PREVALENCE AND INCIDENCE OF TUBERCULOSIS INFECTION AND DISEASE IN INDIA
Item
- Title
-
PREVALENCE AND INCIDENCE OF
TUBERCULOSIS INFECTION AND DISEASE IN
INDIA
- extracted text
-
RF_DIS_5_B_SUDHA
DISTR
I
WORLD HEALTH OR( iANIZATION
DISTR
. LIMITED
1.1M11 ( E
WHOrrB/97.231
Original: English
ORGANISATION MONDI ALE DE LA SANTE
- VV LO
s
-/
T i: I i-i I
PREVALENCE AND INCIDENCE OF
TUBERCULOSIS INFECTION AND DISEASE IN
INDIA:
A Comprehensive Review
by A.K. Chakraborty
BIKALPA, Koramangala, Bangalore-560 034, India
*
This document IS not issued to the general public, and all rights
are reserved by the World Health Organization (WHO),
The
document may not be reviewed, abstracted, quoted, reproduced or
translated, in part or in whole, without the prior written permission
of WHO. No part of this document may be stored in a retrieval
system or transmitted in any form or by any means • electronic,
mechanical or other • without the prior written permission of
WHO.
Ce document n'est pas destine a etre distribue au grand public et tous
les droits y afferents sont reserves par I’Organisation mondiale de la
Sante (OMS). II ne peut etre commente, resume, cite, reproduit ou
traduit, partiellement ou en totalite, sans une autorisation prealable
ecrite de I'OMS. Aucune partie ne doit etre chargee dans un systeme de
recherche documentaire ou d iff usee sous quelque forme ou par quelque
moyen que ce soit • electronique, mecanique, ou autre - sans une
autorisation prealable ecrite de TOMS.
The views expressed in documents by named authors are solely the
responsibility of those authors.
Les opinions exprimees dans les documents par des auteurs cites
nommement n'engagent que lesdits auteurs.
»
Reproduced from originals provided.
( CONTENTS
SUMMARY
01
INTRODUCTION
03
MATERIAL
04
CRITERIA FOR DEFINING INFECTION & DISEASE
05
A.
B.
05
C.
PREVALENCE OF INFECTION
INCIDENCE OF INFECTION
ARI
TUBERCULOUS DISEASE
D
ii)
iiD
lv)
v)
06
07
07
08
08
I
RESULTS
A.
07
07
Screening for bact. diagnosis
Definition of a case
Prevalence ratio by variation in
screening procedure in a survey
No. of bact. specimens and prevalence
Definition of x-ray active bact. negative case
09
09
TUBERCULOUS INFECTION
1.
1.1
1.2
1.3
PREVALENCE
Prevalence in different areas
Prevalence by age and sex
Situation over a time
09
09
09
09
2.
2.1
2.2
INCIDENCE
Incidence over time by age and sex
ARI
09
09
09
i
B.
C.
D.
E.
PREVALENCE OF DISEASE
09
1.
1.1
1.2
1.3
RADIOLOGICALLY ACTIVE CASES
Area-wise prevalence
Age, Sex-wise rates
X-ray active bact. negative case rates
10
10
10
10
2.
2.1
2.2
2.3
2.4
2.5
2.6
2.7
BACTERIOLOGICALLY POSITIVE CASES
Prevalence rates
Age-Sex distribution
Prevalence by age-sex and method of diagnosis
Change in distribution with time
Bact. case prevalence & socio-economic dimensions
Culture : Smear positivity proportion
Bacterial drug resistance
10
10
10
10
11
11
11
INCIDENCE OF DISEASE
11
1.
1.1
1.2
INCIDENCE OF X-RAY ACTIVE CASES
Definition
Incidence rate
11
11
11
2.
2.1
2.2
2.3
2.4
2.5
2.6
INCIDENCE OF BACT. CASES
Definition
Age-Sex distribution
Incidence over a time
Proportion smear positive in incidence cases
Incidence of infection In relation to case incidence
Conclusion
12
12
12
12
12
12
12
3.
3.1
3.2
3.3
INCIDENCE BY EPIDEMIOLOGICAL GROUPS
Rate of incidence of cases by infection
Rate of incidence by radiological abnormality
Incidence in epidemiological groups
12
12
12
12
PROGNOSIS OF CASES
13
1.
1.1
1.2
PROGNOSIS OF PREVALENCE CASES
Fate
Natural dynamics
13
13
13
2.
PROGNOSIS OF INCIDENCE CASES
13
3.
3.1
3.2
TUBERCULOUS MORTALITY
Proportional mortality
Mortality rates
13
13
13
TUBERCULOSIS IN PAEDIATRIC AGE GROUP
13
1.
2.
PREVALENCE
MORTALITY
13
14
ii
F.
MORBIDITY & MORTALITY: SOCIO ECONOMIC CONSIDERATIONS
1.
1.1
1.2
1.3
1.4
1.5
1.6
STUDIES NOT SPECIFICALLY DESIGNED FOR ECONOMIC
INFORMATION ON POPULATION
Information prior to NSS
NSS
Tumkur Survey
BCG Trial
Other studies
Mortality
14
14
14
15
15
15
2.
2.1
2.1.1
2.1.2
2.1.3
2.1.4
2.1.5
SPECIALLY DESIGNED BY SOCIO-ECONOMIC CRITERIA
Wardha study
By age and sex
By literacy standard
By employment
By Income-group
By area and living standard
15
15
15
15
15
16
16
3.
TUBERCULOSIS AMONG THE WORK FORCE
16
DISCUSSION
1.
17
SUMMARY OF TUBERCULOSIS PROBLEM
2.
RELIABILITY IN SAMPLING : PROBLEMS IN EXTRAPOLATION
2.1
Sampling and stratification
2.2
Reconsideration on average prevalence rates
Infection
2.2.1
2.2.2 Radiological cases
2.2.3
Bact. cases
2.2.4 Current estimates on pulmonary tuberculosis
2.2.4.1 Estimates for Andhra Pradesh, South India
2.2.4.2 Estimate on patients on treatment
2.2.4.3 Estimate on morbidity and mortality from SEARO
workshop 1996
2.2.4.4 Use of estimates
3.
14
17
17
17
17
17
18
18
18
18
18
19
19
3.1
3.2
CHANGES IN TUBERCULOSIS SITUATION OVER A TIME:
ITS MEASUREMENT
Indices for measurement
ARI
19
19
19
4.
4.1
4.2
4.3
CONTROL PROGRAMME & EFFECT-EVALUATION
Objective of an Indian programme
Continued surveillance through ARI and effect-evaluation
Epidemetric models
20
20
20
21
iii
REFERENCES
22-26
FIGURES
1 -9
TABLES
1 - 14
I
APPENDIX
ACKNOWLEDGEMENT
I-VI
VII
iv
SUMMARY
The data available on tuberculosis morbidity and mortality from various surveys in India are
summarised in this review, with a consideration of their socio-economic aspects. A brief
interpretation of the epidemiological situation in India, along with the use of the information on
epedemiology in designing the tuberculosis programme in India (NTP) is given in Appendix (vi).
1.1.
Tuberculosis problem is more or less spread all over the country. Differences in prevalence
rates of disease as observed between area to area did not appear to be statistically significant
(barring a few exceptions e.g., Tamil Nadu, Raichur and urban slums as in Calcutta).
Prevalence rate of disease was observed to be the same for the rural and urban areas (Fig.5).
1.2.
Of persons in all ages both sexes about 38% were infected. In males, almost 70% of persons
above 40 years of age were infected. About 2% had pulmonary tuberculosis, but only 0.4% could
be the average prevalence rate of bacillary cases (In 5+ age group). Mortality rate was observed
for the last time in a population survey way back in 1966. It was reported to be about 90 per
100,000 (could be half of that, as hypothesised in recent times).
1.3.
Prevalence rates of Infection, disease and mortality were more in males than in females:
a third of the cases in all ages could be females. However, of cases In females, about half could
be in the child-bearing age.
1.4.
. In a rural area in Bangalore, there was a marginally decreasing trend in the infection rate:
about 2% per year over a long period of observation ranging from 5 to 25 years, among children
(Fig 3). But the same was not seen in the adjoining areas of Tumkur, Doddaballapur and in the
neighbouring state of Tamil Nadu (the last, over a 15 year period).
(comprising about 50%
1.5.
fPulmonary tuberculosis is an adult disease. Population In 0-19 years
.
of total population) could be estimated to contain only 7% of total prevalence cases. Remaining
93% of cases could be distributed in population aged 20 years over (Fig.6). Relatively higher and
higher concentration of cases in higher age groups was observed In later surveys as compared
to earlier surveys in Bangalore rural area. In later surveys conducted in 1984 in Bangalore rural
areas, about 80% of cases were detected among those In 40+ age group as compared to about
50% in earlier surveys (Fig.8).
Prevalence rates of cases had shown almost no change over a period of over 20 years
1.6.
from different surveys in different areas (Table 9 & 10).
1.7.
The Incidence of cases was observed to be a third of the prevalence, on the average,
which could be interpreted to be due to failure of Intervention and pooling up of previous cases.
The incidence of smear-positive cases was estimated to have dwindled in 23 years in Bangalore
rural area from around 0.65 to 0.23 per 1000 (Table 15). No such observation was made elsewhere.
1.8.
Nearly 10% of all causes crude mortality in the community was contributed by death due
to tuberculosis (tuberculosis mortality rate: about 90 per 100,000). The highest proportional mortality
in women in the reproductive age could be attributed to be due to tuberculosis, higher than even
that due to peri-natal causes. Nearly 40% of female population being in the age group of 20-44
years, this could amount to be the highest single cause of death in women in thisj/ulnerable
population group.
1
1.9.
Mortality rate due to tuberculosis was observed to be considerably lower than case rate,
and was decreasing with time from survey to survey. The case rates were considerably lower than
the infection rates (mortality rate: case rate: infection rate ::90:400:38,000 per 100.000). The
observed wide gap between the rates has significance with respect to the age of the epidemic
(Fig. 15). The survival rate of cases diagnosed in later surveys were better, comparedTo thdTin
earlier rounds of longitudinal surveys.
1.10. Paediatricians were generally seeing less of miliar/, meningeal and fulminant forms of
tuberculosis. Occurrence of pulmonary tuberculosis in paediatric age-group was minimal (Table
17). However, children in 0-4 year age group was found to be highly vulnerable in a Madras slum,
having conspiculously high mortality to tuberculosis.
1.11. There was a higher prevalence In ’Kutcha' houses than 'Pucca' houses, the former
inhabitated by the poor. There was an observed association of pulmonary tuberculosis with lack
of education, lack of regular source of income as well as per capita income. Of the workers with
tuberculosis, it is recently estimated that about 52% are in the 15-44 years age group (Table 18).
In this age group, the proportion of women among the urban workers with tuberculosis (40.0%)
was estimated to be higher than among the rural (17.9%).
1.12. The rates observed in the various surveys conducted In India, have been used to estimate
the approximate average number of cases and deaths etc.. In the country, to serve as a guide
for planning for the resources for a programme at the National level (Table 19). The numbers
presented in brackets in the table are estimates, revised In the light of more recent observations.
The likely estimate of disease in India could be on average; Bacteriologically (culture) positive
cases 4.4 million (0.6%), Radiologically active bacteriologically negative 2.3 million (0.3%), and
annual tuberculosis mortality 0.42 million (45/100,000).
2
I. INTRODUCTION
In line with the current scientific development and WHO policy1. Govt, of India, had in the recent
years, re-examined the strategy hitherto followed under the National Tuberculosis Programme
(NTP). A revised NTP is currently being formulated with assistance from the World Bank and the
WHO2. While re-defining the strategy and deciding on the priorities/emphasis on programme
activities as well as their scope, the epidemiological aspects of the disease, peculiar to this
country, have to be kept in mind. For instance, because of its recognised predilection for affecting
the relatively underprivileged segments of the society3, a control programme addressed to the
needs of the country, requires also to evaluate the relevant data with particular reference to the
socio-economic and gender differentials, if any. The current report is prepared in this context,
reviewing the tuberculosis situation in India. The data could be made use of not only for planning
but also evaluation purposes, generating fresh thinking in respect of epidemiologic dimensions of
control programmes, especially keeping the circumspection of some of the noted programme
managers in view .
Considering the infectiousness of the pulmonary forms of the disease, the present report confines
itself to the data on pulmonary tuberculous and tuberculosis infection only. Available data both
published as well as unpublished to the extent available, are relied upon to present:
1.
Prevalence and incidence of tuberculosis infection as well as of sputum positive and
sputum negative pulmonary tuberculosis, by age and sex.
2.
Prevalence of tuberculosis infection and disease by socio-economic status of population.
3
c
IL MATERIAL
___________
The review mainly derives its material from the published sources, originating from some of the
leading tuberculosis research institutions of the country. However, wherever required, unpublished
work from these institutions have also been freely referred to. Apart from the main papers
originating from some of these studies, a considerable number of subsidiary papers were published
as well. These have been referred to in the text, as and when required.
Other lesser known organisations have, especially In comparatively recent times, carried out some
significant research; a few of them still to be published. These have been duly evaluated and
referred to, if found authentic enough! The major research efforts in epidemiology of tuberculosis
are described briefly in Table 1, giving the name of the study, reference, year of study, population
represented, the design and methodology (age group studied, population-screening and
diagnostic methods used).
4
(JII. CRITERIA FOR DEFINING INFECTION AND DISEASE)
(A)
PREVALENCE OF INFECTION
The estimate of prevalence of tuberculous infection in the community is based on the
interpretation of the results of tuberculin testing of persons without BCG scar. The indurations,
measured ideally, between 48 and 96 hours of the tests, are distributed in the form of a histogram
(FIGURE-1). From these distributions, it is possible to demarcate the positive reactors from those who
are negative to the test dose, since the Indurations tend to He around two discernible modes
separated by a dip. called the antimode. The nature of distribution of the indurations however
differ to an extent from study to study, depending on the tuberculin (or antigen) used, their
dosages, testing and reading variations, as well as on cross-reactions on account of the
non-specific sensitisations that may be present in a given area due to environmental factors
(including non-tuberculous mycobacterial sensitisations). In fact the dip between the two modes
is sometimes less recognisable due to a high prevalence of non-specific reactions in India,
especially south India6*2132. Due to these problems, it is not possible to have a uniform size of test
reaction identified with tuberculin test positivity In every survey.
Besides the type of tuberculin and their dosages used form survey to survey not being the same,
the problem of defining the Infected is compounded by the changing levels of demarcation
between the infected and uninfected from one survey to the other' . The problem of changing
demarcation levels from survey to survey had no doubt engaged the attention of researchers'2-33.
It is suggested after a field study that the demarcation line between the infected and
non-infected persons would require to be shifted from survey to survey to take care of the
observer variations or the rising proportions of BCG scars in the population being wrongly classified
as unvaccinated from one survey to the next, thereby affecting the classification between the
infected and uninfected33. It is not Intended to go Into more detailed analysis here owing to
space-constraint.
For this review, persons ore considered as infected, if the indurations to tuberculin test are
observed to lie to the right of the demarcation line as decided on the basis of the distribution of
the indurations, for the given study.
Non-specific sensitivity: Frimodt Moller had recognised that there could be two kinds of infection
in south India ('specific' and 'non-specific1) and that ft could have serious implications in
understanding infection risks6. The prevalence of the non-specific Infections were hypothetically
considered to be due to other mycobacterial infection. Chakraborty et al had shown that the
non-specific infection increased with age, so much as to leave only a small proportion of the
population in the age group 15-24 years without any kind of sensitivity to tuberculin32. Figure 2
shows that in the age group 15-24 years, where the proportion infected with Mycobacterium
tuberculosis was 33.2%, of the uninfected 93.4% were showing evidence of other Mycobacterial
infection. Raj Narain had shown its prevalence to be widespread In some areas23.
Thus the interpretation of the tuberculin test results in older age groups could be unreliable owing
to a high prevalence of non-specific sensitivity.
5
(B)
INCIDENCE OF INFECTION
^he results of tuberculin testing surveys are usually presented only In the form of prevalence figures
for past tuberculous Infection. Prevalence figures do not indicate, when in the past the first
infections had occurred34. Incidence of Infection, which gives the most recent tuberculosis
transmission situation in the community, Is defined as the number of fresh infections occurring in
the community between two points in time, among those uninfected initially. It calls for direct
measurement of the conversions of uninfected persons to infected status, in the intervening period
between two succeeding surveys, say a year or so apart. Information on incidence from some of
the surveys in India are reviewed here 6’0-’5 J5,
Apart from the operational constraints In carrying out the incidence surveys on tuberculous
infection again and again, there were inherent problems in interpreting the repeat test results35.
The latter was proposed to be solved by distributing the differences between the two tests in each
individual. In younger age group, where the situation was not likely to be complicated by high
non-specific sensitivity etc., the distribution was found to be bimodal. the antimode separating the
'new infected' (occurring to the right of the antimode), from those who were 'not newly infected'
(to the left of the antimode). In the BCG TRIAL for Instance, the antimode was at 12 and 10mm,
between the initial test and at retests at 272 years and 4 years respectively’5. Thus persons showing
more than 12 and 10 mm increases over the first test induration size, when retested after 27? and
4 year period respectively, were identified as freshly infected in BCG.Trial. Somewhat similar
method was adopted for NTI.LS10.
Annual Risk of Infection (ARI): The problem of obtaining data on incidence of infection could be
obviated by computing these from one or more prevalence figures, if these are obtained by
similar testing methods and if the prevalence surveys 'cover a range of ages'34. The computed
estimates of the incidence, called the 'Annual Risk of Infection1 (ARI), give the risk of tuberculous
infection in successive calendar years. These estimates may be regarded as an alternative
method of presentation of the tuberculin surveys. ARI could be defined as the proportion of the
population which will be primarily infected or re-infected (in those previously infected) with
tubercle bacilli in the course of 1 year, and Is usually expressed as a percentage or as a rate. A
tuberculin survey carried out in a representative sample or a succession of them could be
converted into the estimated risks (ARI). The ARI at a particular time indicates the current
magnitude of the incidence and prevalence of infectious cases as well. An observed decline in
the ARI would be the earliest Indicator of a decline In the epidemic cycle of tuberculosis and
would therefore be a suitable indicator for evaluating tuberculosis control programme. A rising
risk on the other hand could be the portent for a change to the other direction consequent on
adverse epidemiological situation e.g., HIV supervening.
The ARI could generally be worked out on the basis of the method described in TSRU Report34 22.
By this rr^ethod, the ARI (R) for a group of average age (A) was derived from the prevalence (P)
by R = 1 - (1-P) 1/A
Incidentally, the corroboration that ARI represents annual Incidence of infection is available from
the study by Chakraborty et al in the LS-Follow up’3(Figure 3).
6
(C)
TUBERCULOUS DISEASE
(i)
Screening for bacteriological examination: In the classical epidemiological surveys for
estimating the case-load, the population Is x-rayed at first, mostly using a mobile mass miniature
radiography (MMR) unit. Only the persons with shadows on MMR of chest assessed as abnormal,
usually by two readers Independently, are subjected to sputum test (x-ray-screening). Alternatively,
the persons in the community are questioned for presence of chest symptoms (cardinal chest
symptoms being cough, pain chest and fever of one or two weeks duration and haemoptysis);
those having symptoms (symptomatics) are subjected to sputum examination
(symptoms-screening). In the MS, NSS, TUMKUR STUDIES, MTI-LS, BCG-TRIAL and NEW DELHI STUDY,
the x-ray screening procedure was adopted6-7<L9’0’52’G'able 1). On the other hand, in the
WARDHA. RAICHUR, TRC-TRIBAL and CARNICOBAR Studies, symptoms-screening procedure was
followed2526■27-28. In the NTI.PERI URBAN STUDY and LS-FOLLOW UP, again, the population was
screened for their eligibility to undergo sputum examination, using a criteria, different from the
above mentioned two, and casting a wider net'424.
Varying screening procedures from survey to survey affect the observed prevalence rates and
therefore the comparability of results between one survey and the other3457.
(ii)
Definition of a bacteriological case: It was shown that cases positive by smear alone,
detected in an epidemiological survey, were hardly the real cases of tuberculosis, the
reproducibility by a second smear or by way of radiological confirmation as active TB in these
cases being very poor'5,33. In the BCG-TRIAL of the 567 persons positive on smear but negative on
culture, 76% had only 1-3 bacilli on both smears examined. The x-ray confirmation in these persons
having actively tuberculous pulmonary shadows was a mere 13%'5. Cases with positivity on culture
only, without radiological abnormality at the corresponding survey, were also a group whose
tuberculous etiology was not acceptable in the NTI-LSn. In most of the epidemiological surveys
carried out at the NTI using the classical survey method, a case of tuberculosis was defined as a
culture positive person, with a radiological abnormality at the current survey". Persons not
confirmed on culture (l.e., smear alone positive) were not defined as bacteriologically positive
cases, even if they had an x-ray-abnormality, (they were however eligible to be defined as x-ray
positive cases, if shadows on their chest-radiography were classified as actively tuberculosis).
