 17, 443
Market Facts, Inc.: 52, 442
Monahan, Michael: 264
Marks, IN: 68, 442
minitoring: 293
Marks, JW: 260
performance: 19, 152, 156, 173-76 (reports), 297
Marks, RG: 120, 442
treatment effects (also referred to as safety monitoring):
Marple, CD: 68*, 451
4, 19, 70, 206, 208-16, 233-34, 236, 297, 306, 345,
Marshall, EK Jr: 5*. 440
350, 368-69, 421-24
Martin, GL: 125, 442
frequency: 209, 345
Maryland Medical Research Institute: 353-54
history: 4
mask(ed): 6. 14, 16, 68-70, 96-105, 114-16, 164, 172, 200,
methods: 70, 209-16, 350, 368-69
203, 205, 288, 293, 302, 306-7, 333, 350, see also
rationale: 206, 208-9, 233-34, 236
unmask(ed)
reports: 209-11, 421-24
considerations: 114-16, 200
when needed: 208-9, 234
double: 6, 14, 16, 288, 333
Monte Carlo simulation: 212, 293
effectiveness: 69, 164, 200, 205
Montgomery, B: 261, 443
level: 16, 69, 333, 350
Morgan, DC: 17», 120*, 439, 442
principles: 68-70
Morgenstern, H: 281*, 301, 443
procedures: 97-100
Morris, RA: 58, 443
readings and determinations: 172, 203
Morris, W: 281*, 443
safeguards: 101-5
Morton, William TG: 5
single: 6, 16, 302, 333
Moses and Oakford (randomization) algorithm: 97-99,
triple: 307
107-11
uses: 68-70, 96
Moses, LE: 97, 443
Mathematical Reviews (MATHFILE): 272-73, 293
Mosteller, F: 15, 74, 89, 187*, 440, 443
Mather, A: 172*, 439
Mount, FW: 7, 443-14
Mather, HG: 17, 442
Mount Sinai School of Medicine: 351
Matta, RJ: 50*, 432
Muench, Hugo: 113
Matts, JP: 152, 442
Muggia, FM: 49*. 432
Mazzullo, JM: 44*, 444
Multicenter Investigation of the Limitation of Infarct Size
McCarn, DB: 13, 442
(MILIS) (Research Group): 234, 354, 444
McDermott, W: 58*. 431
Multiple Risk Factor Intervention Trial (MRFIT)
McDill, M: 18, 19, 31,442
(Research Group): 7, 28, 35-36, 50 (cost), 117,
McFarland, SG: 120, 442
148, 150, 152, 160 (access to data), 173, 189, 204,
McFee, JG: 17*. 439
206, 212, 214, 237, 309, 313, 317, 323, 327-48, 354,
McHugh, PM: 17*, 442
444
McLaughlin, GH: 374, 442
multiple
McMahon, BT: 4*. 6*, 430
comparisons: 78, 80, 213-14, 294
McPeek, B: 15*, 74*, 89*, 443
regression: 294, see also analysis, adjustment procedures,
McPherson, CK: 73*. 212*, 430
multiple regression
McPherson, K: 94*. 187*, 190*, 446
looks: 80-81, 211-12, 294
Meadows, AJ: 281*, 442
outcomes: 81, 204, 212-13, 294
medical device: 143, 293
Mundy, GR: 44, 444
Medical Device Act (Amendment) of 1976: 8
Munro, DF: 68*, 435
Medical Literature Analysis Retrieval System on Line
Murphy, J: 17*. 439
(MEDLINE): 9, 13-14, 259-60, 265, 269, 271-72
Murphy, ML: 17, 444
Medical Research Council (of Great Britain): 4, 7, 443
Murray, JAH: 281*, 444
medical subject heading (for clinical trials): 9
Medical TYibune: 52, 263
Medicare and Medicaid: 43-44, 52
Meier, P: 94, 187, 292, 440, 443
N
Meinert, Curtis L: 8, 14, 16*, 26, 33-34, 36, 37*, 41, 47-48,
Naggan, L: 17*. 438
52*, 58*. 173, 181-82, 242*, 268, 281, 353, 443, 446
Napier, JA: 7*. 438
Merendino, KA: 66*. 433
National Cancer Institute (NCI): 4, 7, 11-13, 24, 27, 40-43,
Miller, DT: 172*, 439
294, 444
Miller, JP: 52*. 56*, 274, 440
National Center for Health Statistics (NCHS): 54, 163,
Miller, Max: 351
294*. 444
Miller, RG Jr: 213, 443
National Cooperative Dialysis Study (NCDS): 354
National Cooperative Gallstone Study (NCGS) (Research
Group): 21, 31,91,96, 114-18, 120, 134, 142-43,
145, 154, 176, 187, 197, 212, 214-15, 238, 242, 260,
313, 315, 319-20, 327-48, 354, 417-19,444
National Death Index (NDI): 45, 155, 163, 165, 176, 294
National Diet-Heart Study Research Group: 206, 444
National Disease and Therapeutic Index: 52, 59
National Eye Institute (NEI): 13, 24, 34, 41-43, 51, 149,
294