Three categories of bacterlologically positive cases were presented in the BCG-TRIAL: © positive
on two cultures (ii) positive on one culture (iii) smear-positive, excluding those showing 1-3 bacilli.
There was, of course, the additional group of radiologically positive bacteriologically negative
group15. In the NSS, the sputum was collected contingent on an abnormal shadow being present
on chest x-ray. It appears that the definition of a bacteriological case, under NSS and NTI classical
surveys were, essentially similar. In the Madanapalle study, on the other hand, there was too much
stress on radiological examination, bacteriological results being comparatively of less prominence6.
(iii)
Variation in screening prodecure and diagnostic fest influencing prevalence rate: In some
of the more recent prevalence surveys carried out in India, mobile MMR units were not employed
to screen the population (MMR-screenlng). An alternative method of screening the population
was followed in order to identify the population eligible for sputum tests, by questioning the
persons on the presence of cardinal chest symptoms on a house to house basis, followed by
sputum tests, either culture or smear microscopy2425,2428,30.
A recent NTI-study, addressing itself to the problem of comparability of culture positive prevalence
rates obtained by employing population-screening by MMR vs. symptoms questioning has shown
that the efficiency of the above two screening procedures in diagnosing culture positive cases
was different from the best estimate obtained of employing multiple screening (MMR and/or
symptom question)24 The correction factor to obtain the best estimate of culture positive cases
7
from that obtained on symptoms screening could be between 1.30-1.63 (approximately 1.5’).
However, the significant finding from this study, which to an extent seemed to extenuate the
problems of estimating cases based on variable screening procedures, was that the prevalence
rates of the culture positive cases in the community were the same, irrespective of whether
symptoms or MMR screening was the procedure of choice in screening the population for
identifying the eligibles for sputum test. The same was found to be good even for the estimates of
smear positive cases. Moreover, it was observed that there were no age-sex wise differences in
observed rates between the two procedures. It was Immaterial for the estimate on the basis of
symptoms screening, whether the Interviews of the population for Identifying symptomatics was
carried out by a qualified social investigator or a field health worker, as long as adequate training
on the procedure was given prior to the deployment for the investigation. The above could be
utilised in evaluating and drawing inferences from some of the latest studies on prevalence.
(iv)
No.of bacteriological specimens examined and prevalence of cases: The number of
specimens of sputum collected from among the eligibles after screening were variable from survey
to survey, which would affect the estimate of the bacteriologically positive cases in the
community. The extent of difference in rates could be estimated from the finding, that instead of
two sputum specimens, if four were examined in a survey, nearly 25% more number of cases could
be added to the prevalence39. This was found to go as high as 39%, when eight specimens were
examined37. Needless to say that the findings from surveys with variable number of sputum
specimens need to be suitably revised, before comparative assessment of the rates are made.
(v)
Definition of a radiologically positive bacteriologically negative case of tuberculosis: Apart
from its use as a screening tool for the identification of persons eligible for bacteriological
examination in a tuberculosis survey in the community, MMR-results are also relied upon to
diagnose and label persons as radiologically positive cases of pulmonary tuberculosis ('persons
with active and probably active shadows'-NSS). Estimates of the problems of tuberculosis on this
basis has some serious limitations, principally owing to the Inherent 'reader-variations' in
interpretation of the x-ray shadows (as well as their Identification) their tuberculous etiology as well
as the activity. The procedure, usually followed, was to have two independent interpretations of
the chest x-ray shadow on MMR (single picture) and to collect sputum from persons with a
shadow (‘x-ray-screening'). The disagreements between the two independent interpretations were
then subjected to a third reader's assessment (or, an umpire reader's arbitration) regarding the
nature and etiology of the shadow. However, despite the panel of readers, used in these surveys,
being composed of highly experienced persons in their field, there was considerable disagreement
in interpretations’5.
A five-year follow up of the persons, diagnosed to have active tuberculous shadows (as also other
MMR abnormalities in their chest), was carried out by repeated x-ray of chest, smear exam for AFB
and culture for M.tuberculosls (three examinations at 1’Z?, l’/2 and 2 years after the initial one)40.
Based on an overall consideration of all the follow up Investigation results, at one time, a panel
of x-ray-readers had Jointly re-evaluated the Initial MMR-interpretations (Joint Parallel
Reading-J.P.R.). Of the initial 385 sputum negative x-ray active persons Identified initially, only 22.0%
could really be classified to have sputum negative x-ray active tuberculosis at the first survey,
when such a re-evaluatlon of the MMR-shadows was carried out by J.P.R.40. Unpublished
i information from the TRO Madras on the BCG.TRIAL FOLLOW UP studies recently corroborated this;
inot even a third of the radiologlcally positive cases were later assessed to be having active
Ituberculosis needing treatment41.
It could be suggested that the generally accepted prevalence rate of radiologically active
bacteriologically negative pulmonary tuberculosis In India, largely estimated on the basis of data
from the NSS and other similar studies (using a single MMR and the pulmonary shadows evaluated
by 2 readers and umpire), could be nearly three times the likely estimate of the problem.
8
IV. RESULTS
. (A) TUBERCULOSIS INFECTION
(PREVALENCE AND INCIDENCE)
1.
PREVALENCE OF INFECTION
1.1.
Prevalence in different areas (Tables 2 & 3): That the tuberculosis infection was more or less
spread throughout the country was hypothesised on the basis of studies done by Ukil and from
information available from the mass BCG - Campaign results42. However, differences were
observed from area to area, e.g., hilly areas. In later surveys carried out in parts of India, different
epidemiological situations could be observed between areas In adjoining states (Tamil Nadu and
Karnataka)' 015, contiguous districts in the same state (Tumkur and Doddaballapur in Bangalore
district)822 and between areas within the same district (economically backward northern part of
Tumkur compared to the southern part: infected, 46.0% and 30% respectively8), as also between
two panchayat unions of Tiruvallur district In BCG.Trial: Infected 1-9 year old, Kodambathur 8.7,
Thiruvalangadu 12.3 percent in 197916 (Not on Table). Prevalence by socio-economic criteria is
presented separately under Section IV F (see infra).
1.2.
Prevalence by age and sex (Tables 2 & 3): Between 25-38% of population in all ages both
sexes were infected, as per data available from Tumkur Study (males 42.8, females 33.9 per cent)
(Not on Table)8, infection rising with age, more in males (Fig.4)6-8J0. Prevalence rates were almost
similar in males and females upto about 14 years In age, after which males had higher
prevalence. Whereas the peak in the males was observed at around the age of 30 years, it was
so in females by about 40. Similar phenomenon was observed both in Tumkur, as well as in the
BCG Trial area (Not presented)'6. In the BCG.Trial overall prevalence was found to be 50% (males
54.0, females 46.0%)16.
1.3.
Situation over a period of time: There was a declining trend in prevalence with time in
NTI.LS area (Appendix table i), not seen elsewhere.
2.
INCIDENCE OF INFECTION
2.1.
Incidence by age, sex and over time: The incidence of infection was found to be 1% per
year and did not seem to vary age-wise In 0-4, 5-9 and 0-14 year age group in the NTI.LS (Table
4)10. In rural areas of Bangalore there was a marginally declining trend observed over a period of
5 years (Fig.3)10.
2.2.
A.R.I.: ARI calculated from different parts of the country was found to be between 1 and
2% (Tables 5 & 6). The decline In Bangalore area at about 2% per year annually for over 23 years,
as seen in Figure 313, was not seen elsewhere, including in BCG.Trial area over 15 years period of
observation .
(B)
PREVALENCE OF DISEASE
This report confines itself to pulmonary tuberculosis. No significant population-based study on extrapulmonary TB is available, except study on glandular forms in small population groups of Nicobar
island and Sheriff garden In Bangalore28M. Being In specific and selected groups, data from them
could not be representative.
9
1.
RADIOLOGIC ALLY ACTIVE CASES:
1.1.
Area-wise prevalence: Table 7 presents the prevalence of pulmonary tuberculosis,
including bacillary cases, across the country from NSS data. Table 8 presents the bacteriologically
negative radiologically active case prevalence, derived from rt. Tables 9 and 10 present the rates
from limited area surveys conducted after the NSS. Prevalence rates of disease did not appear
to be significantly different from area to area, whether rural, seml-urban or city. There were
however pockets of high prevalence, possibly related to economic situation. For example, the
prevalence rate of 58 and 50 per thousand in Block No.39 and 8 Calcutta city (slum-dominated)
was higher than 2.48 per thousand in Block No.34. Similar was the case with Delhi city.
1.2.
Age-Sex-wise rate: The rates were seen to be rising with increasing age in both sexes.
Irrespective of areas of study (Fig.5).
I
1.3.
Radiologically active abacillary case-rate: The bacteriologically negative radiologically
active case prevalence rate varied from 10-19 per thousand. These could be over-estimates to
the extent of about 75% (See section III C.V.),40. The rates, after correcting for possible
over-estimates, is presented in the last column of Table 8. The MTI-rural study which had followed
the JPR method of interpretation of MMR and the Madanapalle study, also following a similar
method, had reported prevalence rate of 5.40 and 4.23 respectively (Table 9)6 4043.
2.
BACTERIOLOGICALLY ACTIVE CASES:
2.1.
The prevalence rates are presented in Table.7 (for NSS), Table.9 (for, limited area studies
using MRR-screening) and Table 10 (for, limited area studies with symptom-screening). There was
no difference between the rates, either from area to area or with time, provided the same method
and criteria were followed. It should be appreciated that large sample sizes of population are
required to be followed up, so as to be able to discriminate small differences in the already low
case-prevalence rate of about 4.0 per thousand^47 (Appendix Table ii).
2.2.
Age-sex distribution: Figure 5 shows age, sex distribution by areas. The rates rose with age
in both sexes, more in males. Based on the age-distribution of the cases, it could be estimated
that the population in 0-19 years age (comprising 50% of total) could contain only about 7% of
total prevalence cases (Fig 6). Remaining 93% of the cases could be found distributed in th©
population aged 20 years and over (l.e.. In remaining 50% of population). On the other hand, of
the total cases, nearly 50 per cent could be distributed among the population aged 40 years and
over, constituting about 20 per cent of the total population.
Prevalence of bacteriological cases In each of the age and sex groups studied, is presented in
Table 11, as computed on the basis of BCG.Trial. Prevalence in each of the age groups was much
’e^s ,r}
Of the total bacillary cases 79.0% could be in males and 21% in females in
G. rial. Whereas males in 20-54 years age group i.o., the wage-earners, had constituted 39%
o to fa! male popu.ahon, nearly 70% of all male cases wore In this age-group. For females, on the
nSlvh4n%
In.1110 ^productive age group of say 20-44 years, had constituted
<
Lh
° a fornal® Population, with about 56% of all the female bacillary cases
distributed in this age group.
and ca,e9°ri9s of case: Figure 7 presents the distribution of cases
bvcriteda
°"a~
SoS pSss
ssTsas01"
10
voa,!-1 hM
10
4
2.4.
Change in distribution with time: The observed age-sex distribution of cases with time in the
2.4.
NTI.LS area wqs seen to reflect a change In situation as shown in Figure 8 47. In the first of the
longitudinal surveys in Bangalore rural area, about 50% of cases was found In the population
above 40 years in age, constituting about 20% of total population. About 43% of the cases were
distributed in 20-39 year age group, constituting about 39% of population. In later surveys in the
same area, about 70-80% of cases were found among those in 40+ age group. Relatively higher
and higher concentration of cases was observed to take place in higher ages in later surveys.
2.5.
Bacteriological case prevalence and socio-economic dimensions: Table 12 along with
Appendix Table (iii) and (iv) present the various socio-economic categories in the population and
prevalence rates in them along with the proportion of cases contributed by each to the total
prevalence cases in the community25 These aspects are dealt with separately under Chapter F
(see infra) of this Section.
The Wardha study25 and the NTI.peri -urban study24 had shown, that as different from other studies,
the rural areas had higher prevalence rates. The respective rates per thousand were:- Wardha
study: Urban 1.62, Rural 1.98, NTI.Peri-urban: 3.4 (interval estimate 2.7 to 4.2) as compared to
Bangalore rural 5.7 (interval estimate 5.4 to 5.9). Whether the differences were due to
urban-peri-urban and rural divide or had represented a favourable trend In favour of the urban
group is a matter for discussion.
2.6.
Culture: Smear positivity proportion: Table 13 shows the data from NTI.LS14.Inspite of the
fact that there was no difference in the age standardised case-prevalence rates from survey to
survey in the area for a 23 year period, the smear positivity among the prevalence cases had
come down to 16% at the survey in 1984, from the initial proportion of 47.0% observed in 1961.
More importantly, notwithstanding a very comprehensive screening method in the 1984 survey,
the smear positive case prevalence rate had also come down to 68 from 189 per 100,000,
.observed in 1961 survey. It is debatable, whether this could be attributed to the National
Programme, implemented in the area after the Survey 4, in 1970. The trend may be the result of
a low-efficiency programme running over a long term in the area14. In other surveys conducted
elsewhere in India, the smear positivity proportions were more or less similar to NTI.LS first to fourth
surveys (e.g., Bangalore Peri-urban area 44.9 to 46.7%24, survey 1 of BCG.Trial 57.4%u).
2.7.
Bacterial drug-resistance: Table 15 and Figure 9 shows the bacterial drug resistance in the
community. Unpublished information from New Delhi TB centre, Bangalore Peri-urban area and
rural area in Raichur (Karnataka) shows high initial resistance to INH in recent times.
(C)
1.
INCIDENCE OF DISEASE
INCIDENCE OF RADIOLOGICALLY ACTIVE CASES
Definition: Sputum negative persons with a normal MMR of chest or those with
1.1.
non-tuberculous or inactive tuberculous shadows at the initial survey, who had a radiologically
active tuberculous shadow at a later survey, but sputum was negative, were classified as
incidence of radiologically active bacteriologically negative cases (also called "suspect cases").
1.2.
Incidence rate: In the Nil rural area study, of 35,876 persons aged 5 years and over,
incidence was found to be 2.24 per thousand between two points of observations in 3 months on
JPR43 (not on table). The incidence rates, as calculated for 3 months, was not different from that
estimated for one year44.
11
2.
INCIDENCE OF BACTERIOLOGICAL.CASES:
2.1.
Definilion: Culture negative persons or those without a radiological opacity in a previous
survey, who were detected to have culture positive disease at a later survey, were termed as
"bacillary case incidence" between the two surveys40. The respective Incindence rates expressed
as an annual average between two surveys are available from New Delhi, NTI.LS and BCG.Trial
areas2021’’0’18. On the other hand, new cases detected in the community following continued
surveillance are reported from Nicobar and CMC-Vellore studies 28’45 (Direct observation of
incidence).
2.2.
Incidence of cases and age-sex distribution: Table 16 shows the incidence from various
surveys in India. Age-Sex distribution of incidence cases over a 5 year period is shown in Figure 10.
The incidence in BCG.Trial area was higher than elsewhere, as it was for prevalence also. On an
average, incidence of cases between two points of observation however, was a third of the
prevalence at the initial point of observation, similar to that In NTI.LS. In the NTI.LS about 50% of
total incidence was observed to be in males aged 35 years and over. In females the contribution
was IS^^’0”.
2.3.
Incidence of cases with time: In the NTI.LS, among the younger population in the age group
5-4 and 15-34 years, a decrease with time was observed, with corresponding rise in those aged
2.4.
Proportion of smear positivity In Incidence cases: The proportion in Incidence cases,
positive on culture alone (negative or smear), in relation to culture and smear positive cases in the
NTI.LS was 75.0% and 55.8% for surveys 2 to 3 and 3 to 4 respectively (not on table)10”. The
incidence of smear positive cases was estimated to have declined in 23 years in the NTI.LS area
from 0.65 to 0.23 per 1000 (not to table)14. No such observation was made elsewhere.
2.5.
Incidence of infection in relation to that of casesJable 4 shows, that the relationship of
annual incidence of infection of 1% as observed In various age groups upto 14 years,
corresponded to smear positive case incidence of 45/100,000 in survey 1 of NTI.LS10 ”, and in line
with observations made elsewhere49. However, the relationship was not stable in later surveys of
NTI.LS (not presented).
2.6.
Conclusion: The considerably higher prevaience
prevalence of cases as compared to Incidence (3:1)
could oe Interpreted to be due to failure of intervention and consequent pooling of previous cases
in the community.
3.
INCIDENCE OF CASES BY EPIDEMIOLOGICAL GROUPS:
3.1.
r
* ■
Case •incidence
by infection: Figure 11 depicts a higher rate of case incidence among the
infected, rising with age, especially in males.
. 3.2.
Case incidence by radiological abnormality: Bacteriologically negative persons having
I rodiologically active tuberculous shadows had the highest Incidence rate of bacillary cases (2.6%
1 per year), on a five year observation, more so if they were tuberculin positive”.
'
3.3.
incidence of cases by epidemiological groups: Various population groups, classified by
some epidemiological attributes, and incidence of cases in them, are presented in Figure 1 250 The
two highest risk groups had constituted only 5.6% of the population size, but still contributed 46%
to the total new cases arising in the population in a year. However, one must also balance it with
the observation that 48% of the new cases would arise from amongst those who had no shadow
in their chest x-ray, albeit, with a much lower rate, and was attributable to the relatively larger
zj > A t ve Jc,
^7.
12
group size (89% of population). It could be concluded that surveillance of the two highest risk
groups could be useful, if they were action-taking50.
(D)
1.
PROGNOSIS OF CASES
PROGNOSIS OF PREVALENCE CASES:
Fate: Fate of cases in a situation without active intervention is presented in Figure 13. Based
1.1.
1/2
year period observation, 20%, 18% and 62% of the cases were dead, became sputum
on 1
negative and remained sputum positive respectively, in a year's time, subject to the hypothesis
that the dynamics within the I’/z year period, had remained uniform.
1.2.
Natural dynamics: Figure 14 Is a stylised presentation of the dynamics of tuberculosis in the
community without active intervention, • utilising the Information available from NTI.LS47. The
proportion excluded annually from the existing pool of cases by reasons of death and cure, was
rounded up to be nearly a third of the initial pool (d:20%, c:18% of pool total: say, 1/3 of pool).
The exclusions would get balanced with the estimated addition by way of annual incidence (i:
to the extent of about 1/3 of size of initial pool). Thus, year to year, the size of the pool would
remain unaltered, and 2/3 of It would be formed by the continuing cases (the so-called "left
overs"5’).
2.
PROGNOSIS OF INCIDENCE CASES:
Of the incidence cases in the NTI.LS between surveys 1 and 2, the proportion of dead, cured and
remained positive were 14%, 52%, 33% respectively, thus having a better prognosis than that in
prevalence cases” (not on table).
3.
TUBERCULOSIS MORTALITY:
Proportional mortality due to tuberculosis: Nearly 10% of all causes crudejrrortality in the
3.1.
community was shown to be due to tuberculosis In the NTI. LS area (Table 17)42.
3.2.
Mortality rate due to tuberculosis In the community: Mortality rate in the NTI.LS area was
reported to be 95/100.000, not changing with time In 5 years (1961-68)52 (Table 17). In the New
Delhi area, between 1972-76, the rates were about 40 per 100,000, consequent on a well
organised programme20 21. In the Madanapalle area, the mortality was reported to have declined
from 253 (in 1949) to 64 (1952-53) and then to about 21 per 100,000 (in 1954-55), the latter
hypothetically attributed to be due to a well-organised programme6. Murray C.J.L., in his draft trip
report (Geneva WHO CDS 1992) estimated that women In their reproductive age (15-44 years)
had about 70,000 deaths from tuberculosis, every year, higher than that attributed to peri-natal
causes related to pregnancy and child-birth (un-published). Further, unpublished report derived
from the Sample Registration System (SRS), and available with the TB section. Government of India,
estimates around 400,000 annual deaths from tuberculosis in India. The currently estimated
mortality due to tubercuolosis could be In the range of 50-80 per 100,000 Le., between 0.3 to 0.5
millions annually (say 0.4 milllons).The above projection agrees with that by Dholakia w.
(E) TUBERCULOSIS IN PAEDIATRIC AGE GROUP
1.
PREVALENCE:
Pulmonary tuberculosis in children was reported to be less of a problem in the paediatric
population as compared to those aged 15 years and over54,56. Table 17 presents prevalence and
incidence of cases from NTI.LS, upto the age 14 years. Average rate of incidence of
bacteriologically positive cases among children was ’/5 of prevalence, whereas it was ’/3 in those
13
oged 15 yeors and over. In the Nicobar study the best estimate of all forms of tuberculosis in
children was 0.6%, including histo-pathology confirmed glandular tuberculosis (smear positive case
prevalence 0.4%)5d.
2.
MORTALITY:
In a study by Rajnarain and Diwakara, considerably higher annual mortality rate was reported from
a Madras urban population group aged 1 -4 years old (239 per 100,000) as compared to between
52-55 in the rural areas under BCG-Tiral and NTI.LS57. Of the total causes deaths in that age group,
nearly 50% were estimated due to tuberculosis, as against between 4 and 5% in the rural areas
in the same age group. The study had concluded that the special risk pertained only upto 4 year
of age and not beyond.
(F)
1.
MORBIDITY AND MORTALITY: SOCIO ECONOMIC CONSIDERATIONS
STUDIES NOT SPECIFICALLY DESIGNED FOR ECONOMIC INFORMATION ON POPULATION:
1.1.
Information prior to NSS: Not much reliable information on infection by socio-economic
criteria are available in general. It was however known from BCG vaccination campaign results,
upto early 1950s. that the Infection rates were higher In Industrial towns than elsewhere - 50 per
cent of those aged 10 years and more and 75 per cent of those 15 years and more were
Infected7.
1.2.
NSS: In the NSS the only information collected on the economic strata of the population
was place of residence (urban/rural) by type of dwelling houses ("Kutcha"/ "Pucca" houses)7. In
the cities, there was higher prevalence of disease (x-ray active as well as bacillary) among persons
living in the 'Kutcha" houses than in the ’Pucca" houses. The differential in prevalence rates by
type of houses did not exist in rural areas. It was taken to indicate the possible effect of economic
and sanitary conditions. The NSS had also shown that there were areas within a city (as in
Calcutta), where the prevalence of tuberculosis was as high as 40 or 50 per thousand. These areas
were invariably inhabitated by the poorest segment of the population. For example. Block 39 and
8 of Calcutta city had prevalence of 58 and 50 per thousand respectively, against 2.48 per
thousand in Block 34 (comparatively affluent) as being the lowest estimate. Delhi city also had
several blocks with prevalence between 30 and 50 per thousand, and so had other cities too in
every zone.
Even though, generally speaking, the bacillary case prevalence rates in urban and rural areas
were similar within each zone. Bangalore city, forming part of the Madanapalle zone, had a lower
prevalence (2.40 per thousand: confidence limit 1.64-3.16) than the rural areas of the zone (6 11
per thousand; confidence limit 5.02 - 7.20). In all likelihood. Bangalore being one of the
economically betterof cities, with considerably less slum problem within its environs, had something
to do with it. The Information Is further substantiated by the observation in recent times that a
sample population of Bangalore peri-urban area had a lower case prevalence rate than found
in the sample survey conducted in the Bangalore rural areas24.
1.3.
Tumkur Survey: The tuberculosis prevalence study In Tumkur district, which had immediately
fo.lowed the NSS had, as one of Its objectives, to investigate the area-wise difference in
tuberculosis case-rates8. It had made the significant observation, that the southern half of the
district, consisting of six subdivisions (talukas), had "strikingly* large differences in the tuberculosis
situation over the southern four talukas. (Tuberculous infection rate all ages: Nothern half 46,
Southern half 30%. Prevalence of x-ray active cases: North 2.3, South 1.4% and Prevalence of
bacillary cases: North 0.58, South 0.24%). Moreover, there was a preponderance of male bacillar/
coses in the north than In the south. It Is well known that the northern part of Tumkur district is
comparatively backward than the sourthern areas studied. There was no difference of course due
♦k c2Lerag0 by ag0' sex or dU0 to sfee of villages between the two zones in the study, to which
the differences could be attributed.
14
1.4.
BCG Trial: In the Chingleput study of BCG.TRIAL there Is the consistent finding, that in terms
of prevalence of Infection, the problem of tuberculosis was higher in Thlruvalangadu area as
compared to Kadambathur area, the annual risk of Infection In the 1-9 year old children in the
former being 1.6 times higher (prevalence of Infection 10.9 and 7.6% respectively)'6.
1.5.
Other Studies: In the study conducted In the tribal area In Madhya Pradesh
the tribal
population had a significantly higher prevalence rate of bacteriologically positive cases (15.0 per
thousand) compared to the non-tribal residents of the same area (9.7 per thousand). This was the
nearest approximation to a study of the bacillary case rates by economic stratification, as one
could have, without consciously designing for it. provided the hypothesis is true that the tribals
were economically weaker in the area than the non-tribals. In the Isolated Andaman Nicobar
islands territory, the prevalence of smear positive cases among the primitive tribal population of
Nicobar island, was found to be 4.1 per thousand, higher than that seen In CMC Vellore study
(Table 10) and NTI.LS (Table 13).
Mortality: Apart from these studies, one could consider Rajnarain's finding57, that there was
1.6.
a considerably higher annual tuberculosis mortality rate in children aged 1-4 years old in the
Choolai area in Madras city, dominated by slums (239 per 100.COO), as compared to the rural
areas of BCG.TRIAL or in NT1.LS In Bangalore district (between 52-55 per 100.000).
2.
SPECIALLY DESIGNED BY SOCIO-ECONOMIC CRITERIA:
2.1.
Wardha Study: The survey carried out in Wardha district is the only tuberculosis prevalence
study, specifically designed to observe the socio-economic aspects of population, as related to
tuberculosis prevalence rate25. It gives information on various socio-economic strata, re: place of
residence (urban/ rural); type of dwelling house (kutcha/ pucca), education, occupation and
income strata. (Table 12 and Appendix Tables iii & iv). Of all the socio-economic groups studied,
the highest prevalence rate was found to be among the urban female professionals (8.49 per
thousand) as shown in Appendix Table Iv. The rural women service group (5.20 per thousand) and
women cultivators (6.80 per thousand) were also particularly vulnerable.
Other salient features of the study are summarised as under:
2.1.1. By Age and sex: The prevalence In males was higher than in females (2.39 vs. 1.32 per
thousand population), both In urban and rural areas (Not on Table).
In rural areas, the prevalence was higher than In urban areas. The males had twice the
prevalence than among females.
The age group 55-59 and 60+ years had the highest prevalence, rising with age in males. In
females the rise was upto 39 years, falling after the age of 50-55 years.
2.1.2. By Literacy standard: The tuberculosis prevalence per thousand population was the highest
among the illiterates (2.49) and lowest among the graduates (0.74) (Table 12). However in the rural
population, the high school group (1.79) had higher prevalence than those educated only upto
primary level (1.42). This has been interpreted to be due to the hardship that a rural student has
to go through, leaving rural environs in pursuit of higher education.
For calculation of the respective prevalence rates by education levels, the eligible population
group in the denominator were considered (Narang P, Personal Communication).
2.1.3. By employment: Prevalence per thousand (Table 12) was the highest among the
professionals (including the petty shop keepers) (4.08) followed by cultivators (3.12) and
agricultural labour (2.45). The housewives had a comparatively low prevalence rate, but had
15
contributed a high proportion to the total prevalence of cases, owing to the group-size being
relatively larger. Of all cases In both sexes, about 70% were among those either classified os *
non-worker (24.9%), cultivators (24.8%) or agricultural labour (21.4%). All these could be persons
without a regular source of Income. Of the total cases In females, about 48% was among those
unemployed (called non-worker* which Included housewives), followed by agricultural labour
23.9% and cultivators 15.2% (Not on Table).
2.1.4. By income-group: The prevalence of pulmonary tuberculosis showed inverse relationship
with increase in per capita income from 2.04 per thousand in <Rs.lOO group to 1.09 in ) Rs.300
group (Table 12). The inverse relationship had held good both for urban and rural population
(Appendix Table iv).
In urban areas, prevalence in >z Rs.300 group was fairly high (2.18 per thousand).
The grouping by income, used in the survey was decided after a preliminary socio-economic study
in the rural areas of Wardha. The economic scenario, reflected by per capita income is different
from India as a whole and gives a grim picture (Narang P, Personal Communication).
By arGa and ,ivin9 Standard: The prevalence (2.4 per thousand) as found in those living in
Kutcha houses in urban areas was the highest. In rural areas, there was no difference in rates
between Kutcha" - "Pucca" houses (Appendix Table iii). It was probably due to small number of
pucca houses and also to the almost similar quality of life and the level of health consciousness
among rural population, affluent or otherwise.
3.
TUBERCULOSIS AMONG THE WORK FORCE
Dholakia'4 contends that the proportion of workers among the tuberculosis
ukc;y
cases is likely to be
more than among total population. This may have more to do with the distribution by age-sex in
e population than anything else, l.e., more among males than In females and among adults
han in children. Evidence, according to him. Is lacking to assume a differential prevalence of
tuberculosis among workers and non-workers. His estimate of workers with tuberculosis in the
oase-year 1993-94, among population aged 15 years and over is presented in Table 18 Of the
workers estimated to have tuberculosis in India, about 52% were in the age group 15-44 years. In
is age group, about 40% of the workers with tuberculosis were estimated to be women in th®
urban areas The proportion, however, was only 17.9% in rural areas. There was much lower
proportion of women among workers with tuberculosis in higher ages, especially in urban areas.
16
V. DISCUSSION
1.
SUMMARY OF TUBERCULOSIS PROBLEM:
The average rates of tuberculous morbidity and mortality, hypothetically considered to be relevant
to the country as a whole, are utilised to estimate the problem in absolute numbers, so as to serve
as a guide for planning for resources at the national level (Table 19). However, it is recommended
that the estimates be considered in the light of the comments made in para 2 (vide infra).
A brief interpretation of the epidemiological situation in India, along with the use of the
epidemiology in designing the tuberculosis programme in India (NTP) is given in Appendix (vP.
2.
RELIABILITY OF DATA AND PROBLEM OF EXTRAPOLATION:
2.1.
Problem in Sampling and identification of strata: The near continental dimension of India
and the variability in the socio-economic situation from area to area and even within as small an.
area as a district, raises the question of representativeness of the data and the wisdom of
extrapolation of the findings obtained from epidemiological studies carried out in limited areas to
other areas or groups. While some of the surveys had taken samples for ensuring a degree of
representativeness, others had arbitrarily selected the population. Even in the sample surveys,
owing to lack of hypothesis, an appropriate stratification by relevant variables could not be
carried out, thereby rendering the study population less representative for the area, to that extent
Examples of both NSS and Tumkur surveys could be cited in this respect810. No doubt suggestions
were available from the results of these studies that the prevalence rates could vary by the
socio-economic strata in society, yet the Information could not be reliable for lack oi
representativeness of samples. The data from the Wardha study25 as well as Dholakia's study53,
however, could now be useful in this regard. Hypothesis on socio-economic aspects, gendei
differentials and areawise distributions of cases besides the pressing problem of tuberculosis
among the workforce could now be formulated for designing representative sample surveys ir
other areas of the country, apart form putting it to use In programme planning. Priorities may have
to be redefined, keeping in view the relatively deprived sections of the society. It appears that
tribal or urban slum dwelling population groups, living on the fringes oHhe society, could neec
specially monitored programme, may be on the lines of that followed in the Nicobar or North
Arcot tribal population groups28 29-45. The higher prevalence in the urban slums, as observed in the
NSS. needs to be viewed now In the light of the projection that the proportion of urban slum
dwellers. already high at about 37%. Is likely to escalate, to be about 50%, by the turn of the
century.
2.2.
Reconsideration on average prevalence rates: The review has highlighted the need fc:
re-consideration of the average prevalence rates for the country.
2.2.1. Infection: The question of considering the infection rate in the country to be 6070%, in both
sexes of all ages needs to be reviewed. Given the inappropriateness of tuberculin testing fa
discriminating the population as Infected and uninfected In the age group beyond 10 years (a
say 15+ years) in age, makes the estimates on infection rates beyond this age inaccurate. Fror
Figure 2 it could be seen that the proportion of positive reactors to tuberculin could be c
phenomenon related to age. In the age group of 15-24 years, 93% of those who were negative
reactors to 1 TU, were still positive reactors to a higher dose, i.e., its only a small proportion whc
would be negative reactors to any dose of tuberculin. Because of this problem, the estimates a
infection could only be made In younger age group, with any degree of accuracy.
17
2.2.2. Radiological cases: The revised estimates of radiologically positive bacteriologically >
negative tuberculosis prevalence as per NSS are presented In Table 8 along with their revised
estimates, correcting them as suggested on JPR40 (see section III.C.V.). As per the corrected rates.
the bacteriologically negative but radiologically positive cases prevalence In the community
would vary from 2.6 to 4.7 (say, 3.0 per 1000, as average), instead of 10.3 to 18.6 (say, 16 per 1000)
without correction (Table 8).
2.2.3. Bacteriologically positive cases: The above Information, taken together with the finding,
that the estimate of bacteriological case rate (culture positive) should be corrected to be 39%
more than the rate obtained from surveys examining only two specimens of sputum37, would mean
that the proportion between bacteriologically positive cases and radiologically active and
bacteriologically negative cases should be revised. The bacteriologically positive case rate in the
NSS, taken to be between 2 and 8 per thousand (say, 4.0 on an average) varying from area to
area, should be corrected to be approximately between 3 and 11 per thousand (say, 6.0). The
\7 ratio between bacillary cases and radiologically active bacteriologically negative cases would
A then no longer be 1:4 (l.e., 4.0 vs 16.0 per 1000) as currently estimated for India, but as 2:1 (i.e, 6.0
vs 3.0 per thousand on an average). The use of the revised rates (given in bracket in Table 19),
in preference to the oft-quoted NSS rates hitherto followed in respect of radiologically positive
cases in the community, has over-riding implications for the planning process of NTP as well as in
respect of resources-management under it.
2.2.4.
Current estimates on pulmonary tubercusis:
2.2.4.1. Estimates for Andhra Pradesh, South India: In a recent report, the prevalence rates of
bacteriologically as well as total problem of positive tuberculosis In Andhra Pradesh is estimated,
based on two studies of comparatively recent vintage, rural area in Medak (1992) and tribal area
in Khammam (1982)60. The smear positive prevalence among those aged 15 years and over in the
former was 1.62 per thousand. It was 5.13 In the latter. The coverage of sputum exam of eligibles
in Medhak study was only 70.8% and radiolograpic coverage, a mere 55%. To what extent the
rates obtained from low coverage of population could be taken to be representative is
questionable15. In most studies presented In the current review, the coverages were of the order
of 90% or more7’10’12-1415.
In the Andhra Pradesh estimate, presented by Ramanna, the above rates were adjusted (Total
prevalence in the state 890 per 1000 In 15+ age), using the correction factors for varying methods
of screening and sputum tests, based on unpublished data from North Arcot study (TRC,
unpublished)61. This was to make the estimates comparable to NSS, NTI.LS and BCG Trial data7’10’15.
However, whether these correction factors were seeking to correct rates, falling within 95%
confidence limit, as was done In an earlier stud/2,63, could not be ascertained, owing to
unpublished and brief nature of the material under reference61. Correction factors for prevalence
rates based on statistically discriminated rates were provided In a later NTI study, revising the
earlier correction factors24,62. The adjusted rates provided In Ramanna's report are recalculated,
adjusting for age-difference, screening method and bacteriological specimens examined.
Accordingly, the prevalence rates for culture positive cases are found to be 300.0 per 100,000 for
5+ age group in Andhra Pradesh (270.0 to 330.0 per 100,000). These are similar to rates found in
NSS7. The prevalence of smear positive cases could be 130, 370 and 150 for Medak. Khammam
and Andhra Pradesh respectively, for the population aged 5+.
2.2.4.2. Estimate on patients on treatment: Based on data from secondary sources, as well as
qualitative study in 26 villages spread over 13 selected states in India, the National Council of
Applied Economic Research New Delhi has estimated the point prevalence of tuberculosis (all
forms) to be 4.23 per thousand, with an estimated total number of 3.8 million tuberculosis patients
on treatment in the country64. Since information on method, population, coverage and diagnostic
criteria are not readily available In the report, (as already discussed in Section IIIC, these have vital
18
bearing on the results), the above should be viewed more as an estimate of the load on the
health services than epidemiological estimate of the problem In the population at large.
2.2.4.3. Estimate on morbidity and mortality from SEARO workshop 1996: In the workshop on
country-specific estimates of tuberculosis morbidity and mortality, organised by the WHO SEARO
at New Delhi In November S’d, all available country-specific data were used to estimate the
problem, by using four models, namely Notification Method. ARI Method, Incidence Study Method
and Triangulation Method (DISMOD). The current estimate on mortality for India, as arrived at. is
shown in bracket in Table 19. The disease rates are being further developed and not presently
included herein.
2.2.4.4. Use of estimates: The average rdtes. as estimated and presented above, could be of great
use in the planning for provision and utilisation of resources, as well as for monitoring of
programme output In terms of the problem In the population at large. It should be realised,
however, that these estimates, owing to the nature of their computation and large range, could
be of limited purpose in effect-evaluation of Intervention. This Is especially so, since in tuberculosis
control, one is called upon to discriminate between low Initial prevalence rates with small amount
of change consequent on intervention (say between 5-7%) (see para 3.1, intra).
3.
CHANGES IN TUBERCULOSIS WITH TIME: ITS MEASUREMENT:
Indices for measurement: Starting from the NSS In 1955-58, several surveys were conducted
3.1
in different areas In India at different times (Table 1). It was observed that the comparatively low
prevalence rate of cases between 2-8 per thousand, (say, 4 per thousand) had more or less,
remained unchanged over the years. The reason for this Is clear from an understanding of the
natural dynamics of tuberculosis (Figure 14). It could be observed that the pool of cases remained
unchanged over short period of time, without active intervention. On account of the relatively low
initial case prevalence rates (Table 9 and 10) and the expectation of a very small change in it,
if at all. the population sample size had to be considerably large to identify a statistically valid
change (Appendix Table ii )46’48. Even mortality rates In tuberculosis did not present itself to be a
sensitive index of actual epidemiological time trend. However, on interpreting the above survey
results on infection and disease by areas, especially their distribution by age and sex as well as
over time. Indian epidemiologists had taken the viewpoint that the disease could be in an
endemic phase in India, on a slow declining trend,taking their cue from Grigg's work5459 (see Fig. 15
for the hypothetical secular epidemic curve).
It is to be noted In this connection, that the wide gap between mortality rate, case rate and
infection rate in the community (90:400:38,000 per 100,000 respectively) is also considered
significant with respect to the Interpretation of the Interpretation of the age of the epidemic in
India (Fig. 15). While dwelling on the cunent epidemiological situation In India, one may not miss
the rather disturbing recent findings on drug resistance as shown in Table 15, even though
information on this is still meagre.
3.2
A.R.I.: It needs to be recognised that bacteriological case-prevalence, in being the pool
of cases leftover and carried over time. Is more of an index of a failure of the anti-tuberculosis
effort, than a representation of the secular epidemiological trend as such. Incidence of cases, on
the other hand, represent the current risk of developing disease among the previously infected,
and therefore give a cumulated risk over a long time, as it has to be observed in higher ages. As
distinct from these, risk of infection is the index, which represents the direct and a near-immediate
consequence of the presence of bacteriological cases in the community. Following extensive work.
'on the epidemiology of tuberculosis in the western countrieTin the'comparatively recent times,
ARI has been recognised to reflect the current epidemiological situation in an area, in preference
to the disease rates54. It is also possible to derive from it an estimate of the current rate of
incidence of the infections cases, as suggested by Styblo^, also shown in Table 4. In the Bangalore
19
rural area studied by NTI in a 23 year period. ARI was seen to have declined at 2.3% per year
(Figure 3). The smear positive case incidence was 23 per 100.000 at the last survey, as estimated
tom the prevalence rate of cases in 1984. It could thus be interpreted as coming down during
the period at a rate nearly corresponding to the foil In ARI. However, there was no other evidence
of ARI declining in India, apart from Kashmir valley (Fig 16). Thus. India could be Identified as a
country of high transmission and inadequate decline, a situation which it shares with the
sub-Saharan countries (ARI 1 -2.5%, annual decline 0-3%).
4.