National Foundation for Infantile Paralysis: 7
National Heart, Lung, and Blood Institute (NHLBI): 7-8,
13, 24, 33, 36, 40-43, 45, 52, 219, 242-13, 247, 294,
444
National High Blood Pressure Education Program: 52
National Enquirer: 261
National Institute of Allergy and Infectious Diseases
(NIAID): 13, 24, 41-43, 294
National Institute of Arthritis, (Metabolism) (Diabetes),
and Digestive (and Kidney) Diseases (N1ADDK):
13, 24, 41-43, 220, 294, 349-51, 444
National Institute of Child Health and Human Develop
ment (NICHHD): 13, 24, 41-43, 294
National Institute of Dental Research (NIDR): 13, 24, 4143, 294
National Institute of General Medical Sciences (NIGMS):
24,41-43,295
National Institutes of Health (NIH): 4, 7, 11-13, 24-25, 28,
38, 40, 47-48, 208, 220, 223, 238, 245, 294, 309,
444-45, see also entries for individual institutes
National Institutes of Health Clinical Trials Committee:
208, 445
National Institute of Mental Health (NIMH): 11
National Institute of Neurological and Communicative
Disorders and Stroke (NINCDS): 13, 24, 41-43,
295
National Library of Medicine (NLM): 9, 13, 269, 295, 445
National Prescription Audit: 58
National Research Council (of the National Academy of
Sciences): 17, 445
National Technical Information Service (NTIS): 256, 262,
273, 295, 310
natural history: 6, 46, 114, 294
of disease: 6, 46, 294
study: 114, 294
Nelson, Lord: 11
New Drug Application (NDA): 8, 294
New England Journal of Medicine: 257, 263
Newgate Prison: 4
Newman, IM: 125, 442
Neyman, J: 73, 445
Nicolle, Charles: 90
NIH Guide to Grants and Contracts: 223
Noren, J: 281*, 448
Norusis, MJ: 179,445
Notice of Claimed Investigational Exemption for New
Drug: 295
Notice of Grant Award: 230
null (treatment) hypothesis: see hypothesis, null and
alternative
Nuremberg Code for Human Experimentation: 4, 153
o
O’Brien, PC: 212-13,445
O’Donnell, R: 9*. I3*. 451
Oakford, RY: 97, 443
observation unit: 287, 295
Office of Management and Budget (OMB): 144-45, 296
Office for Protection from Research Risks: 141, 153, 154*,
156, 445
Office of Technology Assessment (OTA): 49*, 445
officers (of the study): 296
Olansky, S: 153*, 447
Oldendick, RW: 120*, 431
Onions, CT: 281*, 444
organization and management practices: 4, 8, 18-22, 26,
31-33, 38, 47, 62, 69, 74, 144, 147, 159-60, 203,
232-38, 240-51, 339-41, 343-45, 351, 363, 365-66
committee composition (rules, procedures): 234-35, 238,
241-48, 339-41, 343-45, 351
decision making: 233
deficiencies: 232-33
division of responsibilities: 4, 8, 18-21, 31-33, 38, 47, 62,
69, 144, 147, 203, 232-37, 243-47
maintenance activities: 159-60
priority assessment: 21
Orleans, M: 17*, 439
Ott, WJ: 17, 445
outcome (measure): 3, 10, 15, 25-26, 50, 66-67, 75-77, 8185, 197, 203, 206-7, 215, 283, 290, 296, 299, 302-4,
336, 350
binary: 76-77, 82-83, 84-85, 283
choice: 25-26, 50, 66-67, 75-76, 197, 203, 206-7, 215
composite: 25-26, 67, 197, 203
continuous: 77, 83-85
event: 3, 10, 66, 296
measure: 296
primary: 10, 15, 66-67, 81, 299, 336, 350
secondary: 67, 81, 302, 350
surrogate: 26, 66, 304
outlier: 171-72, 184, 205 (how to handle), 296
Overton, HH: 244, 445
Owen, DB: 188,441
P
Paasikivi, J: 51, 445
package insert (for licensed drugs): 296
Packard, FR: 3-4, 445
paper writing: see publication(s)
parameter: 296
Part, Ambroise: 3
Park, WH: 6, 445
Parsons, RJ: 17*. 440
participant: see study, participant
patient(s): 3, 8, 297
close-out: see close-out, patient follow-up
compliance: see treatment, adherence
contact: 45, 149-51, 288, 291
enrollment: see enrollment, patient
entry and eligibility criteria: 56-57, 67, 116-18 (choice
of), 197-98 (modification of), 202-3, 350
Index 465
464 Index
patient(s) (continued)
evaluable: 186, 275, 289
examination: see examination
follow-up: see follow-up
identification number: see identification number (patient)
log: see recruitment, log
monitoring: see committeefs) (of clinical trials),
treatment effects monitoring; monitoring,
treatment effects
population: see study, population
recruitment: see recruitment (patient)
registry: see recruitment (patient), log
safety monitoring: see committee(s) (of clinical trials),
treatment effects monitoring; monitoring, treatment
effects
selection: see patient(s), entry and eligibility criteria
transfers: 161
Patulin Clinical Trials Committee: 4, 7, 445
payline: 297, see also funding
Payne, SL: 120, 445
Pearson, ES: 73, 445
Pearson, NG: 17*. 442
Pearson, NJ: 120*, 439
Pellegrino, Edmund D: 119
performance monitoring: see monitoring, performance
Perkins, Elisha: 6, 15
Perry, GT: 66, 445
Persantine Aspirin Reinfarction Study (PARIS) (Research
Group): 33, 39, 44, 78, 92, 115, 117, 176, 182, 18889, 195, 209, 231, 236-37, 242, 260*, 265, 268-69*,
313, 316, 321-22, 327^8, 354, 374-77, 417, 423-24,
445
Peto, J: 94*, 187*, 190*, 446
Peto, R: 94, 187, 190, 446
Pharmaceutical Manufacturers Association: 8
Philadelphia Inquirer: 55
Physicians’ Health Study (PHS): 23, 117, 309, 313-14, 319,
327-48, 354
Pike, MC: 82*. 84*. 87*, 94*. 187*, 190*, 432, 446
pilot study: see study, feasibility
Pines, WL: 297-98, 446
Pinner, M: 4*. 6*. 430
Pitman, EJG: 68, 446
placebo: 4, 6, 69, 98, 115, 152, 202, 293, 298
effect: 6, 298
history: 4, 6
matching: 98, 115, 293
origin of term: 4
reaction: 69
reactor: 298
single-mask(ed): 152,202
Plackett, RL: 73, 446
Planck, Max: 15, 49
Pledger, GW: 80, 186, 448
Pocock, SJ:91, 187, 290, 446
Poisson test (of significance): 81, 83-86
policy (-advisory) board: see committee(s) (of clinical
trials), advisory-review
Poole, W Kenneth: 354
post-marketing surveillance: 298
post-stratification: 298, see also stratification
power: 15, 66-67, 75, 81, 84-85, 88-89, 289, 295, 299, 338,
351
assessments: 88-89, 338, 351
examples: 75, 88
formulas: 81, 84-85
Pre-Market Approval (PMA) Application: 299
Pentice, RL: 187, 189,440
Presberg, J: 204, 446
press queries and releases: 51, 55, 151-52, 257, 262-63
principal investigator: 240, 242, 299, see also directors) (of
centerfs])
priority score: 220, 230, 293, 299
Professional Standards Review Organization: 52
Program on the Surgical Control of Hyperlipidemia
(POSCH) (Research Group): 117, 149-50, 152, 247,
313. 317, 325-48, 354, 442
program
director: 299
office: 38, 299, see also centerfs) (in clinical trials), proj
ect office
officer: 299, see also project officer
project
director: 299
office: see centerfs), project office
officer, health scientist administrator: 39, 299, 341, 344
protocol: see study, protocol; treatment, protocol
Prout, BJ: 281*, 282*, 448
Prout, TE: 52*, 58, 446
Psychological Abstracts: 272
Public Affairs Information Service (PAIS): 272, 296
publication(s) (presentation[s]): 9, 11-16, 51, 62, 68, 204,
209, 214, 216, 238, 256-77, 286, 289, 293, 302, 308,
310, 319-26, 345, 352, 355-62, 365, 368-71, see also
communications
authorship: 23, 259-61, 286 (types), 310, 345, 352, 365
choice of journal: 257-58
clinical trials publications: 319-26, 345, 352, 355-62
contents: 11-16, 68, 204, 209-11, 264-69
credit rosters: 260
critiquing: 14-16, 272-77
identification: 9, 13-15, 264-65, 271-72
interim: 216, 256
preparation (procedure): 51, 238, 260-61, 264-70, 368-72
referencing procedures: 268-69, 310
what to publish: 257-58
when to publish: 256
where to publish: 257-59
writing team: 308, 352, 369