*
CONTROL PROGRAMME AND EFFECT-EVALUATION:
i
5 |GC,
. a tub®rcljlosis Programme in the Indian context: One may wonder whether
the epidemiologists are not over-zealous In expecting a control programme, like it is in India wrth
a current efficiency at 33% or less, to bring in the epidemiological returns in a relatively short time
For countries like India, with the likely current average annual number of smear positive incidence
o
iStrlCt Of L5 million-population, the task of reducing it ®
°
,a d|rter®ot proposition, as compared to. say. a country like the Netherlands, with an
incidence level of 12-15 smear positive cases In a million population (Table 20)47 65.
The
Indian
be viewed in the light
eonilnu.
to situation
con^toshould
« r—
I?.h' of
»' the
,h8 fact
'?« that
“ already diagnosed cases
major proportion of the prevalence (2/3rd) year after year (Fig.l4) l.e. it
could be a sort "of an epidemic
left-.
apidemic of the left-overs".
This needs to be transformed through the
operation of a highly efficient treatment p't
programme, as experts globally contend, with a
well-thought out case-finding network In place.
Some salient features of the epidemiological situation, as used for planning a relevant control
AfS’endkTaSe vi®
tO9ether With their lnterpretation with regard to the trend. Is given in
su™e'ra2Ca ttUOU3h ARI and e«ec*-«va|Uatlon: In view of the large population
Appendlx Table 11)for
information on disease from the repeXXs to
bJ
dS r®vfewln9 of th® epidemiological situation from time to time or effectina
f
a.re2 tO arSa followin9 a tuberculosis programme, the alternative of carrvina
cut Infection surveys instead, could b® considered. For this purpose, sub-district level samplX ol
unvaccinated children could be selected (cluster samples). Care should be taken to obtain
stra™icafl°n- in
°f differences in rates as seen even between contiguous
areas. Study by Bleiker and unpublished information developed in association with the WHO
sTudv°n thPVa"ab'T
the reviewer- Sh0WS that the relevant s^Pl® sizes requked ^r
s udy in the population, could. Infact, be a manageable activity64. The question of an
nSm Vr-MK °f unvapcinated chlldren available In The communitj was naHound to be a
Pu ?Ik ’ ev®nif ,he proportion vaccinated was found to be as high as 80-90%d7. It was also shown
ahffec?ThT~fmatO«VMiable proportior's of vaccinated children from infection surveys would not
ou the fitld 3® >’
°/ 0,7 9enefal hePtth WOrl<erS wrth simPle training could carry
PPnrtoTof d i k
■ The finding f10™ the N11 as well as BCG.Trlal that the prevalence of various
oHhf nPa ?
enrT KV malnutritlon among younger children had not influenced the estimates
de^o^^coi^rteV 4^-^011 OUS
°f SpeCial sigp^ance. in this context, to the
n
P!P.
°,n record here that a study on ARI as related to some well-known health incides in the
PPmZfataH of'tn''
rn°rtality. infant mortality and tuberculosis case fatality rates etc.) was
ch° P ® ed at
NT'Bangalore in 1993-94. (Nil, peri-urban follow up). When analysed, it could
general health as well, besides for
tuberculosis
I vj
I
VJ I
w lw«
20
4.3
Epidometric Models: Over end above the direct measurement of the situation through ARi.
it Is also suggested that repeat surveys in some areas be carried out In order to provide inputs for
construction of epidemiological models, feeding into them the data on operational efficiency of
the programme also as a variable, as done by the present reviewer's group48. The trend obtained
from such a model (Fig. 17) had shown a marginal decline, relevant to the current programme
efficiency over a period of 50 years or more, the decline almost aetting arrested thereafter
corroborating the observation of epidemiologists on a slow decline*969. It is suggested that the
exercise of mathematical approximation could be perfected from the data obtained from
longitudinal surveys in selected areas, of course with adequate provision for reflecting socio
economic changes with time.
I
21
REFERENCES
01.
WorlcrHeatth Organisation: WHO Expert committee on tuberculosis. Ninth report. WHO tech
Rep Ser, 1974.552,1-40.
02.
Sarin R & Dey LBS: Indian National Tuberculosis Programme ; Revised strategy Ind J
Tuber.. 1995.42,95.
03.
Ziedberg LD, Gass RS. Dillon Ann, Hutcheson RH: The Williamson county tuberculosis study
- a twenty four year epidemiological study - Amer Rev Respiratory Disease, 1963,81 (No.3
Part 2)1-88.
04.
Pio Antonio: Impact of present control methods on the problem of tuberculosis. Reviews
of Infectious Dis. 1989,11,suppl 2. S-36O- 365.
05.
Frimodt-Moller J: In Proc, of the TB workers conference held in Bombay, 1949, TB Association
of India New Delhi, p.40.
06.
Frimodt-Moller J: A community wide tuberculosis study In a south Indian rural population,
1950-55. Bull World Health Organisation. 1960,22.61-170.
07.
Indian Council of Medical Research: Tuberculosis in India - A sample survey 1955-58,
Special report series No.34, ICMR, New Delhi 1-121.
08.
Raj Narain, Geser A, Jambunathan MV, Subramanian M: Tuberculosis prevalence survey
in Tumkur district. Ind J Tuber., 1963,10,85-116.
09.
Gothi GD. Chakraborty AK. Nair SS. Ganapathy KT. Banerjee GC: Prevalence of
tuberculosis in a south Indian district twelve years after initial survey, Ind J
Tuber..l979.26,121-135.
National Tuberculosis Institute. Bangalore: Tuberculosis In a rural population of south India:
a five year epidemiological study. Bull Wld Hlth Org.. 1974,51.473-488.
11.
Olakowski T. Assignment report on a tuberculosis longitudinal survey, NT1 Bangalore WHO
Project India 0103. 1973, SEA/TB/129 WHO SEARO Indraprastha Estate. New Delhi.
PP.83+hv+35 pages tables (unpublished).
12.
Chakraborty AK. Singh H. Srlkantan K. Rangaswamy KR. Krishnamurthy MS. and Stephen JA:
Tuberculosis in a rural population of south India: Report on five surveys. Ind J Tuber.
198Z29.153-167.
13.
Chakraborty AK. Choudharl K. Sreenivas TR. Krlshnamurthy MS. Shashidhara AN,
Channabasavaiah R: Tuberculosis infection in a rural population of south India : 23 year
trend. Tubercle and Lung Dis. 1992,73,213-218.
14.
Chakraborty AK. Suryanarayana HV. Krlshnamurthy W, Krishnamurthy MS. Shashidhara AN:
Prevalence of tuberculosis in a rural area by an alternative survey method without prior
radiographic screening of the population. Tubercle and Lung Dis. 1995,76,20-24.
22
15.
Tuberculosis Prevention Trial Madras: Trial of BCG vaccines in south India for tuberculosis
prevention, Indian J Med Res, 1980, 72(suppl), 1-74.
16.
Mayurnath S. Vallishayee RS. Radhamanl MP, Prabhakar R:
Prevalence study of
tuberculosis infection over fifteen years. In a rural population In Chinglepet district (south
India). Indian J Med Res.. 1991.A 93,74-80.
17.
Tripathi SP: Fifteen years follow up of Indian BCG - Prevention trial. In Proc.XXVI 1UAT
WORLD CONFERENCE AT SINGAPORE 4-7 Nov 1986; Professional Post Graduate Services KK
Japan Tokyo. 102.pp.69.
18.
Tuberculosis Prevention Trial Madras: Full report of 15 years follow up (unpublished Draft).
19.
Choudhari K. Krishnamurthy MS, Shashidhara AN, Channabasavaiah R, Sreenivas TR.
Chakraborty AK: Tuberculin testing in the community through general health services in
preparation for tuberculosis surveillance - a study of feasibility. Ind J Tuber. 1991.38,131 -137.
20.
Pamra SP, Goyal SS. Mathur GP : In Proc. 23 TB & Chest Dis. workers conference, Bombay.
Tuberculosis Association of India. New Delhi 1968. p.61.
21.
Goyal SS. Mathur GP, Pamra SP : Tuberculosis trends in an urban community. Ind J
Tuber.. 1978,25.77-82.
Cauthen GM, Pio A,
WHO/TB/88.154.,pp.l-34.
ten
Dam
HG..
Annual
risk
of
tuberculous
infection, '
23.
Rajnarain, Krishnamurthy MS, Anantharaman DS: Prevalence of non-specific sensitivity in
some parts of India, Indian J Med Res., 1975,63(8), 1098-1109.
24.
Chakraborty AK, Channabasavaiah R. Krishnamurthy MS, Shashidhara AN. Krishnamurthy
W and Choudhary K: Prevalence of pulmonary tuberculosis In a perl-urban community of
Bangalore under various methods of population screening. Ind J Tuber., 1994,41, 17-27.
25.
Nayar S, Narang P, Tyagi NK et al: Field trial of short term intermittent chemotherapy
against tuberculosis, an ICMR project. Department of community Medicine & Department
of Microbiology MG Institute of Medical Sciences Sevagram Wardha. 1989. Report to
ICMR, pp. 1-325 (unpublished).
26.
Tuberculosis Research Centre Madras: Tuberculosis prevalence survey in Raichur district
Karnataka (unpublished).
27.
Tuberculosis Research Centre Madras: Tuberculosis survey in a tribal population in North
Arcot district(unpubllshed).
28.
Directorate of Health Services A & N Islands: Intensified tuberculosis control programme in
the isolated tribal population of Car-Nicobar island WHO - Government of India HSR
project No. IND/HSR 001/D., A&N Administration Port Blair 744101, 1989. pp.1-64 (+xviii),
(unpublished) restricted supply mimeograph.
29.
Chakraborty AK, Channabasavaiah R. Krishnamurthy MS, Shashidhara AN. Motiram G:
Tuberculin skin sensitivity following chemoprophylaxis in an island community, Ind J
Tuber., 1991,38, 201-211.
23
30.
Chakma T. Rao PV, Pall S. Kaushal LS. Datta Manjula, Tlwarl RS: Survey of pulmonary
tuberculosis in a primitive tribe of Madhya Pradesh. Ind J Tuber.. 1996,43,85-89.
31.
Siddiqui Diba. Ghose S, Krishnamurthy MS. Sashidhara AN: Tuberculous infection rate in a
rural population of Bikaner district, Ind J Tuber., 1996.43.91-97.
32.
Chakraborty AK Ganapathy KT. Nair SS, Kul Bhushan: Prevalence of non-specific sesrtlvrty
to tuberculin in a south Indian population. Indian J Med Res.. 1976,64/57,639-651.
33.
Chakraborty AK, Chaudhurl K: Tuberculin testing In a partly vaccinated population. Ind
J Tuber.. 1992,39,149-158.
34.
Styblo K. Meijer J, Sutherland I: The transmission of tubercle bacilli its trend in a human
population. TSRU report No.l. Bull Int Union Tub. 1969.XUL 1-104.
35.
Rajnarain. Nair SS. Chandrasekhar P. Ramanatha Rao G: Problems connected with
estimating the incidence of tuberculosis infection. Ind J Tuber.. 1965.13.5-23.
36.
Chakraborty AK: The use of screening tools for the estimation of tuberculosis case rates in
a community. Indian J Pub Hlth.. 1980. 24.115-120.
37.
Nair SS. Gothl GD. Naganathan N. Rao KP, Banerjee GC. Rajalakshml R.. Precision of
estimates of prevalence of bacterlologically confirmed pulmonary tuberculosis in general
population. Ind J Tuber., 1976.23,152-159.
38.
Rajnarain. Nair SS, Naganna K. Chandrasekhar P, Ramanatha Rao G, Pyarelal. Problems
in defining a 'case' of pulmonary tuberculosis in prevalence surveys. Bull Wld Hlth
Org., 1968.39. 701-729.
39.
Chandrasekhar P. Nair SS. Rao P. Rao KR, Pyarelal: Examination of multiple sputum
specimens In a tuberculosis survey: Tubercle, 1970.51,255.
40.
Gothi GD. Chakraborty AK. Banerjee GC: Interpretation of photofluorograms of active
pulmonary tuberculosis patients found in epidemiological survey and their five year fate.
Ind J tuber., 1974,21.90-97.
41.
Tuberculosis Research Centre: Report on continued follow up of Tiruvallur population,
unpublished. In: TRC Annual report. 1992. TRC Madras India.
42.
Gothl GC: Epidemiology of tuberculosis In India. Ind J Tuber., 1982.29,134-148.
43.
Gothi GD, Chakraborty AK Krishnamurthy W. Banerjee GC: Prevalence and incidence of
sputum negative active pulmonary tuberculosis and fate of pulmonary radiological
abnormalities found in a rural population. Ind J Tuber.. 1978.25.122-131.
44.
Gothl GD. Chakraborty AK. Parthasarathy K Krishnamurthy W: Incidence of pulmonary
tuberculosis and change in bacteriological status of cases at shorter Intervals. Indian J Med
Res. 1978,68.564-574.
45.
Ray D. Abel R: Incidence pf smear positive pulmonary tuberculosis from 1981-83 in a rural
area under an active health care programme In south India. Tubercle & lung Dis.,
1995,76.190195.
24
46.
Chakraborty AK: Some problems of epidemiological assessment of a district tuberculosis
programme. NTI News letter, 1973.10,57-61.
47.
Chakraborty AK: Tuberculosis situation in India: measuring it through time. Ind J Tuber..
1993.40.215-225.
48.
Balasangameshwara VH, Chakraborty AK. Chaudhuri K: A mathematical construct of
epidemiological time trend in tuberculosis - fifty year study, Ind J Tuber. 1992.39,87-98.
49.
Styblo K: The relationship between the risk of tuberculous Infection and the risk of
developing infectious tuberculosis Bull Int Union Tub, 1985,60.117.
50.
Gothi GO. Chakraborty AK. Jayalakshml MJ: Incidence of sputum positive tuberculosis in
different epidemiological groups during five year follow up of a rural population in south
India.. Ind J Tuber., 1978.25.83-91.
51.
Grzybowski S: Natural history of tuberculosis epidemiology. Bull Int Union Tuber Lung Dis,
1991,66,193.
52.
Chakraborty AK. Gothi GC, Dwarkanath S, Singh H: Tuberculosis mortality rate in a south
Indian rural population. Ind J Tuber.. 1978,25,181-186.
53.
Dholakla RH, Estimates of potential economic benefits of DOTS In India. WHO Project, 1996
(unpublished).
54.
Gothi GD, Nair SS. Pyarelal: Some epidemiological aspects of tuberculosis disease and
infection in paediatric age group in a rural community, Indian Paediatrics, 1971.8.186-194.
55.
Gothi GD. Isaac B, Chakraborty AK, Rajalakshml R, Singh S: Tuberculosis in a slum
community. Ind J Tuber., 1977,24.68-74.
56.
Motiram G. Chakraborty AK: Prevalence of childhood tuberculosis and problems in its
diagnosis in a community. NTI News letter, 1990.26,8-17.
57.
Rajnarain, Diwakara AM: Mortality due to tuberculosis in children in India. Indian
Paediatrics, 1975.12,529-537.
58.
Grigg ERN: The arcana of tuberculosis with a brief epidemiologic history of the disease in
USA: Amer Rev Tub. 1958.78 Part I & 11,151, Part III 426.
59.
Nagpaul DR, Tuberculosis In India - a perspective. J.lndian Med Assoc. 1978,71.44.
60.
Ramanna GNV: Preliminary results of disease burden. 1995 (unpublished).
61.
TRC: Tuberculosis prevalence survey in North Arcot district. Annual Report of TRC, 1990.107118 (unpublished) quoted by Ramanna GNV, as above.
62.
Gothi GD. Radha Narayan. Nair SS, Chakraborty AK. Srikantaramu N: Estimation of
prevalence of bacillary tuberculosis on the basis of chest x-ray and/or symptomatic
screening, Indian J Med Res, 1976,64,1150-1159.
63.
Chakraborty AK: The use of screening tools for the estimation of tuberculosis case rates ir
a community, Indian Jour Pub Htth, 1980, 24,116-120.
25
64.
National Council of Applied Economic Research New Delhi: Human development Profile
in India: Inter-state and inter-group differentials. 1996 (unpublished).
65.
Styblo K: The estimate of tuberculosis In the Netherlands: Bull Int Union Tuber, Lung Dis,
1990,65(23)49.
66.
Bleiker MA, Sutherland I, Styblo K, ten Dam HG, Misljenovic O: Guidelines for estimating the
risks of infection in a representative sample of children. Bull Int Union Tuber Lung Dis 1989,
64.7-12.
67.
Salekar A. Edo Narain, Porobo Jalram, Dessai H P, Aranjo: BCG scar prevalence among
rural children in Goa. NTI Bulletin. 1993, 29(1 &2), 17-20.
68.
Chakraborty AK, Ganapathy KT, Gothl GD: Prevalence of Infection among unvaccinated
children for tuberculosis surveillance. Indian J Med Res. 1980,72.7-12.
69.
Grzybowskl S: In reply. Drugs are not enough. Tubercle & Lung Dis. 1994,75.317-318.
26
(
11ST OF FIGURES
F.l
Distribution of Indurations to tuberculin test
01
F.2
Proportions of persons infected with M.tuberculosis and
other mycobacteria In a south Indian rural area
01
F.3
Estimated ARI and observed annual Incidence of infection
over time
02
F.4
Tuberculous infection in Tumkur district by age and sex
02
F.5
Active and probably active tuberculosis as well as bacillary
cases in Delhi zone of NSS
03
F.6
Distribution of prevalence by population proportion in various
ages
03
F.7
Prevalence of cases of age, sex and method of diagnosis
04
F.8
Age-wise proportional distribution of prevalence cases In an
area with time
--
05
F.9
Drug resistance (INH) in tuberculosis cases in the community
05
F.10
Incidence of bacterlologically positive cases in a community
over a period of 5 years of observation.
06
Average incidence of bacterlologically positive cases by
tuberculin test induration size.
06
Epidemiological groups in population and incidence of
tuberculosis.
07
F.13
Prognosis of bacterlologically positive cases detected in a
survey over time.
07
F.14
Natural dynamics of tuberculosis In a community over a short
span of time.
08
F.15
Development of the wave of tuberculosis epidemic through
time.
08
F.16
ARI across India
09
F.17
Mathematical model showing hypothetical trend In
tuberculosis for a 50 years period.
09
F.ll
F.12
•• m
IN A rOI’ULA I ION AGI-J) 0-9 \ RS.
AGE 0J4
X
96.'i
20
N » 3950
A
15 .
2.0
I
10
s
5
1-1 -r-rf-r-r-y P 7'/' J it ? t
3
c
0
•3
X
20
e
cj
__ _______AGF 5-9
85.2
N - 3890
15
8.3
105
0^
0
5
10
15
20
25
30
35+
Indurations in mni.'A' - Antimode of distributions : Arrow to its right infected proportion
SOURCE : OLAKOWSKl"
Fig. 2. PROPORTIONS OF PERSONS INFECTED WITH M. TUBERCULOSIS*
AND THOSE SHOWING NON - SPECIFIC SENSI TIVTY** IN A RURAL POPULATION
’OO
4
[TTTI
•••
to
SO
O
TO
co
k:
co
u
40
30
:o
II
’O
J •
0
------ 0-4
•----- 9 - 9 --------•----- 0-34
A O C
’ tv McoArivc
PEAf TOHJ
( O• • mm I
’ TV ^OSiTfvc
’UC TOPS
{ ’O mm or mor* |
• ;
70 TV negative
* * *1 Pr AC ron$
<0'7 mm.
I 8 mm or m or t )
* Infected - Persons who me I TU positive reactors
•• Non - Specfic sensitvity - Persons 20 TU positive reactors
among those with negative reactions to I TU.
SOURCE : CHAKRADORTY et al U
‘ Fie 3 ANNUAL ESTIMATES OF RISK OF INFECTION (1962 - 1985) AND OBSERVED ANNUA
INCIDENCE OF INFECTION (1962 - 1967)
I
5
______ _ ESTIMATED ANNUAL
RISK RATE (X)
UJ
13
I
O
-------- OBSERVED ANNUAL
INCIDENCE OF INFECTION (X).
z
UJ
u
cr
UJ
O- 5
a.
O- I OO1 -Up
i i
I ' r-
P
3
3 a»-.
5 <■
5
0*
2»
| 55 "
I
X o i
i
S
2
z
□
TIME OF SURVEY
SOURCE : CHAKRABORTY, AK et al
13
<
Fig. 4. PERCENTAGE OF REACTORS BY AGE AND SEX
(TUMKUR DIST. 1960)
•0
---rcMAif
/O'
5J
U
u
1,0
o
io n x' » «o m
Af.f
in
n
f(a«t
Reactor : 10mm and over
SOURCE : RAJNARA1N el al*
•'OC UN ULLU1 ZONE BY AGE & SEX
A ACTIVE & probably ACTIVE CASES ON MMR
DELHf
C I TY
VILLAGES
14
/
s9.