publicity: see press queries and releases
p-value: 195, 212, 214-15, 296
Pyke, R: 307, 447
Q
quality assurance, quality control: 152, 166-76, 300, 36569, 371, see also monitoring, performance; replica
tion, reproducibility
laboratory: 171-72
procedures: 152, 365-69, 371
reading center: 171-73
R
Rambo, DS: 153*. 447
Rand Corporation (The): 96-97, 100, 106, 109, 111-12, 446
Rand, LI: 21,446
random: 9, 68, 96, 100, 105-7, 110, 295, 300
numbeifs): 96, 100, 105-7, 110, 300
permutations: 96, 105-6
variable: see variable, random
randomization (randomized): 3-4, 6, 9-10, 12, 14, 16, 2526, 65, 67-68, 91-112, 193, 197-200, 203-4, 281-83,
290, 295-96, 300-1, 303, 307, 334-35, 350, see also
allocation
adaptive: 91-92, 198-99, 281-83, 295-96, 307, 334
baseline: 91, 282
biased coin: 91, 198-99, 283, see also randomization (ran
domized), adaptive, number
number: 91, 295, 334
outcome: 91, 296
play the winner: 91, see also randomization (random
ized), adaptive, outcome
two-armed bandit: 307, see also randomization (random
ized), adaptive, outcome
administration: 101-5, 199-200
block(ing): see block(ing)
breakdowns: 68, 105, 199-200, 203
conditions for: 65
documentation requirements: 100-1, 198
examination: see examination, randomization
examples: 105-12
fixed: 91-112, 290
history: 4, 6, 9
methods: 67-68 , 96-112, 198-200, 334-35, 350
rationale: 67-68, 198, 204
stratified: 107-10, 197, 303
unit: 300
visit: 300, see baseline, visit
rate: see event, rate
Ray, A A: 179, 446
Read, CB: 281*, 441
Read, KLQ: 17*. 442
recruitment (patient): 21, 45-46, 74, 118, 149-53, 198,
288, 291, 297, 301, 333, 350, 363-64, 366-67, see
also enrollment
goals, quotas: 21, 45-46, 74, 149, 297
ethics: 153
log: 152, 198, 301
methods: 118, 149-52, 288, 291, 297, 333, 350, 363-64,
366-67
regression to the mean: 301
Reitman, D: 300*, 432
relative betting odds (RBO): 212, 301
Reiman, AS: 17, 259, 446
Remington, RD: 28, 259, 446
Renne, KS: 120, 439
Renou, P: 17, 446
repeated significance testing: see multiple, comparisons;
multiple, looks
replication (reproducibility): 27, 38, 50, 171-73, 269
clinical trials: 50, 173
data analyses: 171, 269
laboratory tests: 38, 171-72
readings: 38, 171-72
treatment effect: 27
repositories: 164-65, 182-84, 211, 261-62, 270, 273, 295
request for application (RFA): 220, 301, see also grant(s),
application(s)
request for proposal (RFP): 20-21, 47, 223-25, 301, see
also contract(s), application(s)
Research Award Index (RAI): 273, 300
Research Development Committee, Society for Research
and Education in Primary Care Internal Medicine:
271, 446
research group: see investigative group
Research Triangle Institute: 354
response variable: see variable, outcome
review group: 231, 281, 301, see also contract(s), review;
grant(s), review; study section
Rheumatic Fever Working Party of the Medical Research
Council of Great Britain and the Subcommittees of
Principal Investigators of the American Council of
Rheumatic Fever and Congenital Heart Disease,
American Heart Association: 7, 8, 446
Rhodes, IN: 125*. 451
Rich, R: 52*. 450
Riley, IS: 17*. 442
risk factor (identification): 94, 301, 317
Rivers, E: 153*, 447
Robbins, H: 298, 307, 446
Rogot, E: 80*. 439
Romm, FJ: 120, 446
Roper, FW: 271,446
Rosenbluth, MB: 6*, 445
Roth, A: 120, 447
Rothman, DJ: 153, 447
Rowe, BC: 73*, 212*. 430
Royall, RM: 212, 215, 447
Rozencweig, M: 49*. 432
Rubin, J: 17*. 443
Ryden, VMJ: 206*. 209*. 432
S
Sackett. DL: 289, 293, 447
safety monitoring: see monitoring, treatment effects
Sales, BD: 58*, 443
Salisbury (ship): 5
sample size: 10, 12-13, 15-16, 24-26, 46, 66, 71-89, 197,
202, 281, 288-90. 295-96. 301-2, 327, 334, 33738, 349, 351
calculations: 15, 71-89, 202, 301, 338
examples: 74, 85-88
formulas: 81-84
considerations: 46, 66
deficiencies: 74
design: 10, 15, 72-89, 284, 290, 295-96, 302, 327, 349
I
Index 467
466 Index
I
I
1
'■I
I
■ •1
sample size (continued)
fixed: 10, 72-89, 290, 327, 349
nonsequential: 10, 295
sequential: 10, 15, 72-74, 284, 296, 302‘, closed: 73, 284;
open: 72-73, 296
reported in publications and research proposals: 12-13,
15-16, 24-26, 89, 334, 338, 351
specifications: 74-81, 197, 337, 351
Sanders, CA: 68*, 432
Santayana, George: 3
Sarna, G: 17*. 443
Schade, RR: 44, 447
Schatzkin, A: 212*, 435
Schefft, H: 213, 447
Scheibel, J: 68*. 443
Schlesselman, JJ: 74, 283, 447
Schmeck, HM Jr: 53*. 447
Schneeweiss, R: 44*. 436
Schonfield, U: 260
Schoolmaker, RC: 284*, 430
Schor, S: 52*. 53, 56-58, 204, 277, 447
Schriesheim, CA: 120, 447
Schroeder, B: 15*. 300*, 432
Schuman, SH: 153, 447
Schwartz, D: 289, 293, 447
Schwartz, TB: 52*, 447
Schwarz, RM: 125*, 451
Science Citation Index (SCI): 271-72
Sciotti, H: 271, 447
Sciotti, S: 271*. 447
Scott, DB: 68*. 447
Scott, DHT: 68, 447
Scott, YB: 13, 440
secular trend: 172-73, 198, 205, 302
Seigel, D: 8, 212, 447
selection criteria: see patient(s), entry and eligibility criteria
Selmer, ES: 284, 447
Seltzer, HS: 52-53*, 56-58, 204, 277, 447
sequential analysis: see analysis, sequential
sequential design: see sample size, design, sequential
sham procedure: 4, 66, 302
Shapiro, ED: 4, 435
Shapiro, SH: 187, 447
Shareck, EP: 15*. 432
Shaw, DB: 17*, 442
Shaw, George Bernard: 23, 185
Shaw, LW: 260
Sherwin, R: 20, 148, 447
Shindell, S: 281*. 448
side effect(s): 144-45, 282, 302, 305-6
adverse: 144-45, 282
toxic: 305
Siegelaub, AB: 120*, 433
significance level, significance testing: 296, 302, see also
test of significance
Silke, B: 281-82*, 448
Silverman, B: 15*, 300*. 432
Simon, R: 91, 281, 446, 447
Simpson, James Y: 5
site visit(s): 152, 156, 175-76 (procedures), 175, 303
Skouby, AP: 17*. 432
Smith, AGL: 68*. 435
Smith, CA: 153*, 447
Smith, CV: 307, 447
Smith, H Jr: 15*. 50*. 74*, 89*. 194, 300*, 432, 435, 438
Smith, Nathan: 15
Smith, PG: 82*. 84*. 87*. 94*, 187*, 190*. 432, 446
Smith, TW: 120, 447
Smith, WM: 19, 448
Smythe, M: 284, 448
Snedecor, GW: 74, 448
Snell, ES: 73, 448
Society for Clinical Trials, Inc: 4, 7, 62, 448
sponsoring agency: 19, 303
stage(s) (of a clinical trial): 18-20, 32-34, 163-65, 291, 29799, 303-4, 306, 327, 363-73
initial design: 19-20, 32, 291, 327, 363-64
patient close-out: 19, 32, 34, 163-64, 297, 327, 370-72
patient recruitment: 19, 32, 34, 297, 327, 366-68
post-trial follow-up: 19, 33, 165, 298, 327, 373
protocol development: 19-20, 32, 34, 299, 327, 364-66
termination: 19, 32, 34, 164-65, 304, 327, 372-73
treatment and follow-up: 19, 32, 34, 306, 368-70
Stallones, RA: 58, 448
Staquet, MJ: 49*. 187*, 432
statistical significance: see p-value
Steed, GR: 17*. 442
steering committee: see committee(s) (of clinical trials),
steering
Stokes, J: 281*, 448
stop condition (stop item): 130, 303, 398, see also forms,
construction
stopping boundary (stopping rule): 72-73, 197, 215-16,
288, 303, see also monitoring, treatment effects
stratification: 80, 93-95, 197, 199-200, 303, 334-35, 350,
see also post-stratification
advantage, disadvantages: 93-95
choice: 93-95, 197, 199, 334, 350
considerations: 80, 93-95
number of strata used: 93, 200, 335, 350
variable: see variable, stratification
Strauss, MB: 15, 448
Stress, JK: 52, 448
study: 303
ancillary: see ancillary study(ies)
chairman: 240-41, 243, 303, 339, 341, 344
function: 240-41
selection and qualifications: 243
comparative: 3, 285
database: see database