-h
s---AGE
* C, E
*• G t
B^BACTERIOLOGICALLY positive cases.
DELHI
CITY
towns
villages
S’
I
0
£
J. G C
•MALES
FE.MALS
SOURCE : NSS (7)
FIG 6
distribution of age wise prevalence
OF BACILLARY CASES IN THE COUMMUNITY
POPULATION
AGE GROUP
ACE CROUP
40 ♦
5 CH
20 ♦
93
0-39
50’:
TB CASES IN THE POPULATION
Source: Author (Unpublished)
50T
Fig. 7. PREVALENCE OF DISEASE BY AGE SEX
AND METHOD OF DIAGNOSIS
2
OO-
Joo of a o <2-
jj o o —
^00-'
■» O 0 —
o
o o3 00 0-1
Q
7^0 O —
O
/3OO—
3 0 o—
^OO-
O
«*c
* OO -
S’ 3 o o-
<
<9 0 0 o-
-* • o o-
3^0 o—
/ 3 OO~
o
O’
T T
irt
O
>■
CM
7 7
O
o
'*■
o
o»
O»
77TITT7T-+cnouv {'tiaps')
Categories of eases :
a) Culture Pos. on 2 specimens
b) Culture Pos. on one specimen only
c) Culture Neg. smear Pos. ( 3 or more AFB)
d) Abacillary MMR - active (by 2 readers)
SOURCE : BCG. TRIAL1'
Fig. 12. INCIDENCE OF SPUTUM POSITIVE TL'BERCULOSIS IN DIFFERENT
EPIDEMIOLOGICAL GROUPS
EPIDEMIOLOGICAL
ANO
GROUPS
IN c I O E N C E
ANNUAL
annual
S'2C
INCIDENCE
RATES U/..)
OP EPIDEMIOLOGICAL
CROUPS (•/■•>
OF CASES IH
PROPORTIONAL
CONTRIBUTION
TQ
incidence from
1’73
EACH
GROUP
30-
J 6-33
f.
THEM
TUB.
% POS.
21 • 5
76-3
63-i
-I
=J 4 6
I -03
24.4
3-7
2-2
\ 5?
0-5
POPMILATIQIJ
>r o
Sodiory
TUOERCULl^
NEGATIVE
N :
X-ray normals.
fEnT'.Tn a a f
AB :
CD :
TUB.
NEG.
17. 0
|cd
n cziA9
T USERCULIN
POSITIVE
23-7
i
Cases
nun
inactive tuberculous lesion and non.tuberculous shadows.
Probably active tuberculous shadows.
SOURCE : Gothietal50
Fig. 13. PROGNOSIS OF TUBERCULOSIS CASES DURING 5 YEARS
OF OBSERVATION. ALL AGES AND BOTH SEXES.
2-4
3^
*
£3
■ *
5 6
UUIIIIIII
27-0
13-5
itWr
I26«
1007.
—"
9 5
t:
34-9
: S-3
13-5
I
32-5%
I
12-7
_L_
15-1
10 2 7.
421
31
30-1
1-5
0
15
40 •5
1-5
i
2
3
Non-injected
Rad io I oq ica 11 y liurmal
ID.Sujduws
In/ected
Radi oloq ical !y .lunnjl
Udder io logical ly
Pus i ti ve
AO S"udUw$
UtdUis
49-3 %
Hi
’I
5
SOURCE : 0LAK0WSK1 T. UNPUBLISHED11
<
POOL OF TUBERCULOSIS CASES 4N THE COMMUNITY
(NATURAL DYNAMICS)
POPULATION
ADDITIONS TO
root or cases
AT 013". OF POPN.
DEAD(d)
annually
ELIMINATED
/ CURE(c)
!NCIDENCE(i)
38 Z
20%
1^3; 7^'
62%
62%
PREVALENCE CASE CONTINUING
62% (0.62)
CASE PREVALENCE
INITIALLY
ONE YEAR
-0.20
pto 1.0
CASE PREVALENCE
NEXT YEAR
pit i o
-0.18 *0.38
(-8—c*i)
Source Author (Unpublished)
Fig. 15. DEVELOPMENT OF THE WAVE OF TUBERCULOSIS EPIDEMIC
THROUGH TIME
TU BERCUIOSIS WAVES
o
°
5X
tel
:
w
S
INFECTION
DISEASE:
1 000 -
. ■'
100 -
UR BAN
• 00 700 • 00 SOO
AOO
3 00
200
RURAL
MORTALITY
-
-1000
■ 900
1 00 -
u
Mb
i
SO
8
> 8
5*
i"
i
i
i
i 1
100 150 200 290 900
YEARS
(SEMI DIAORAMATie - ADAPTED- FROM
w
<
X
E.RN. GRIGG)
SOURCE : CHAKRABORTY AK
47
The tuberculosis epidemic curve appears similar in form to those of other infectious
diseases. However, the former develops through centuries. It has an ascending limb (phase
of spread), a peak (phase of transition) and a descending limb (phase of decline), followed
by endemicity. The essential proximity of infection, disease and mortality curves
characterises the phase of spread (shown with arrow 4a). Wide gaps between one and the
other rate develop at the peak and descending limb (shown with arrow b). In India, gaps
similar to the latter, exist now. An inference that could be derived is obviously of an
advanced epidemic curve, probably in declining phase. The urban - rural epidemic curves
are different entities - but could cross at some point in the descending phase. The urban rural distribution presently observed in India, may be viewed in this light.
S.; /.
I IC 16. ANNUAL RISK ()|- IM I.C I ION (ACROSS INDIA)
x
\
C
(
5
r
j
K.
VALLCY
1 .4
'T
Jb
rt
r
ARI:
J EHLINE
SLAHC
n/
I KULU 1.8
j
/
I
HARYANA 1.5
a
RAJASTIIAM
AGRA
1.8
(0 4) - (I 4)
)
J
z-4
tC
)
<s
)
-1
, l
BANGALORE 1
d-
1
>!22T
I 6G |
b 0 99 f
tn j
.*•c 1 12
0 !H
■
•i • b
\DANGALOWE
a*
MADRAS
1.79 (STATIC)
i ►
(x) 1.4
(y) 2.23 i ►
\
0
TIME Of SURVEY
1972 &
RAJASTHAN
<
1991
(Separate) j
ax co
bl 1^4-79 *•-
RANGALOHE (a) 1966-73
(C) 1962-85
1969-1984
OTHERS
MADRAS
Sea'* lJ7.uCO.CXJO lcm t 170 »mi
lambert Conical O»mo^oip4>«c Projection
1972
SOURCE : CAUTHEN (WHO)
HQ
Fig. 17.MODEL DEPICTING HYPOTHETICAL TIME-TREND
OF TUBERCULOSIS IN BANGALORE RURAL AREA.
0 40 -i
0-31 zt
UJ
0
Z
■*
0 It -
a
0 H0 IS
o
co
5b
20
IS
JO
TEARS
MO DISTRICT TUQERCUIOS'S PROGRAMM €( 0 TP )
NO OTP < 0-5 r°S ) C* 1J > SR 100 *4 (I- SO tRS )
I «-S0 »RS )
HQ OT p ( 0-5 » RS ) C f 40 *4 SR ’00
5R-$t*NCAR0 RgG'MEN
ZF -CASE Finding
10
47
SOURCIt : CHAKRABOR I Y AK
(
)
Particulars
SI.
No.
*
LIST OF TABLES
Page No.
01.
Prominent epidemiological studies
01
02.
Prevalence of infection (Reagent RT-23)
04
03.
Prevalence of infection (Reagent PPD-S)
04
04.
Annual incidence of Infection and smear positive case
incidence
04
05.
ARI (NU.LS and BCG.Trlal)
05
06.
ARI across the country
05
/ 7
.2'
07.
Area-wlse prevalence of pulmonary tuberculosis by sex
06
08.
Area-wise prevalence of radlologlcally positive
bacteriologically negative tuberculosis
07
Prevalence of bacteriologically positive cases
(M.M.R.-Screening)
08
Prevalence of bacteriologically positive cases
(Symptom-Screening)
08
11.
Prevalence of bacillary cases by age and sex
09
12.
Prevalence by selected socio-economic attributes
10
13.
Proportion smear positive in prevalence
11
14.
Proportion drug-resistance In different studies
11
15.
Incidence of tuberculosis cases
12
16.
Estimated annual tuberculosis mortality rates
12
17.
Tuberculosis prevalence and Incidence in paediatric
population
13
18.
Estimates of workers with tuberculosis of lungs in India
13
19.
Problem of tuberculosis in the country* (on average)
14
20. .
Indian situation set against advanced countries.
14
09.
10.
Table 1: PROMINENT.EI’IDEMIOLOGICAL STUDIES IN INDIA
REI’ORTED/UNREPORTED
Rcfere
ncc
Year
Aie*
Design
Examined popn A age
gtoup
Method A invest! - gallons
Outcome
Short Thle
• Madanapalle
Town study
5
1948 -49
Madanapalle town
Aibilrarily
selected
Popn size
14,000
All ages-Tubercuiin test
PPD-S
ITu-10 Tu-100 Tu MMR
Sputum Culture
i) lnfection rare
ii) X-ray active case
rate
iii) Bacillary case rate
6
1950 -55
Madanapalle town
A rural are*
Sample Survey
60,000
5* in age
a) House to house census
b) All ages-tuberculin test, PPD-S, ITu-lO Tu-100
Tu; c)5* age MMR Sputum Culture
- Do -
• Madanapalle
• National Sample
Survey(N.S.S.)
7
1955 -58
6 Zones:
Hyderabad
Madanapalle,
Patna, Trivan
drum, Delhi,
Calcutta, Towns,
villages A I city
in each zone
Sample Survey
representing 40%
. of Indian popn
Cities-131,319
Town-59,541
Villages-137^271
5* in age
a) House io house census
b) MMR 5* in age
FROM MMR-aboomuls:
c) Direct Smear of Sputum -2 Specimens
d) Two laryngeal swab culture
i) X-ray active case
ii) Culture positve
case
•Tumkur Study
8
I960 -61
Turakur district
Sample Survey
About 30,000
IO* age
a) House to house census;
b) MMR 10* age
c) Tubcrculin testing all ITu RT 23
i) lnfeclion
ii) Xray active case
iii) Culture* case
9
1972 -73 .
- Do -
- Do •
- Do -
- Do -
- Do -
•Tumkur
Resurvey
• Long’Survey
(LS.NTT) Su I
Survey 2
Survey 3
Survey 4
Survey 5
10.11
12
1961 -68
1961 -63
1962 -64
1964 -66
1966 -68
1977
3 sub-districts
(taluks) of
Bangalore dist
Sample Survey
Popn 66,000
Age 5*
a) House to house census
b) TUberculin test to all
ITU RT 23
c) MMR-5* age
- Do Fate of infected,
suspects, cases A
other risk-groups
Sub-'Santple
Popn abotfl 14,500
• LS:Followup
13. 14
1984 -86
Fresh Sample in 2
of lite above 3
Taluks
Sample Survey
Popn. about 30,000 age 15*
Tuberculin 0-44 yrs ITu RT 23
Chest symptom questioning: I5» yrs 2 specimens
culture A smear exam from chest symptomatic;/
tuberculin reactors
Inlection A case
rales
swdy(M.S.)
A
y.-
Table P No. 1
Cliingleput district
Tamil Nadu
Selected lor
study of BCG
efficacy
All agcs(l») Ap-proi
360.000 in 209 contiguous
villages and 1 town
a) l louse to house census
b) AJI ages tuberculin lest PPD-S
c) MMR IO» age, d)2 specimens of sputum from
MMR abnormals - direct smear &. culture
a) lnlectnxi rate
1978
1983
- Do -
- Do -
Popn: I-9 yr. 10 yr trend:
Survey I. 8703 -IS yr
trend: Survey I, 4808
Tuberculin lest PPD-S
Annual infection rale
18
1968 -83
- Do-
- Do -
As in BCG trial with
selected group for followup
MMR-5* in follow up surveys
Incidence of cases
19
1983 -84
Anantpur dist AP
Dharmapuri dist in
TN
Arbitrary selected
population
Aged:0-9 yrs
Anantpur-4350 in 12 vils
Dharmapuri-2077 in 6 vils
Tuberculin testing by general health workers
ITu RT 23
Infection rate
• Infeaion
surveillance
feasibility study
•New Delhi study
6 surveys
20^1
1962 -75 (resu
rveys at 2Vfc yr
intervals except
Su6: after 4yrs
from preceding)
New Delhi area '
under New Delhi
TB centre
Population
covered by
Centre
Aged-5*
Approx 30,000
a) House to house census
b) MMR
c) MMR abnonnals-culture of sputum &. laryngeal
swab
d) Followup exam by xray, repealed sputum
Xray case. Bacillary
case rale
Prevalence &.
incidence of cases
•All India
Tuberculin survey
21
1972
Villages in Kash
mir, Kulu, Lohaghat Pilhoragarh,
AgraJIaryana,
Rajasthan
Selected for
altitude contrast
1-4 yrs Between 76-769
children in each
PPD-S, 5 Tu
a) Infection rate (ARI)
• Kashmir Survey
22
1978
Kashmir valley
Sample Survey
0-4 yrs popn 2448
‘hiberculin lestinf PPD-S, 3TU
Infection rate (ARI)
• Bangalore dist
study on nutrition
2 surveys
22
1974
1979
Dodballapur
subdist of
Bangalore
Sample survey:
repeated in same
villages
Aged 0-4 yrs
a) Tubeculin testing I Tu RT 23
b) Nutrition assessment,
Infection rate by
nutrition grades
• Bangalore peri
urban study
24
1986 -89
Bangalore peri
urban area
Random Sample
of villages in a 5
km radius 19 km
from city centre
Aged 15*
56,000 persons
aUlouse to house census
b) Chest symptom screening
c) MMR
d)One specimen of sputum culture
from eligibles by (b)/(c)
Symptom screened
culture positive case
rate
- Do -
- do -
a) Aged 0-14
b) Old cases of previous
survey
Tuberculin lesi I Tu RT 23
Inlection rate (ARI)
Fate of cases
• BCG Trial
(BCG.TL)
15
• BCG Trial
(Infection Trend)
16,17
• BCG Trial
6 resurveys
• Bangalore peri
urban repeal study
(under analysis)
i
1968
Followup every
IVa for 7'/a yrs
b) Xray active cases
cjBacillaiy cases
b) Non-specific
sensitivity
Table P.No 2
• Wardha study
25
1981
Wardha dist in
Maharashtra
Central India
Puqiosive sample
(whole dist)
Aged 5*
Popn 773,500
a) House to house census
b) Socu>-econunuc stratification
cjChest syiD|Xom questioning
d)2specimcns of sputum from sym[xorna(ics culture St smear
Symptom screened
culture positive case
rate
• Raichur
study(Analysis not
completed) by
TRC Madras
26
early eighties
Raichur dist in
Karnataka
Sample Survey
Aged 15*
Ap|Mox 70.000 in 56
villages and 21 town blocks
a) House io house census
b) Chest symptom questioning
c) 2 specimens of sputum horn symptomatic culture &. smear
- Do -
•Tribal study TRC
27
1980
Jawadhu hills of
N. Arcot dist in TN
Sample Survey
Aged 15 ♦
Tribal popn of 96,000 in 56
panchayats: 24 selected
a) Tuberculin test: 1-9 yr (N-6702) ITuRT 23
b) MMR 15* age (N-12745)
c) Cbest symptom questioning 15* age (N-15075)
d) From symptomalics/MMR aboorroals: 2
specimens of sputum-culture A smear
Culture positive case
rale
•Nicobar study
28,29
I9S6
Car Nicobar island
in Bay of Bengal
Entire island
Entire popn of 17^277
resident of 15 villages
a) Tubcrculin test 0-14 (popn 5907) ITu RT 23
b) Chest symptom questioning 15* age (popn 9514)
Infection rate, smear
positive case rate
•Madhya Pradesh
Tribal study by
Regional Health
office
30
1991
Mocena dist in MP
Sample Survey
Aged 15* Tuberculin test 19 yr (N-7642) Popn 23,000
in 37 villages
a) Tuberculin test ITu RT 23 • 1-9 yr age
b) Chest symptom questioning 15* age
c) 2 sputum specimens by culmre/smear lor chest
symptomatics
Infection rate, smear
positive case rale
• Rajasthan study
by URMUL in
Lunkaransar
31
1991 -93
Bikaner dist in
Rajasthan
Arbitrary selec
tion: 16 of 33 vil
in Lunkaransar
sub-dist
Popn 0-9 yr (N 2482)
Tuberculin lest 1 Tu RT 23
ARI
NOTE: Other studies referred to in the text al respective places.
Table P.No.3
NTP
NTI Bulletin 1999, 35/1-4, 39-41
Performance of the National Tuberculosis
Programme -1998
Vishweswara Sharma*
INTRODUCTION
PHI Implementation
IVTational Tuberculosis Programme (NTP) of northern
regions was monitored by Northern Regional Centre
(NRC) at DGHS, New Delhi and southern regions by
Southern Regional Centre (SRC) at National Tuberculosis
Institute (NTI), Bangalore, till 1978. Since then, NTI has
?n monitoring the programme for the whole country.
...onitoring is a continuous assessment of performance by
certain key indicators of the programme through periodic
(quarterly and annual) reports. These reports provide
information on implementation, case finding and
treatment activities and other related aspects. The basic
organisational unit of NTP is the District Tuberculosis
Programme (DTP), which consists of District Tuberculosis
Center (DTC), usually situated at the district headquarters
and Peripheral Health Institutions (PHIs), located in rural
areas. The salient features on performance of NTP in terms
of implementation, reporting, supervision, performance of
case finding and treatment activities for the year 1998 are
given in this report. Four quarterly reports on the above
mentioned activities and one annual report on treatment
completion pattern and outcome constitute the material for
this report. In all, 1234 quarterly reports, 124 Standard
Regimen (SR) annual reports and 154 Short Course
Chemotherapy (SCO annual reports were considered for
analysis.
All medical and health institutions other than DTC, in a
district are referred to as PHIs. The principal activities of
PHIs are case finding, treatment, case holding, recording
and reporting. Depending upon the facilities available,
implemented PHIs are categorised as X-ray Centres (XCs),
Microscopy Centres (MCs), and Referring Centres (RCs).
From these PHIs, 70% of the expected reports were
received in 1998. Among the reported DTPs, a total of
11202 PHIs are implemented under NTP.
Implementation of DTP
Out of 548 districts in the country, programme is
implemented through DTPs in 440 districts. Among the
440 DTPs, 254 (58%) are providing treatment with both
SCC and SR and 145(33%) are providing only SR
treatment. The remaining 41 (9%) districts are
implemented under Revised National Tuberculosis
Control Programme (RNTCP). The population coverage
under SCC, SR and RNTCP are 61%, 22%, and 17%
respectively.
*
Computor, National Tuberculosis Institute. No.8, Bellary Road,
Bangalore 560 003
4^9
I-
Reporting by DTCs
NTI monitors the programme through the periodic
(quarterly and annual) DTP reports received from DTCs.
The reporting efficiency in 1998 is 70%. Reporting
efficiency of quarterly reports decreased marginally from
71% in 1997 to 70% in 1998. This is because some of the
RNTCP districts were submitting the quarterly reports
directly to Central TB Division, DGHS, New Delhi.
Moreover, most of the RNTCP districts have discontinued
to report non- DOTs cases to NTI.
Supervision
The key personnel of the DTC viz., District TB Officer
(DTO)ZMedical Officer (MO), Lab Technician (LT) and
Treatment organiser (TO) are expected to supervise all
PHIs, at least once in each quarter. However, the
supervision visits were made only in 42% of the PHIs,
during 1998.
Case Finding Activity
Performance during the year 1998 with regard to X-ray
and sputum examinations, new cases detected, total cases
reported per lakh population and quality of case finding is
shown in tables 1,2,3 and 4. On an average, 3984 new
X-ray examinations and 10750 new sputum examinations
have been carried out by DTPs. Of these, 58% of the
39
X-ray examinations and 59% of the sputum examinations
have been contributed by PHIs (Table 1).
Among the total patients subjected for sputum
examination, 863 were found positive by smear, 2408 by
X-ray, and 211 were having extra pulmonary disease, per
DTP (Table 2). Compared to last year 1997, the case
finding has declined by 3.8% (Not given in table). This
decline may be due to:
*
Progressive coverage of the districts under RNTCP,
which do not report NTP case finding.