design: 196-97, 363, see also sample size, design; treat
ment, design
director: 240
feasibility: II, 289
follow-up study: 290
group: 304
handbook: see handbook
intervention study: 292
investigator: 304
manual of operations: see manual of operations
participant: 304
patient: 304
physician: 304
plan: 113-18, see also study, protocol
population: 26-27, 51, 116-18, 197-98, 206, 304, 333,
350
prospective follow-up study: 11, 299
protocol: 7, 141-48, 304, 364-66
section: 220-21, 281, 304, see also review group
subject: 304
vice-chairman: 233, 243, 304, 339-41, 344
subgroup!ing): 80-81, 193-94, 204, 214, 286, 304, see also
analysis, subgroup
cut-point: 214 (choice), 286, 304
examples: 193-94
groundrules: 204, 214
sample size calculations: 80-81
variable: see variable, subgrouping
Sudman, S: 120, 448
Summers, William: 5
Sundaresan, PR: 44*. 444
Suntar, A: 284, 436
survival analysis: see analysis, lifetable
Sutton, GC: 281-82*, 448
Sutton, HG: 4, 6, 448
Sylvester, RJ: 187*, 432
Szklo, M: 52, 448
T
Takaro, T: 17*, 444
Tamminen, Tytti: 354
Tanur, JM: 281, 441
Taylor, SH: 281*, 282, 448
Teacher's Word Book of 30,000 Words: 120
technical group: see investigative group
Temple, R: 80, 186, 448
terminology conventions: 8-9
test group (test-treated group): 305
test of significance: 201, 296, 305, 307, see also chi-square
test (of significance); Fisher’s exact test (of signifi
cance); hypothesis, null and alternative; inverse sine
transform test (of significance); log rank test statis
tic; Mantel-Haenszel test statistic; Poisson test (of
significance); r-test (of significance)
test treatment(s): 3, 8, 10, 50, 92, 305, 332, 350, see also
control(s), treatment
choice of: 65-66
number for sample size calculation: 74-75
test-control (treatment) difference: see treatment, difference
therapeutic regimens (mentioned in text)
dietary substances and regimens: 3-6, 49, 314, 317-18
beta-carotene (cancer): 314
cholesterol reduction (heart disease): 317
cyder (scurvy): 5
elixir vitriol (scurvy): 5
King’s meat (appearance): 3
lemon juice (scurvy): 4
lemons and oranges (scurvy): 5
mint water (rheumatic fever): 6
nutmeg (scurvy): 5
potassium enriched foods (blood pressure): 318
pulse and water (appearance): 3
seawater (scurvy): 5
sippy diet (ulcers): 49
sodium reduction (blood pressure): 318
vinegar (scurvy): 5
weight reduction (blood pressure): 318
drugs: 5-7, 11, 13, 17, 44, 51-61, 78, 258, 263, 274, 31417
ACTH (rheumatic heart disease): 7
anesthetic (surgery): 5
antihypertensive: 52, 317
aspirin (heart disease): 7, 78, 314, 316
chemotherapeutic (cancer, tuberculosis): 7, 13
chenodeoxycholic acid (gallstones): 315
cimetidine (ulcers): 44
clofibrate (heart disease): 316
cortisone (rheumatic heart disease): 7
dextrothyroxine (heart disease): 316
diethylcarbamazine (dog heart worm): 17
dimethyl sulfoxide (DMSO) (strains, pains): 17
dipyridamole (diabetic gangrene, heart disease): 78,
314, 316
estrogens (heart disease): 258, 316
isulin (diabetes): 258, 314
Laetrile (cancer): 17
nicotinic acid (heart disease): 316
patulin (common cold): 7
penicillin (bacterial infection): 5
phenformin (diabetes): 53-55, 59-61, 258, 314
psychopharmacologic agents (mental illness): 11
sanocrysin (pulmonary tuberculosis): 6
streptomycin (pulmonary tuberculosis): 7
tolbutamide (diabetes): 51, 53-61, 258, 263, 274, 314
surgical procedures: 15, 17, 50, 315, 317-18
bladder surgery (vesicoureteral reflux): 318
blood letting (blood pressure): 15
coronary artery bypass surgery (heart disease): 17, 317
ileal (partial) bypass surgery (heart disease): 317
liver transplant: 50
photocoagulation (eye disease): 315
other procedures: 3-7, 15, 17, 49-50, 317