*
Decreased diagnosis of smear negative TB cases.
*
Non reporting by newly created DTCs.
An overview of the statewise performance in total
cases reported per lakh population (excluding under 5
years age group) during 1998 as shown in Table 3 depicts
that some states like Arunachal Pradesh, Gujarat,
Maharashtra, Pondicherry and Sikkim show better
performance in case finding activity. The contribution of
the above four states is 42% of the total case finding.
It is observed from Table 4 that the smear confirmation
rate i.e., the percentage of sputum positive cases
Type
Table 1
New Examinations done
DTCs
PHIs
Total
Table 2
DTCs
PHIs
Sputum positive
144046
390213
42919
577178
121801
351478
22068
495347
Extra Pulmonary
Total Cases
10750
Average/
Total
DTP
863
265847
2408
741691
211
64987
1072525 3482
Table 3
Total cases reported statewise per lakh
population during 1998
Range
250 and
above
Name of the State
250-100
< 100
40
Smear Rates
Obscrved(%)
Expccted(%)
Confirmation Rate
27
32 - 35
Positivity Rate
5
8- 10
confirmed among pulmonary cases is only 27% at national
level, which is far below the expectation of 32% to 35%.
This reflects the poor quality of X-ray reading and/or
laboratory services. The sputum positivity rate at PHIs i.e.,
the percentage of sputum positive cases out of total
sputum examinations made is 5% compared to the
expectation of 8 to 10%. The poor performance of PHIs in
achieving the desired sputum positivity rate calls for the
need to strengthen the laboratory services. Proper and
regular supervision by DTC staff is very much needed in
this respect.
Treatment Efficiency
Treatment outcome of the cohort of smear positive
cases diagnosed during 1996 from the annual reports
received from the DTPs are given in Table 5. The analysis
is done separately for the patients put on SR and SCC
regimen. Patients making ten or more collections are
likely to have favourable outcome in terms of
Tabic 5
Treatment Result by cohort analysis
58
59
New Cases detected
Category of Patients
Sputum Negative
3984
Quality of Case Finding
bacteriological conversion.
Average/ Contribution
ofPHl%
DTP
X-ray 513569 713475 1227044
Sputum 1342969 |1967962 3310937
Table 4
Arunachal Pradesh, Gujarat. Pondichery,
Sikkim, Maharashtra __________________
A & N Island, Andhra Pradesh. Chandigarh,
Goa, Harayana, Himachal Pradesh, Karnataka,
Kerala, Mizoram, Madhya Pradesh, Punjab,
Rajasthan, Tamil Nadu. Uttar Pradesh,
Assam, D & N Haveli, Bihar, Jammu &
Kashmir, Orissa, Tripura. Manipur, Meghalaya,
Nagaland, West Bengal
Regimen
Category
SR
see
Total Implemented Districts_______
440
No. of reports analysed___________
124
No. of patients initiated on treatment
79221
Number
45221
254
154
47029
14061
%
57
30
Treatment Completed
-The patients making 10+ collections were considered
to have completed the treatment successfully. While
treatment completion rate is 57% in 1997 for SCC
patients, it is 30% for SR patients (Table not given).
It is also observed that the proportion of smear positive
cases put on SCC compared to SR has increased during
1998. There is a trend that more number of patients have
been put under SCC than SR.
Key Personnel
Out of 440 districts, 231(51%) districts have reported
on position of the key personnel which shows that nearly
50% of all the key personnel viz., MOs, Statistical
Assistants (SAs), TOs X-ray Technicians (XTs), LTs have
been trained in TB control programme.
Limitation
Conclusion
1. The performance needs to be improved, by increasing
the trained manpower and supervision by STC and
STO.
1.
RNTCP patients put on DOTS are not included in the
1998 analysis.
2. The reporting of cohort analysis of treatment outcome
should improve.
2.
In Revised SCC format, PHIwise coverage of the
programme and important information on new
examinations of sputum and X-ray are not available.
3
STRENGTHENING
The "Cure" rate under NTP is not available. It can now
be obtained from revised SCC formats of quarterly
reports.
PRIMARY HEALTH CARE
The districtis the frontline unit for planning, organizing and managing primary
health care. Programmes have to be devised by governments, the voluntary
sector and communities, all planning and working together. While
considerable decentralization of authority is called for, overall national
guidance and monitoring have to be provided by government. District health
systems cannot be strengthened in isolation; the development of the whole
system is essential to Unfortunately, there are increasing pressures to organize
rersources for health along traditional, vertical lines the functionong of the
different parts.
The holistic approach to health care will not work without determinmation
and bold enlightened leadership. Unfortunately, there are increasing pressures
to organize resources for health along traditional, vertical lines and to pursue
goals and programmes in isolation. The organization of health systems based
on the comprehensive objectives of primary health care is difficult, and
practical experience in this area remains limited,particularly in international
organizations. The sceptics who claim that the district approach is woolly and
unmanageable should become more closely acquainted with what is going on
in the field and should help to encourage joint action. Those working at
government level in developing countries should be aware that the giving of
special attention to primary health care in the district is a logical step towards
health for all.
Source: Tarimo E & Fowkes FGR : Strengthening the backbone ofprimary
health care: Wld Hlth Forum 1989, 10, 79.
41
RNTCP
NTI Bulletin 1999.35/1-4. 14-17
Quality Assurance of Sputum Smear Microscopy in
Revised national Tuberculosis Control Programme
M Jayasheela*
SUMMARY
The paper describes the broad framework of the
steps in Quality Assurance of sputum microscopy and
how these have been built in under the Revised
National Tuberculosis Control Programme (RNTCP).
KEY WORDS : QUALITY CONTROL, PROFICI
ENCY TESTING, QUALITY IMPROVEMENT,
SUPERVISION.
INTRODUCTION
r 11he success of RNTCP depends upon proper
categorization of Tuberculosis (TB) patients and their
treatment under Directly Observed Treatment Short course
(DOTS). Sputum smear microscopy is essential for
proper diagnosis & categorization of the patients and is
thus invaluable for assessing the treatment outcome.
Quality Assurance Programme (QAP) has been given
due priority in the programme. The programme has
identified some national institutions as central laboratories
for this purpose1. National Tuberculosis Institute (NTI),
Bangalore is one such Institute to which eight State
Tuberculosis Centres (STCs) have been attached; they are:
Agra, Ajmer, Ahmedabad, Bangalore, Calcutta, Kangra,
Patiala, Srinagar, irrespective of their status of functioning
under National Tuberculosis Programme (NTP) or
RNTCP.
MERITS AND
MICROSCOPY
DEMERITS
OF
SPUTUM
The merits of sputum microscopy are (i) it is simple
and inexpensive and can be established in any part of the
country easily (ii) results are available quickly and are
easy to interpret (iii) can identify an infectious patient
and (iv) helps in assessing the effectiveness of the
treatment. The main demerit of sputum microscopy is low
sensitivity. Neither serological tests nor radiology can act
*
14
Former Consultant Microbiologist, National Tuberculosis Institute,
8. Bcllary Road. Bangalore 560 003
as gold standard because of their own inherent
weaknesses.
Though histopathology and cultural
techniques can be definitive in diagnosing TB. these are
not widely available and are difficult to establish. Hence,
the only practical way available as a Quality Control
measure for sputum microscopy is to ensure that a set of
given slides arc read by two readers and slides with
discordant results are read by an umpire reader whose
findings will be treated as final. It is not that the umpire is
always right. He may also make mistakes. But that is to
be accepted as a shortcoming of the system for want of a
better procedure.
QUALITY ASSURANCE PROGRAMME
International Union Against Tuberculosis and Lung
Diseases (IUATLD) has described Quality Assurance in
sputum microscopy2 under three heads viz., (a) Quality
Control (b) Performance Testing and (c) Quality
Improvement.
Quality Control: This essentially refers to the internal
monitoring of a laboratory that allows the frequency of
errors to be estimated against established limits of
acceptable test performance and helps in determining the
competence of their sputum microscopy services. Quality
Control of sputum microscopy envisages the following as
important inputs viz.,
i) Infrastructure: For a well equipped laboratory, it is a
must that it should have a good binocular microscope,
staining facilities, continued supply of quality
stains/reagents, slides, slide boxes, forms / registers and
such other necessities, besides running water supply and
power supply,. Care must be taken to ensure that there are
no shortcomings in this component.
ii) Sputum: Sputum brought out properly by the patient as
described in the programme manual3 is essential if sputum
microscopy has to be of any use to the programme. Every
effort has to be made to obtain proper sputum sample and
not merely saliva from a patient.
iii) Procedure : All the laboratories participating in the
programme should follow a single Standard Operating
ft
Procedure (SOP) as described in the guidelines for
preparation of sputum smear, staining, reporting and
recording so that the results of different laboratories arc
comparable.
3.
iv) Performance (human factor) : Sputum smear
microscopy for AFB is not difficult to learn; but its
challenge lies in the perseverance of continued high
quality reading. How well a technician performs and
continues to perform under routine conditions is of vital
importance.
Performance Testing : It is a method available for a
laboratory to assess its diagnostic services by comparing
its results with those obtained from other laboratories i.e.,
performance monitoring with inputs from an external
agency.
Performance testing can be carried out in three different
ways i.e.,
1.
2.
Centre to Periphery: The simplest way of assessing
the performance of a peripheral laboratory is to send
a panel of slides from a central laboratory which has
requisite expertise for reading/reporting and then
compare the results. As ready made slides are sent,
this procedure does not evaluate smear
making/staining abilities of the peripheral unit.
Similarly, as the technician has unlimited time for the
examination of slides sent and is aware of being
tested; this method does not allow the assessment of
his work ur ?r routine conditions. It can only
measure the . .lily of the technician to read the slides
correctly. This method can be used after training the
new technicians or conducting refresher courses to
assess the performance.
Periphery to Centre : This is a method of proficiency
testing wherein a sample of routine smears from a
peripheral laboratory are re-read at a designated
centre. Re-examination is done using the same
technique as used in the peripheral laboratory,
specially the same number of fields as specified in the
guidelines and not more to ensure proper comparison.
Findings of the umpire reader will be taken, as final
wherever there is discordance. This method is the
most suitable method of proficiency testing as it
allows routine work of technician in all its entirety i.e.,
smear, staining and reporting quality to be assessed
but not just the ability of the technician under test
conditions, w'ithout any extra work load on the
peripheral laboratory. As his routine work is going to
be assessed at a designated level on a regular basis, a
technician is bound to be extra cautious in carrying out
sputum microscopy.
Supervision : Visits by staff from a designated
laboratory to a peripheral laboratory would be of great
use in getting a realistic picture about (a) Personnel,
infrastructure, safety and equipment (b) stocks of
supplies/consumables (c) Registration, recording and
reporting and (d) performance, so that deficiencies can
be identified and remedies found. However, it is
difficult to organize visits to all the laboratories by a
national laboratory and such visits are necessarily to
be carried out by intermediate (State or District)
authorities.
Quality Improvement ; Il is a process of removing
obstacles for conducting quality sputum smear microscopy
successfully and ensuring non-recurrencc of the problems..
RNTCP encompasses all the above three components
viz.. Quality Control, Performance Testing and Quality
Improvement, to ensure quality sputum microscopy to be
carried out under the programme conditions3.
Quality assurance in TB control programme carried out
by NTI (centre to periphery) & from microscopy centre to
the TB unit & DTC (periphery to centre) and supervisory
visits to the TB unit and microscopy centre is shown in the
flow chart.
a)
Centre to periphery: NTI sends a panel of 100
stained sputum smear slides (made as per guidelines
given for the purpose) once in six months to these
STCs for reading/reporting. The panel consists of
negative slides as well as positive slides of various
grades, all arranged in a random fashion. Each
technician in the STC reads the slides independently
and the results are conveyed to NTI within a month,
which compares the results of each technician with its
own findings. Discordant results (false negatives or
false positives) when it is less than 5% are considered
satisfactory. The evaluation results with remarks will
be conveyed to the STC concerned. STCs in turn
would repeat such an exercise by sending a panel of
25 slides (with predetermined percentages of negative
and positive slides of various grades) to different
District Tuberculosis Centres (DTCs), once in a
quarter, under their jurisdiction. The DTCs should
ensure that each technician reads them independently
and send the results to STCs within a month.
b) Periphery to Centre: The Senior Tuberculosis
Laboratory Supervisor (STLS) after his monthly
supervisory visit to the microscopy unit, sends all the
slides with discordant results to a designated
laboratory (DTC) for verification. The results of re
examination with comments should be sent back to the
STLS/microscopy unit within a month.
15
I
Flow chart
r
Quality Assurance in TB Control Programme J
Centre to
periphery
Periphery to
centre
NATIONAL INSTITUTF ] |
100 Slides
| / 6 Months
STATE TB CENTRE
25 Slides / 3 Months
DISTRICT TB CENTRE
25 Slides j / 3 Months
Supervision
TB UNIT
(visits by STLS)
DTC/LT_______ |
Slides withjdiscordant results
TB UNO’(STLS)
Every month
Every] month
MICROSCOPY CENTRE
MICROSCOPY CENTRE
All positives & 10%
negatives
TB UNIT (STLS)
Supervision: Each microscopy centre is expected to
preserve all its slides till a sample from the lot has
been subjected to quality control. All the positive
slides and 10% negative slides will be re-examined by
STLS during his monthly visit to the unit. He will use
this opportunity to understand the plus and minus
points of that unit including sputum collection/smear
making/staining procedures, condition of the
laboratory, stocks of consumables etc., and to
guide/encourage the technician concerned to achieve
optimal performance. He should submit his findings
in the prescribed format given below to his controlling
officer.
c)
Proficiency Testing Periphery to Centre (an example)
Frequency . Quarterly
No. read as Pos. in the microscopy centre
No. confirmed by STLS
No. read as Neg. in the microscopy centre
No. reviewed by STLS
No. confirmed out of above
Evaluating report format
Initial reading
& No. of
slides
*
16
The purpose of quality control is to identify the
laboratories which are under performing for
improvement. Labeling a well performing laboratory
as under performing is of minor significance
compared to labeling an under performing laboratory
as good laboratory.
No. of
positive
slides (40)
39 (a)
01 (b)
(b/[a+bj)
1/39+1 = 1/40 x 100 = 2.5%
False Positive
No. of
negative
slides (40)
02 (c)
38 (d)
(c/[c+dl)
2/2+38 = 2/40 x 100 = 5%
False Negative
*
Scanty positive slides are the ones that are most likely
to be missed. Every effort should be made to correctly
identify such slides. Minor variations in the grading
of smears will not affect categorization of patients.
However, labeling positive slides as negative will lead
to wrong categorization and may deprive the real TB
patient from treatment.
*
Every negative slide that has been wrongly labeled as
high positive needs probing to ascertain the reasons
and to take remedial action. Normally, this will be
very rare and are mostly due to administrative error.
*
Sputum microscopy can neither differentiate
pathogenic and non- pathogenic mycobacteria nor live
and dead organisms.
40
30
398
40
38
Ziehl Neelsen Staining for acid fast bacilli will fade
with time specially in hot and humid conditions.
Under such circumstances, it is necessary to restain
the slides before taking them up for re-examination.
Percentage Discordance*
No.
No.
positive negative
* The permissible limit for both false positive and
negative is <5
CRUCIAL POINTS TO REMEMBER
*
Supervisors
reading
CONCLUSION
The procedure adopted for Quality Control of sputum
smear microscopy in RNTCP is simple and eminently
practicable and if implemented sincerely will ensure
accurate diagnosis and proper categorization of
tuberculosis patients, which is a must for the success of the
programme.
2.
Rieder HL et al: The Public Health Service National
Tuberculosis Reference Laboratory and the National
Laboratory Network, Minimum Requirements; Role
and Operation in a low-income country: ILJATLD
1998.
3.
Central TB Division, New Delhi: Manual for
Laboratory Technicians under RNTCP; DGHS, Min
of Hlth & FW, New Delhi 110 011; September 1997.
REFERENCE
1.
Directorate General of Health Services letter No.
Z.22/98-TB (TB division) dated 25th March 1998 on
Quality Assurance in sputum microscopy under
RNTCP
SEEING THE PATIENT H WHOLE "
Our present system of medical education is founded on the belief that the patient's symptom
arises from "disease". We are taught which symptoms» occur in which disease, we construct
seek by examination and investigation to identify the
differential diagnisis, and these we .
disease which has caused the symptom.
The reality, however, is quite the opposite. The vast majority of symptoms complaind about to
do not aim from disease, and are much more clearly related to "problems of living ", for want of
a bettc ■ phrase ......Disease and emotional problems commonly co-exist. It all feels the same to
bettc' phrase.
the patient. The only useful wayforward is to see the patient, in the round as an individual, with
a family role, a work role and a social role.
Source: Raymond Pietroni, in a Hospital doctor, Vol.3.No.24. 16 June 1983, P.19.&
World Health Forum, Vol.4.P.130
-
17
brief on
NATIONAL TB CONTROL PROGRAMME
Every year 2 million people in India develop tuberculosis (TB) and nearlv son nnn
X
m°re tha" 1000 ever* da* The Htoa" IrnmunodeScv Wo°
(HIV IS estimated to have infected 3 5 million people in India HIV associated TR
and the emergence of multidrug -resistant TB will increase the mannit J6d TB
seventy of the TB epidemic. Tuberculosis has devastating soc^l S as XeT
data suggests that each year more than 300,000 children are forced to lAaL
m0'e ‘han 100'00° 'vomen wi,h TB
TB'
are ^etSb^err'Ss'5
treated patients develop drug resistant and potentially incurable TBh 7
P°Or y
Programme review-1992 :
rogramme
tong accurately diagnosed and most
"patients since 1962, TB patients
National TB Control Programme
Revised National Tuberculosis Control Programme (RNTCP)
Based on the findings and recommendations of the Review the GovPrnmPnt
India evolved a revised strategy with the objective of curing at least 85°/ J * f
RnTp ^an6
Z
3t
ObseIveclSTtTalmentCashonrt Coure^fOOTs"
<
"’icros<:op* °<
atSnding'p':^^
given in
patient-wise boxes'^’Xto
peripheral health functionaries Ncn<= ond atment .thr°ugh involvement of
systematic monitoring, evaluation, and supervision™ levels°'Un,eerS' a"d V)
a-asm^Va^^Xm^S
in ’
•o a population of 13.85 milling? to assets
proved both ,ts technical and operational feasibility, a soft loan of US $142 million
Contd/...
1
3
million Xd SoOasX?35 been eXPanC""9 raP'd'V
reaC''ed aromd 313
First year of.the project i.e. 1998-99 - 89 million
Second year coverage i.e. 1999-2000-136 million
Third year coverage i.e. 2000-2001 upto December 2000
only - 313
By the end of this year it is expected that population coverage will reach
the approved 400 million.
DFID Assistance
'■
Sc^o. sx:
Jhe estlm^TedhcosroHhe'pro^^s'Rs.'hoo'GS^roTes^The^roposafhaT'beehn
approved by the Government of India. The service delivery in the 1s' year
'Str|cts of Andhra Pradesh was started from 3rd quarter 2000 DFID is also
providing assistance for strengthening Central TB Division.
DANIDA Support
Pc ->-i^NIDA has Started RNTCP in 14 districts of the state of Orissa at cost of
^95mXO^Sd Te
strategy Above phase I will expire by December 2001. Now under phase II the
Pjopos31 by Govt of Orissa to cover remaining 16 districts so as to cover whole
of State under RNTCP with help of DANTB is in progress.
Involvement of NGO/Private for the profit health providers.
Cnmm|mpent °f NG°S fnd Pr'Vate Practitioners 'n the National Tuberculosis
kpatmln7?9raTe 'S °f?ltal imPortance as a g°°d proportion of patients seek
treatment from them. Programme encourages participation of NGOs/PPs in
Programme implementation. An NGO policy has been formulated and widely
disseminated. Five different schemes for involvement of NGOs have been
.envisaged. Under the RNTCP, it is proposed to ensure full involvement of NGOs
j^in different activities of TB control, with selection and monitoring of performance
j.^being done at District/State levels. Depending on the capacity of the NGOs it is
•^.bemg proposed to involve them at appropriate levels in planning, programming,
'■r
S'' . .
5
Budgetary inputs in IXth Plan
Expenditure for TB control Programme are as under:
National TB Control Programmo
J .Ninth Plan (1997-2002)
; Approved outlay
. 1997-98
•' (Outlay as Budgeted
1997- 98 (Actual expenditure )
1998- 99 (Outlay as Budgeted)
! 1998-99 (Actual Expenditure )
i 1999-2000
i (Outlay as Budgeted)
| 1999-2000
| (Actual Expenditure)________
i 2000-01 (Outlay Budgeted)
j 2000-01 (Actual expenditure)
; 2001-02 (Proposed outlay)
Centrally
sponsored
scheme
450.00
j Amount (Rs.
in Crores)
80.00
80.00
32.00
72.00
68.50
32.00
72.00
68.50
95.00
95.00
87.50
110.00
110.00
160.62
87.50
110.00
110.00
160.62
450.00
Budget requirements for X Plan (2002-07)
Total Plan (2002-07)
(Proposed outlay)
' 1000.00
1000.00
During IX plan Rs. 450 crore is likely to be fully utilised for coverage of population
around 500 million under RNTCP. Therefore, approximately Rs. 1000 crore will
periodqUired
C°Verage of entire P°Pulation of ^e country during 10th plan
■
Status of RNTCP
1
GE?
Central TB Division
Ministry of Health and Family Welfare
Nirman Bhavan. New Delhi 110 011
Burden of Tuberculosis
"If the number of victims which a
disease claims is the measure of its
significance, then all diseases,
particularly the most dreaded
infectious diseases, such as bubonic
plague, Asiatic cholera, etc., must rank
far behind tuberculosis."
Robert Koch. Berliner klinische Wochenschrift, 24 March 1882; 19:221
India Accounts for Nearly One
Third of the Global TB Burden
TB: Burden of Disease
• 14 million TB cases out of which about 3.5
million are highly infectious sputum positive
cases
• 2 million new cases out of which 1 million are
new smear positive
• 5 lakh deaths/year, >1,000 deaths/day, 1/minute
• Multidrug-resistance and HIV are making the
epidemic much worse
H-
Social and Economic Burden
• Indirect costs to society of $3 billion per year
• Direct costs of $300 million
• Loss of 100 million productive work days per year
due to illness only
• Every year, due to TB deaths, India is losing more
than 130 crore productive work days
• More than 300,000 children leave school as a
result of parents’ TB
• More than 100,000 women rejected by families on
account of TB
TRC. Sodv-aconomA: *np*c< of TB on pM*ils ontf
Status of TB Control in India
• 1950s-60s: Important TB research
• 1962: National TB Programme
• 1992: Programme Review- only 30% of patients
diagnosed and only 30% of those treated
successfully
• 1993: DOTS pilots
• 1998: DOTS scale-up begins
• 2000: > 30% of country covered by DOTS
h lnO». he J Tub Lung Da 1999 3: M9e77
1
T
i
'
■
2
Ma
STU
X
5oo
ic
(2.
103
3
^36
■y
n
fco
y..
u
20-Fold Expansion of DOTS
in India in just 3 years
TB Control- cox erage plans
Population Covered by DOTS. Millions
350
• 2001: 400 million population by end of this
year
• 2004: 700 million population by March
• Entire country as soon as technically and
operationally feasible
■
130
18
July 1998
July 1999
Mar-01
More than 1/3^ of the population of India has access to DOTS
■ArdHa
PrxJeeh
OAmm
OBrfw
■Deta
DOwM
■ Mmmi
Pmdeefi
OKmiiU
i
■KvMB
■MMM
-
■CM
■
■WwtBcrvtf
Case detection and treatment success
RNTCP (1998/1999) and Non-RNTCP (1997/1999) areas
100%
Terpei Z*fte
90%
OrtaMj
80%
2
• Rapid expansion
- covering I/S'4 of country
- 2nJ largest DOTS programme in world
• Accurate diagnosis
- More than 4 lakh microscope exams per month
- Diagnostic confinnation at globally expected levels
• Reliable cure
- More than 8 of 10 patients cured (compared with less
tlian 4 of 10 in previous programme and private sector)
70%
60%
tS
50%
g3
o
40%
co
30%
.
20%
•5'
..
10%
0%
0%
E
10%
20%
RNTCP
Nri-RNTCP
30%
I
I
40%
RNTCP results in India
50% 60%
70% 80% 90% 100%
Detection Rate
2
RNTCP: impact to date
• Each month, nearly one crore patient visits to health
facilities, >30,000 sputum slides are examined &
> 3,000 patients started on treatment
• > 5 lakh patients placed on treatment, saving nearly
I lakh lives & preventing >10 lakh infections
• To achieve above
Systematic monitoring and
accountability
• Good record keeping is the cornerstone of
success /
• The DOTsf recording system enables
->1.5 lakh health workers underwent modular training
- 1.400 supervisors appointed
- 75 lakh microscopic examinations
- nearly 2 crore directly observed doses
- monitoring of patient outcomes
- evaluation of programme performance
- analysis of epidemiologic data
- operational research
• Every level of the health system accountable
for diagnosis and cure
Supervision
Operational problems
• Effective supervision at all levels is key to
success
• Key areas:
/laboratory work
/ patient categorization
/ direct observation
/drug storage and stock
/ record keeping
/ reporting
Evaluation: Joint programme review
in Feb 2000
Key findings:
Successful implementation
-patients accurately diagnosed
-drug supply regular and uninterrupted
-increase in proportion of patients cured
-sound technical and training policies
-intensive monitoring & supervision
• Weak primary health care infrastructure
• Low priority given to tuberculosis in most states
• Extensive use of private sector which provides poor
quality diagnosis, non-standard treatment regimens,
lack of record-keeping, and lack of follow-up
resulting in poor treatment outcomes
• Involvement of private sector, non-govemmental
organizations, civil society
Evaluation: Joint programme review
in Feb 2000 (contd.)
Key findings:
Constraints
-not all health care providers are involved
-difficult to coordinate care in areas with
weak primary health care system
-treatment observation often not
sufficiently convenient to patients
-lack of awareness of the programme
3
Strategic directions for RNTCP
• Patient as VIP in reality and not name onlytreated respectfully
• Involvement of all sectors
• Expansion to cover entire country in phased
manner
• Decentralisation of functions to States
Alternative approach to service
delivery
• Involvement of corporate sector
• Implementation of NGO policy
• Involvement of PPs
• Other health sectors: Medical colleges, ESI,
CGHS, Armed forces
• Community volunteers as DOT providers
Service Delivery by NGOs
• According to RNTCP policy
• Schemes for
- Health education and community outreach
- .Provision of Directly Observed Therapy
- In-Hospital care for TB disease
- Microscopy and treatment centre
- TB Unit Model
Mechanism of Decentralization
Societies formed at States and District levels and
provided with funds
State specific PIPs for progressive decentralisation
of functions
At state level
-Additional financial powers
-Planning & budgeting
-Project monitoring
-Logistic management
Key roles for NGOs
• Community education - awareness of availability
of and right to services
• Making treatment observation patient-friendly and
patient-centred
• Supporting government health services
• Operating health services independently,
following policies of, and coordinating with,
government authorities
• Watchdog function -- constructive engagement to
ensure availability of services as per policy’
Decentralization to States
• States to become primary unit of
accountability for Central Govt, rather than
districts in a phased manner
• Capacity to be built at State level
• Programme, financial, and logistics matters
to be decentralized to States
4
Decentralization: Programme
• Capacity and funds to monitor and
supervise (STC, STDC)
• Quarterly meetings at State level
• Feedback from State to district
• Monitoring at TU level not only district
level
Decentralization:Financial
• Capacity and funds to handle and monitor
financial matters
• Tracking of SOEs and correction
• Annual audits from single agency State
wide
• Preparation of budgets
• Rapid problem-solving by Secretary/others
within State
Current Assistance
Decentralization :Logistics
• Training of State TB Officers
• Specific to States
• Most states all RNTCP loose drugs
• Most states state-wide distribution
• Printing at State level
Donor
Agency
Period of
the Project
World Bank
1997-2002
604.86
DANIDA
1998-2003
31.90
DFID
2000-2005
109.93
Total
Expenditure
assistance
up to
(Rs. In Crores) 1999-2000
133.00
Current Assistance
Year
World Bank
Assisted
Project
20
* DANIDA
Assisted
Project
1998
20
5
1999
135
6
2000
250
10
22___
2001
360
13
45
2002
400
1997
“ DFID
Assisted
Project
75
DANIDA assistance will cover 14 tribal districts of Orissa
with a population of above 13 million
** DFID assisted project will cover the entire population of
about 75 million In Andhra Pradesh
*
■>v
bi
5
100
90 -
E
■
I
: Wl
■. -M
s
<D
O)
5
c
o
80 -
o
!■<
/
Q.
.
c
'q
■■
‘
fe-
□ Andhra Pradesh
□ Assam
□ Bihar
■ Delhi
□ Gujarat
□ Haryana
□ Himachal Pradesh
■ Karnataka
□ Kerala
' I'
□ Madhya Pradesh
J
:
□ Maharashtra
‘
;
•
. '
i
• -
'
70 -
.1
J
i
siif
60
3 month conversion rate(SPP)
States
■ Manipur
■ Orissa
□ Rajasthan
■ Tamilnadu
□ Uttar Pradesh
□ West Bengal
100 1
B Andhra
Pradesh
□ Assam
95 -
□ Bihar
90 -
■ Delhi
□ Gujarat
85 -
80 -
o..
<D
2
0)
B Himachal
Pradesh
□ Karnataka
■ Kerala
75 -
8□
B Madhya
Pradesh
□ Maharashtra
I
70 -
□ Manipur
■
65 -
■ Orissa
■ Rajasthan
60 -
B Tamilnadu
55 -
it
B Uttar Pradesh
□ West Bengal
50 J
States
Government of India
Ministry of Health and Family Welfare
(National AIDS Control Organization)
National AIDS Control Programme
1. Estimated burden (1996, 2002-2006)
The National AIDS Control Programme is a 100% Centrally
Sponsored Scheme (CSS) implemented as a vertical programme through
State AIDS Control Societies (SACS). Keeping in view the complexities of
HIV infection, non-random distribution, occurrence of behaviour influencing
HIV transmission, it is difficult to make exact estimates of HIV prevalence.
It is more so in the Indian context, having its typical and varied cultural
characteristics, traditions and values with special reference to sex related
risk-behaviours. To estimate the total burden of HIV infection in the
country, some efforts had been made in the past by WHO/UNAIDS by using
......
the data generated by NACO and other publications from various research
institutes in India. The different estimates of disease burden iin the country
reported by these organizations are as follows:
1.1
Year
Agency
Estimates
1990
GPA/WHO
0.05-0.2 million
1992
GPA/WHO
1 million
1993
GPA/WHO
2 million
1994 (end)
NACP
1.75 million
1996 (mid)
UNAIDS/WHO
2.5 million
1997 (mid)
MAP
2.5 million
HIV Sentinel Surveillance:
The Sentinel surveillance fo^HIV? infection was initiated in the
country m year 1994 by identifyin^gsentinel sites including, both high
risk group sub-population, i.e. STD clinic attendees and low population sub
•"S’- •
*
I
teb
■
3b V.. v'.vvvv/' /
group at antenatal clinics. In year 1998, the network of sentinel sites was
expanded to 180 sites all over the country. During August-October, 1998,
round of sentinel surveillance, all States/UTs participated. Since then, every
year, during the period of August-October, Sentinel surveillance is
conducted covering the same sites. A sample size of 250 samples for high
risk group, attendees and 400 samples for low risk group of population, i.e.,
antenatal clinic attendees are collected, maintaining “Unlinked Anonymity”
of whole procedure. In year 2000, the number of sites was increased to 232.
Based on the Sentinel surveillance, States/UTs can be categorized into: (i)
High prevalence States- where HIV prevalence among antenatal mothers is
1% or more. These are Maharashtra, Tamil Nadu, Andhra Pradesh,
Karnataka, Manipur, and Nagaland; (ii) Moderate prevalence States, where
HIV prevalence rate in antenatal women is less than 1% but the prevalence
among high risk group is 5% or more. The States/UTs like Gujarat,
Pondicherry and Goa fall in this category; and (iii) Low prevalence States:
where HIV prevalence among high risk group is less than 5% and among
antenatal women is less than 1%. All other States belong to this category.
1.2
Accuracy and Reliability:
In order to maintain the accuracy and reliability of data following
steps have been taken:(•)
(ii)
1.3
Consistency in sites: Sentinel Surveillance is conducted at the
same sites each year during the same period of the year, i.e.
August-October.
Training of medical and paramedical staff: All functionaries of
the sentinel sites as well as the testing laboratories are trained/
retrained each year before commencement of the Sentinel
Surveillance.
Supervision :
Three tier supervision mechanism is adopted to maintain the quality of
the data collected through sentinel surveillance system. These are as:
(a)
Task Force Members: Govt, has constituted a task force on
Surveillance, comprising of experienced biostaticians,
epidemiologists and researchers. They conduct field visits and
provide feed back to NACO as well as State AIDS Control
Societies for any mid course correction.' i
■
Jk'
2.
V
Ofc. ViSSiSBB ««
(b)
(c)
Regional Co-ordinating Teams: National Institute of Health is
the nodal Institutions for coordinating sentinel surveillance
activities. They have constituted a team of experts belonging to
the disciplines of Microbiology, Epidemiology and
Venereology in medical colleges. These teams make field visits
twice during the period of sentinel surveillance to provide
technical support to the functionaries of sentinel sites or
laboratories where samples are tested for HIV.
State AIDS Control Societies: Officers from SACS also visit
sentinel sites randomly to ensure the quality of data being
collected at the Sentinel sites and laboratory procedures being
followed for HIV testing.
For standardization of the procedure, check -list and
manual are provided to functionaries involved in sentinel
surveillance operation.
Based on nationwide Sentinel surveillance results and
certain assumption taken into consideration, the estimated
number of HIV infections in year 1998, 1999 and 2000 is
is as
as
follows:
' A
1998
Point
Estimate
2.97
(In Millions)
20% range
Lower
Higher
2.32
3.51
1999
3.09
2.43
3.72
2000
3.22
2.58
3.86
Year
During the period, 2001-2006, it is estimated that the
number of new infections would be 4-6- million.
1.4
Problems in Making Estimates:
It is an institutional based surveillance and therefore, the catchment
area may vary from are round to another. The samples collected at sentinel
sites may not be the true representatives of the population. Besides, the
sentinel sites may not capture all the subpopulation sub groups practicing
high risk behavmur. The^ssumptions made in HIV estimates need to be
validated time to time.
.
A-T’
1.5
Methods for improvement in 10th Plan Period.
It is envisaged to cover additional sentinel sites in rural areas in order
to estimate the disease burden separately for rural and urban areas. The
assumptions being- used currently in estimation of HIV disease burden would
be validated by conducting community based cross sectional studies.
Besides, reporting of AIDS cases and deaths should strengthened.
2.
Status of National AIDS Control Programme
In order to combat the onslaught of the HIV/AIDS epidemic
effectively, the Government of India established National AIDS Control
Organisation (NACO) in 1992. A National AIDS Board was constituted
under the chairmanship of Secretary (Health), Ministry of Health & Family
Welfare in order to review NACO policies, to expedite sanction, approve
procurement and to undertake and award contracts to private agencies. A
comprehensive programme with World Bank assistance was initiated in
1992 for a period pf fiye years, which was extended till March, 1999.
Achievements included much higher levels of awareness creation, the
putting in place of State level structures for programme implementation and
improvements in "Blood safety".
The launch and consolidation of
successful individual intervention projects such as the innovative work in
Sonagachi amongst commercial sex workers and breakthroughs in reaching
out to college youth through University Talks AIDS (UTA) programme
were amongst its achievements. The scope of these efforts remained,
however, on a limited scale; political acceptance was weak and ownership of
the programme by the States proved difficult to establish. Involvement of
Non Government Organisations (NGOs) at the peripheral level and of
quality needed for sustained interventions proved difficult to obtain. And
while the emphasis was on blood safety and strengthening of infrastructure,
the approach remained primarily medical, with HIV being seen as largely a
health issue. AlthbuglTMhe Is* piiase of the National AIDS Control
Programme was for five years, due to slow implementation in the initial
period it was extended to March, 1999.
In order to strengthen the programme management at the state level,
the State Governments have established their own managerial organisations,
which include State AIDS cells, Technical Advisory Committees and
empowered committees as per the guidelines of the strategic plan.
S'.
1- V
'1: -
ACHIEVEMENTS:
1.
Intensive awareness campaigns through electronic and print media
and the field publicity units of the Ministry of Information &
Broadcasting in both the urban and rural areas resulted in generation
of awareness about the disease both in the high risk groups and the
general population. Awareness levels are of the order of 60-65 per
cent on an average in urban areas and 35-40 per cent in rural areas.
The highest awareness levels are in Tamil Nadu where it is 95% in
urban areas and 75-% in rural areas.
2.
Awareness programs through school and college education has been
taken up on a large scale in 18 States.
3.
To ensure safe blood to the population, 815 blood banks in the
Government and voluntary sector modernised in phases and 40 blood
component separation facilities have been established throughout the
country. Mandatory testing of blood for HIV, Syphilis, Malaria and
Hepatitis B, has been introduced throughout the country. Infection
through blood transfusion has been brought down appreciably over
the last 2-3 years.
4. .
For control ptfS^gjally Transmitted Disease which have a direct
correlation with HIV/AIDS, 504 STD clinics in district hospitals have
been taken up for modernisation. Syndromic management of STD
cases has been introduced and doctors are being put on intensive
training of Syndromic management techniques for STDs.
5.
Over 2200 key trainers for training of doctors have been trained in
clinical management of HIV/AIDS diagnosis. The Indian Medical
Association (IMA) has so far trained 16,000 general medical
practitioners with support from NACO.
6.
For the promotion of voluntary counselling and testing, 141 voluntary
blood-testing centres have been sanctioned. These centres at presently
are mostly located in Medical College Hospitals.
7.
For tracking the epidemic in the country, 180 sentinel sites have been
established. These sites include both high-risk groups like Sex
Workers, Intra-venous Drug Users as well as low risk group pregnant
.
■
•
-
5
•‘i-:
V :
sA
women attendingjjntenatal clinic. One round of sentinel surveillance
is carried out each year during the period of August - October.
8.
9.
Targetted Interventions : Several targetted intervention projects have
been implemented for groups practicing risky behaviour. These
interventions include outreach activities, IEC and interpersonal
communication, condom promotion, and general health and STD
service provision. Targetted intervention projects such as that for
commercial sex workers in Calcutta’s Sonagachi area, the Men who
have sex with Men project in Chennai, truck drivers in Rajasthan and
Injecting Drug Users in Assam, Manipur and Nagaland have increased
the use of condoms and reduced STD, yielding lessons in best
practice.
'
Role of Non-Gpypxoment Organisations : NGOs have played a major
role in initiating and ensuring effective interventions, defending the
human rights of people living with HIV/AIDS and in providing care
and support to people living with HIV/AIDS.
The government of India has also made active efforts to involve
NGO participation. State AIDS cells have been particularly effective
in mobilising community involvement. The Maharashtra State AIDS
Cell, for example, works through a nodal NGO to coordinate the work
of all other NGOs in the state, while the TNSACS has a nodal NGO
officer on its staff. An NGO-AIDS Cell has been established at the All
India Institute for Medical Sciences. The States AIDS Cells are
empowered to sanction grants to NGOs.
10.
Intsrssctoraf 'GoltaboratiGn
NACO has promoted intersectoral
collaboration with other ministries such as Human Resource
Development, Information and Broadcasting and the Railways.
NACO has successfully collaborated with the Steel Authority of India
and the Indian Oil Corporation in HIV/AIDS prevention efforts.
11.
Private Sector Collaborations : Effective collaborations have also been
built with the private sector through the Confederation of Indian
Industry and the Bengal Chamber of Commerce. The Tata Iron and
Steel Company, for example, has incorporated HIV/AIDS prevention
in their ongoing family welfare programmes.
■
■
^
&
r
W'-"'
.
■■
■
ft gf I
■
-MOS
National AIDS Control Project - Phase II (1999 - 2004)
This phase of the programme began w.e.f. April, 1999. Giving a major
focus to targeted interventions amongst groups with the highest risk
behaviors (sex workers (SWs); injecting drug users (IDUs); truck drivers)
and broadening the approach to a multi-sectoral one. The current phase of
the national programme has seen the emergence of a strongly de-centralised
programme with responsibility for implementation clearly placed with the
States. State AIDS cells were created in all the 32 states and UTs of the
country for the effective implementation and management of the National
AIDS Control Programme. However to remove the bottlenecks faced by the
programme implementation at the State level new and more flexible State
structures of State AIDS Control Societies have been formed with strong
mechanism for programme management at state level, including a strong
NGO component of targeted interventions, supported by efforts for
mobilising the community around awareness and treatment of sexually
transmitted diseases/reproductive tract infections. Innovative approaches to
providing technical support to state programmes have been launched through
a network of 12 Technical Resource Groups (TRGs) each covering different
thematic areas of the epidemic. Surveillance has been both expanded and
strengthened. With a new round of resource mobilised from Government of
India, the International Development Agency, major bilateral donors and the
UN system, the programme is moving into an important new phase of
implementation.