bed rest (acute viral hepatitis): 49
boiling oil (wound treatment): 3-4
coronary care unit: 17, 50
dialysis (kidney disease): 50
digestive ointment (wound treatment): 3
electronic fetal monitors (prenatal care): 17
metallic rods (pain): 6, 15
smoking reduction (heart disease): 317
vaccine substances (diphtheria, polio, smallpox): 4-7
Thibodeau, LA: 187*, 440
Thistle, JL: 260
Thomas, Gl: 66*, 433
Thomas, HE Jr: 20*, 148*, 447
Thompson, DN: 120, 441
Thompson, HE: 17*. 439
Thomsen, J: 17*, 444
Thorndike, EL: 120,448
1
468
S'
Index
469
Index
Thorne, MG: 17*. 442
throwaway medical journals: 52, 305
time window(s): 122-23, 171, 305
timetable(s): 223, 225
for initiating clinical trial: 20-21, 49-50, 65
for activities and stage(s) of clinical trial: 20-21, 120-22,
145-48, 363-73
Timnick, L: 204, 446
Tolchinsky, E: 7*, 438
Tonascia, J: 17*. 192*. 438, 448
Tonascia, S: 14-16*. 268*, 443
toxic side effect(s) (toxic drug reaction[sj): see side effect(s)
training (staff): 147^8, 365, 367, 369, see also certification
transfers, see patient, transfers
treatment: 305
adherence (compliance): 78-79, 89, 161, 201 (measure
ment and how to handle), 205 (how to handle), 285,
295, 297, 305, 306-7
administration: 114-16, 200-202
assignment: see allocated (assigned) treatment; alloca
tion; randomization (randomized)
comparison: 203, 205, 306
compliance: see treatment, adherence
control treatment: see control, treatment
cross contamination: 79, 306
crossed treatments: 14, 16, 286, see also treatment,
design, crossover; treatment, cross contamination
design: 10, 16, 115, 286, 289-90, 295-97, 306-7, 331, 350
crossover: 286, 306
factorial: 115, 289-90, 297, 331
noncrossover: 295, 350
nonfactorial: 295, 331, 350
parallel: 296
uncrossed: 10, 16, 307
difference: 76, 295, 305-6
effect, efficacy: 3, 5, 306, see also side effect(s)
effects monitoring: see monitoring, treatment effects
effects monitoring committee: see committee(s) (of clini
cal trials), treatment effects monitoring
failure: 306
group(s): 12, 14, 16, 19, 25-26, 74-75 (number for sam
ple size calculation), 290, 306
interaction: 95, 306
lag: 79-80, 306
mask(ed): see mask(ed)
monitoring: see monitoring, treatment effects
protocol: 20, 114-16, 197, 201, 306
related bias: 306
standard treatment: 303
structure: 306-7, see also treatment, design
study treatment: 9, 196 (addition, deletion during trial),
304, 332, see also control(s), treatment; test, treatment(s)
termination: 216
test treatment: see test, treatment(s)
unit: 307
trial(s): 283, see also clinical trial(s)
triple-blind(ed): see mask(ed), triple
triple-mask(ed): see mask(ed), triple
/-test (of significance): 84-85
Tuchfarber, AJ: 120*. 431
Tucker, WB: 7, 448
Tukey, JW: 213, 448
Tuskegee Syphilis Study (Ad Hoc Advisory Panel): 153,
208, 448
Tygstrup, N: 24*. 187, 440, 448
Tyor, MP: 260
dichotomous: 287, see variable, binary
discrete: 288, see variable, binary; variable, polychotomous
follow-up: 290
observation: 295
outcome: 75-76, 296
polychotomous: 76, 298
primary outcome: 299
random: 300
response: see variable, outcome
secondary outcome: 302
stratification: 93, 303
subgrouping: 214, 304
surrogate outcome: 304
Veronesi, U: 49*, 432
Veterans Administration Cooperative Studies Program
(VACSP): 7, 11, 39, 41, 43, 45, 48, 50, 99, 137, 149,
238, 242*. 309, 450
Veterans Administration Cooperative Studies Program
Number 43 (VACSP No. 43): 96, 117, 209, 313-15,
319, 327-48, 354
Veterans Administration Cooperative Study Group on Anti
hypertensive Agents: 50, 173, 450
Veterans Administration Cooperative Study (Coronary
Artery Bypass Surgery) (Research Group): 17*.
444
Veterans Administration Medical Center (Perry Point,
Maryland): 354
vice-chairman of the study: see study, vice-chairman
Villaine, Castle: 3
visit: see baseline, visit; follow-up, visit
Voight, RB: 7*. 438
Vonderlehr, RA: 153, 450
J
type I and II error: 25, 72, 77-78, 81, 307
U
uncontrolled: see controlled, uncontrolled (observation)
United States Congress: 4, 8, 33, 448, 449
United States Court of Appeals for the District of Colum
bia Circuit: 55*. 449
United States Court of Appeals for the First Circuit: 5354*, 449
United States District Court for the District of Columbia:
54*, 58, 449
United States District Court for the District of Delaware:
8*. 449
United States District Court for the District of Maryland:
154, 449
United States District Court for the Southern District of
New York: 54*, 449
United States Public Health Service: 4, 7, 153, 157
United States Senate Select Committee on Small Business:
52*, 54*, 449
United States Supreme Court: 54-55*, 58, 261, 449
University Children’s Hospital (Essen): 354
University Group Diabetes Program (UGDP) (Research
Group): 7, 19, 27, 44, 50-51, 52-61 (evaluation), 66,
92, 94-95, 98, 100, 114, 117, 142, 150, 152. 157.
163-65, 170-71, 176,183, 187-90, 191*.192, 193*,
194-96, 198, 202, 204-7, 212-13, 216, 242, 247, 25659, 261-63, 269*. 274-77, 313-14, 319, 327-52, 354,
449, 450
University of California: 58
University of Chicago Press: 268, 449
University of Maryland: 154, 157, 349, 353-54
University of Minnesota: 6, 237, 349, 354
University of North Carolina: 354
University of Texas: 353
unmask(ed): 307, see also mask(ed)
how to handle: 200, 205
!
I
f
Weichert, BG: 44, 451
Weik, MH: 281*, 451
Weiner, JM: 206*, 209*, 432
Weintraub, M: 44*, 444
Weiss, DG: 46*, 149*, 433
Weiss, Robert: 354
Wells, Horace: 5
Wenger, OC: 153*, 450
West, KM: 58, 451
West Haven Veterans Administration Medical Center: 44
West Virginia University Medical Center: 349
White, SJ: 91, 438
Wilbraham, A: 4-5, 435
Wilcox, P: 120*, 431
Wilkins, Robert: 8
Williams, ME: 9, 13, 451
Williams, O Dale: 354
Williamson, J: 120, 443
Williamson, JR: 52*. 56*. 274, 440
Williford, WO: 46*, 149*, 433
withdrawal: 308
Wolfe, SM: 52*. 450
Wood, C: 17*. 446
Wood, WB Jr: 5*. 440
Woodward, WE: 157, 451
Woolf, HB: 281, 451
Working Group on Arteriosclerosis: 52, 451
Wortley-Montague, Lady Mary: 4
Wright, IS: 68, 451
Wright, JP: 68*, 442
Wright, P: 120, 127, 133,431,451
Wyndgaarden, JB: 58*, 431
Y
W
Wade, N: 237, 450
Wagner, EH: 52*, 437
Wald, A: 73, 450
Wallace, TJ: 17*, 442
Ware, JH: 187*, 290,435, 440
Warner, R: 52, 450
Washington, George: 15
Waterhouse, B: 4, 450
Watson, BL: 58, 450
Wei. U: 92, 451
I
i
provisions for : 100, 116, 164
Upjohn: 51
V
Valahcevic ZR: 260
validity issues: see generalizability and validity issues
Valle-Jones, JC: 68*. 435
variable: 67, 75-77, 82-88, 93, 214, 283, 285, 287-88, 290,
295-96, 298-300, 302-4, 307
baseline: 67, 283
Bernoulli random: 283
binary: 75-77. 82-88, 283
confounding: 285
continuous: 76, 83-84, 87, 285
•
rwftf'Tft -Tfr
w iar ■
Yale University Hospital: 44
Yates, F: 96, 437
Young, CW: 17‘, 443
Z
Zdep. SM: 125, 451
Zelen, Marvin: 91, 154, 157, 281*, 298, 307, 353, 451
Zelnio, RN: 120, 451
Zogbi, M: 271*, 447
Zukel, WJ; 8, 148, 219, 451
- Media
14150.pdf
Position: 2814 (3 views)