The National AIDS Control Project - Phase - II aims :
(i) To shift the focusfrom,Faising awareness to changing behaviour through
interventions, particularly for groups at high risk of contracting and
spreading HIV;
(ii) To support decentralisation of service delivery to the States and
Municipalities and a new facilitating role for National AIDS Control
Organisation.
Program delivery would be flexible, evidence-based,
participatory and to rely on local programme implementation plans;
(iii) To protect human rights by encouraging voluntary counselling and
testing and discouraging mandatory testing;
(iv) To support structured and evidence-based annual reviews and ongoing
operational research; and
■
■
'T-r
'
•
-
(v) To encourage management reforms, such as better managed State level
AIDS Control Societies and improved drug and equipment procurement
practices. These reforms are proposed with a view to bring about a sense of
’ownership1 of the programme among the States, Municipal Corporations,
NGOs and other implementing agencies.
Project Objectives
Phase II of National AIDS Control Programme has two key objectives
namely:
To reduce the spread of HIV infection in India; and
(a)
(b)
Strengthen Indiaycapacity to respond to the HIV/AIDS on a long term
basis.
Some special features of program delivery and management in Phase II:
•Delegated financial and administrative authority to NACO
•Ownership of the state and decentralised programme at the state level
•Involvement of the community in social mobilization and awareness at the
grass-root level
•Major role of the NGOs in the implementation of intervention programs
for marginalised population
•Involvement of democratic institutions (Panchayati Raj) and youth
organisation at the district, block and village level
Project Scope
Reflecting the extreme urgency with which HIV prevention and
control need to be pursued in India, the AIDS- II project of the National
AIDS Control Programme covers across all State and Union Territories as a
Centrally Sponsored Scheme with 100% financial assistance from
Government of India direct to State AIDS Control Societies and selected
Municipal CorporatiomAfDS Control Societies. The scope of the project
would vary with each intervention, taking into account the need and
absorptive capacity, feasibility and efficiency. The immediate need is to
have a paradigm shift in our response against HIV/AIDS at all levels, with
the overall goal to contain the further spread of HIV at a fairly low level of
HIV prevalence.
Project Targets
The programme has the following firm targets to be achieved during project
period
** -
2
(i)
To reduce blood-borne transmission of HIV to less than one percent
of the total transmissions.
(ii) To introduce Hepatitis C as the fifth mandatory test for blood
screening.
(iii) To set up 10 new modem blood banks in uncovered areas, upgrading
of 20 major blood banks, setting up of 80 new district level blood
banks in uncovered districts, establishing another 40 blood component
separation units, promotion of voluntary blood donation and increase
its share in total blood collected to at least 60%. The total blood
collection in the county which is now around 3-3.5 million units is
sought to be rarsed tO 5-5.5 million units by the end of the project.
(iv) To attain awareness level of not less than 90% among the youth and
those in the reproductive age group.
(V)
To train up at least 600 NGOs in the country in conducting targeted
intervention programmes among high-risk groups and through them
promote condom use of not less than 90% among these groups and
control of STDs.
(vi) To conduct annual Family Health Awareness Campaigns among the
general population and provide service-delivery in terms of medical
advice and provision of drugs for control of STDs and Reproductive
Tract Infections (RTIs). These campaigns will be conducted jointly
by NACO and RCH programme managers at the State level. Through
this it is proposed to reduce the prevalence of STDs/RTIs in the
general commumlyjrom the present level of about 15-20%.
(vii) Promotion of voluntary testing facilities across the country at the end
It is visualised that every district in the country would
/J of the project.
have at least one voluntary testing facility.
(viii) Awareness campaigns will now be more interactive and use of
traditional media such as folk arts and street theatre will be given
greater priority in the rural areas. It is proposed to cover all the
schools in the country targeting students studying in Class IX and
Class XI through school education programmes and all the
universities through the “Universities Talk AIDS” programme during
the project period.
(ix) Promotion of Organisations of people living with HIV/AIDS and
giving them financial support to form self-help groups.
COMPONENTS OF PHASE - II
1. Priority targeted interventions for populations at high risk
/
,
This component of the project aims to reduce the spread of HIV in groups at
high risk by identifying target populations and providing peer counselling,
condom promotion, treatment of sexually transmitted infections etc. This
component would be delivered largely through Non Government
Organisations, Community based Organisations and the Public sector.
2. Preventive interventions for the general population
The main activities would be: (a) IEC and awareness campaigns; (b) provide
voluntary testing and counselling; (c) reduction of transmission by blood
transfusion; and (d) prevention of occupational exposure.
3. Low Cost care for people living with HIV/AIDS
Under this component activities would provide financial assistance for home
based and community based care, including increasing the availability of
cost effective interventions for common opportunistic infections.
4. Institutional strengthening
This component aims to strengthen effectiveness and technical managerial
and financial sustainability at National, State and Municipal levels,
strengthening surveillance activities and building strong Research &
Development component, including operational research etc.
S.Inter-sectoral collaboration
This component would promote collaboration amongst the public, private
and voluntary sectors. 'The activities would be co-ordinated with other
programmes within the Ministry of Health & Family Welfare and other
central ministries and departments. Collaboration would focussed on; (I)
learning from the innovative HIV/AIDS programmes that exist in other
sectors; and (II) sharing in the working, generating awareness, advocacy and
delivering interventions.
6. Outlay for National AIDS Control Programme - II
There are three funding agencies for Phase II of National AIDS Control
Programme. These are :
International Development Agency (IDA) : World Bank has
(a)
provided an account of US $191.00 million as soft loan for the project
io
(b)
(c)
and the domestic budgetary support is US $ 38.8 million. The total
outlay is US $ 229.8 million (Rs. 11550.00 milion).
United States Agency for International Development (USAID) :
United States Agency for International Development has extended its
assistance to the Govt, of Maharashtra for implementation of
'AVERT* Project based on their earlier experience of working in the
State of Tamil Nadu under the AIDS Prevention and AIDS Control
(APAC) project. The outlay for USAID assisted ‘AVERT’ project is
Rs. 1660.00 million.
Department For International Development (DFID) of U.K.
Government : Department For International Development has
extended its assistance for implementation of Sexual Health Projects
in the States of Andhra Pradesh, Gujarat, Kerala and Orissa in
continuation of their earlier projects in West Bengal. The outlay for
this project is Rs. 1040.00 million.
OUTLAY FOR NATIONAL AIDS CONTROL
PROJECT PHASE-II
• IDA credit (1999-2004)
RUPEES IN MILLION
11550
• USAID assistance Tor AVERT Project in
Maharashtra.___________________________
• DFID assistance for Sexual Health projects for
the States of Andhra Pradesh, Gujarat, Kerala
and Orissa.
TOTAL
1660
1040
14250
Monitoring and Evaluation of the Programme.
For the effective monitoring and evaluation to assess the
implementation of the Phase-II of the National AIDS Control Project at
National and State leV'et'J'tftS following mechanism has been envisaged.
j
(i)
(ii)
(iii)
(iv)
■TjuM
•J
—-
Creating a Comuperised Management Information System (CMIS) at
the National and State levels;
Training NACO Staff and Health specialists in evidence based health
programme management.
Conducting base line, mid term and final evaluation;
Conducting the Annual Performance and Expenditure Review
(APER); and
i .
---- TuTT'
-
H
L
(V)
Conducting the National Performance Review (NPR) under the
National AIDS Control Board.
A National level independent outside agency is being identified who
would be assigned the^ssponsibility of development of CMIS, conduct of
base line, mid-term and end term evaluation.
Financial Management System
It is envisaged to maintain an adequate project financial management
system to provide accurate and timely information regarding project
resources and expenditure to facilitate efficient project management. For this
purpose an consultancy agency is being selected for developing a Project
Financial Management System for NACP-II.
The financial management system would be integrated one for the
whole project. A comrqqn jset of policies and procedures would apply to the
entire project and a consolidated set of financial reports for the project
would be prepared from the FMS.
Utilization of funds:
Year
(Rs. In Rupees)
Budget Estimates
Expenditure
1997-98
100.00
123.01
<^’75
1998-99
1999-2000
'1WT
135.25
2000-2001
180.00
168.69
2001-2002
17.
Proposed objectives. Strategies , Initiatives and target for the 10th Plan
Objectives: During the 10Ih plan period the objectives of the programme
would be:
(a) to reduce the spread of HIV infection
(b) to strengthen the capacity of the existing health care system to respond
to HIV/AIDS on long term basis
(c) to make accessible and affordable treatment for opportunistic infections
to people living with HIV/AIDS up to the level of Primary Health
center.
Strategy : to achieve the objectives laid down for 10th five year plan,
emphasis would be on decentralization of the programme at the State level,
making the States/HTS'-TKponsible for implementation of the programme
through the existing health care system. The following strategic themes would
be pursued:
(i)
Expansion of targeted interventions: Based on the experiences
gained from the Ilnd phase of NACP, at least one targeted
intervention will be implemented in each district of the country.
Similarly, in cities, targeted intervention projects would be initiated
for sex workers, intravenous drug users, truckers, migrant workers,
men having sex with men etc. A package of services like education,
counseling, treatment of sexually transmitted diseases/reproductive
tract infection and condom promotion will be delievered under these
projects. Efforts would be made to bring other social sector
departments..-.Jik&- Education, Social Justice & Empowerment
Women and Child development. Police, Urban basic services etc.
with the objectives to provide other services not included under the
targeted intervention projects to the beneficiaries and their families.
This would create an enabling environment which would help these
subgroups to seek services provided under the programme and
motivate them to change their risk behaviour. It is envisaged to ■■
implement 2000 such projects all over the country during plan
period.
(ii)
’
" 1
Strengthen treatment of Sexually Transmitted Diseases & Condom
promotion: As STDs/RTIs have close linkage with that of HIV
transmission, facilities for treatment of such infections upto the
primary health centers would be promoted. The health care
j o
r
functionaries would be trained in counseling skill so such services
can be provided to the patient without additional financial burden.
For the referral services, all 3000 community health services-would
be strengthened by providing laboratory equipment as well as
training of the laboratory staff. A recurring grant of Rs. 25,000/CHL
would be~giverr to each CHC to meet out the expenses on
procurement of RTI/STI drugs and other consumable needed for this
component of the programme.
Condom promotion would be pursued vigorously in coordination with the department of Family Welfare. Both free
distribution and social marketing schemes of condoms would be
promoted by involving NGOs/CBOs and other possible retail
(Hi)
outlets.
Information, Education and Communicatiom
In absence of any cure for HIV infection and preventive vaccine to
protect the community, awareness is only tool for preventing the
spread of HIV infections. Govt, plans to carry out the following
activities to empower the people in general and sub groups of
population^k^k to protect themselves from acquiring HIV
infection.
(a) Electronic and Print media : Govt, will use all available
electronic media channels centrally as well as locally to
disseminate information on HIV/AIDS prevention. Cinema
celebrities as well as politicians would be used for advocacy of
purposes.
(b) Field Publicity units and Song & Drama Division of Ministry of
Information and Broadcasting : There is a nationwide network
of these units of Ministry of Information and Broadcasting.
They will be extensively used in dissemination of information
on Reproductive Health Issues, including Sexually Transmitted
Diseasg^^HIV/AlDS. An integrated programme will be
developed in collaboration with the Deptt. of Family Welfare.
Ministry of Information and Broadcasting would also be
requested to incorporate HIV/AIDS messages in all programmes
being telecast or organized through their various field units.
State AIDS Control Societies will be asked to use local channels
for making the people aware about HIV/AIDS and other
common STDs.
IM
[
(c) AIDS Education in Schools & Colleges : During the Ilnd Phase
of NACP, 90% of the schools and colleges will be covered
under AIDS Education Programme. During the 10th plan period,
the programme will be pursued in collaboration with Deptt. of
Education and NGOs. University Talk AIDS will also be
implemented in all colleges and universities in the country in
collaboration with the Deptt. of Sports & Youth Affairs.
For the coverage of non-school youth in rural and
urban slums, a programme will be chalked out in collaboration
with NYKrorganization of Deptt. of Sports & Youth.
(d) Family Health Awareness Campaign : Govt, has initiated this
programme since 1999, with the objective to sensitise the people
in age bracket of 15 to 49 years of age about Sexually
Transmitted Diseases and HIV and AIDS through house to
house contacts, conducted by health and non-health
functionaries
by
organizing
health
camps
at
Subcenter/PHC/CHC level for treatment of STDs and follow
group education. The strategy is based on the principle of
interpersonal communication between the health functionaries
and the client for dissemination of facts about STDs and
HIV/AIDS : During the 10th plan period, at least one fortnight
eacky^rbe dedicated to this campaign.
(e) Multisectoral Involvement : At present, several Ministries and
Deptts, NGO, Private Sector are involved in implementation of
the prgoramme components in their own settings. Deptt. of
Health provides funds to these organizations for carrying out
EHV/AIDS related prevention and control activities. NACO
would encourage more Ministries/Deptt. to take initiative in
making HIV/AIDS prevention and control activities as one of
the core activities of their ongoing programmes on
training/education and health care services.
.
‘ 3
>
(iv)
'VTLLI.
t
'■C
u >\r*-
) U--.
J
—.O-
J'-*
I
Ki.-
*
Promotion of Voluntary Counseling and Testing for HIV : V oluntary
counseling^nd testing is one of the key intervention to normalise the
epidemicmW^ommunity. The objective is to motivate the people
with high risk behaviour to get themselves tested for HIV in order to
counsel them for a change of their risk behaviour. At the same time,
zu-a.
»t
5
it also gives them opportunity to know about the facilities available
for care and-ssupport. During the find phase of the NACP, all the
districts in the country will have at least one voluntary and
counseling and testing center (VTCT). During the 10lh plan period it
is envisaged to extend this facility up to Community Health Center
(CHC). Efforts would be made to cover all 3000 existing CHCs
during 10th plan period.
(V)
Blood Safety : During the Ilnd phase of the National AIDS Control
Programme, NACO would be able to establish Blood banking
facility in all districts of the county and component separation
facility in 40 additional institutions. Besides, 10 State of art model
blood banks will be established, mostly in backward States of the
country. During 10th plan period, 60 blood component separation
units will b£ established in order to cover most of the public sector
medical colleges of the country. NACO in collaboration with Deptt.
of Family Welfare would establish blood storage facilities at least in
25% of the existing Community Health Centers. Priority would be
given to these centers where First Referral centers have been
sanctioned by Deptt. of Family Welfare.
(vi)
Care & Support of People Living with HIV/AIDS (PLWHA) : It is
estimated that the number of HIV/AIDS patients will increase
manifolds during the 10th plan period. Hence, facilities for treatment
of the patients would have to be created by strengthening existing
health care institutions as well as community based health care
centers. During the 10th plan period, it is envisaged to :
(a) Increase the number of beds in all district hospitals of high
prevaletlce'S’tates by 20%, and 10% in about 25% of the district
hospitals in other States.
(b) Strengthening of diagnostic facilities for diagnosis of
opportunistic infection in all district hospitals of high prevalence
States & 25% of the districts in other States.
(c) Strengthening hospital infection control measures.
(d) Availability of anti-retroviral drugs for the post-exposure
prophylaxis for the use of Health Workers in Public Sector
Hospitals up to district level.
i\
' V'll
'■ • U. « i. '• V-t
S'. :.v Ik4-
I /
c
(vii) Training of Medical & Paramedical professionals :
Though during the Ilnd phase of the NACP, 80-90% of the health
Care functionaries would be trained in HIV/AIDS control programme.
During the 10lh plan period more focused training will be required for
the clinical management of HIV/AIDS cases as well as prophylaxis of
opportunisiib'Trtffeetton; Counseling will be the important aspect of all
such training programme. The target for 10 plan period in training will
be
(a) Refresher training of M.Os and paramedics on HIV/AIDS Control
Programme for functionaries working at Primary Health Care Level.
(b) Training of specialist, on clinical management of HIV/AIDS cases
working at secondary and tertiary level of health care in Public sector
hospitals.
(c) Training of private practitioners by involving Indian Medical
Association and other professional bodies by linking them with the
secondary and tertiary health care institutions.
(d) Training of NGOs involved in community care of PLWHA
(e) Training ofiijchqrgp of blood banks and technical staff in blood safety
programme.
(vi) Mainstreaming HIV/AIDS and other reproductive health problems in the
curriculum of Medical and paramedical education:
The research and development efforts in the field of HIV/AIDS
have been very limited in the country. Government recognizes the need
to encourage and support research and development in the following
areas:
(i)
(ii)
(iii)
(iv)
(v)
(vi)
The Govt, will look out for collaborative research with scientific
groups in developed countries for development of vaccines suitable
for the strains of HIV prevalent in India.
Molecular"'Tnology of HIV-1 and HIV-2 in the country and
sequencing of sub-types prevalent in various parts of the country.
Cohort studies to track the natural history of disease
Development of indigenous diagnostic kits for HIV
Screening of ISM&H drugs having anti-retroviral or immunomodulator properties and conduction of clinical trials.
Impact of HIV/AIDS epidemic at Family and Community level
i
V- V
Socio-cultural determinants of risk behaviour vulnerable to HIV
infection
(viii) Impact of targeted interventions on behaviour change of sub
population groups practicing risk behaviour.
(ix) Community based cross sectional studies to validate the surveillance
data on HIV/AIDS and STDs.
Cost effectiveness of voluntary counseling testing
(X)
Role of “Substitution Therapy” in reduction of HIV infection among
(Xi)
injecting drug users
(xii) Role of Family and community in care and support to PLWHA.
(vii)
Mechanism^***)of
Involvement
of
NGOs/Private
sector/Community/Local self Govt, in implementation and monitoring of
the programme :
NGOs are being involved as critical partners in some of the most
important thrust areas of NACP. NGOs are being specifically involved in
the following activities:
(>)
(ii)
(iii)
(iv)
(v)
Targeted interventions for the high risk and marginalized
populations such as sex workers, intravenous drug users, truckers,
migrant labour and street children
School AIDS Education
Lbw cost community care for people living with HIV/AIDS
Counselmg^*^**Family Health Awareness Campaign
The process of selecting NGOs and for funding has been
decentralized to the respective State AIDS Control Societies. Detailed
guidelines have been formulated to ensure that selection of the NGOs is
undertaken in rigorous and only NGOs with a credible track record are
selected. There is inbuilt system of monitoring and evaluation with each
project proposal. The planned evaluation methodology should be
consistent with the objectives of the project. The evaluation will be
carried out by an independent agency identified by the State AIDS
Control Society.
Private^sectojJs involved in programme implementation through
the organizations like CII/FICCI and ASSOCHAM, Transporter’s
Association of India etc.
I
«
(iii)
(iv)
(v)
Low cost community care for people living with HIV/AIDS
Counseling
Family Health-Awareness Campaign
The process of selecting NGOs and for funding has been
decentralized to the respective State AIDS Control Societies. Detailed
guidelines have been formulated to ensure that selection of the NGOs
is undertaken in rigorous and only NGOs with a credible track record
are selected. There is inbuilt system of monitoring and evaluation with
each project proposal. The planned evaluation methodology should be
consistent with the objectives of the project. The evaluation will be
carried out by an independent agency identified by the State AIDS
Control Society.
Involvement of Private Sector
Private sector is involved in programme implementation
through the organizations like CII/FICCI and ASSOCHAM,
Transporter’s Association of India etc.
Involvement of the community is being sought at present in awareness
generation programme like Family Health Awareness Campaign and
Community based care of PL WHA.
During the 10th plan period, there would be more of
involvement of NGOs in the thrust areas of the programmes as stated
above. Govt, would make efforts to formalize constitution of business
coalition to promote the participation of business houses in
programme implementation, particularly in workplace intervention
projects and caiT'& support component of the programme through
their own health care network.
Collaboration with Bilateral agencies
In the past years several Bilateral Agencies have shown their
interest and have come up with encouraging support to the national
effort for prevention and control of HIV/AIDS in India. Following
are some of the major contributions made by such agencies.
Estimated Burden—NACP-path mccnaksluZ'
20
A
USAID (APAC & AVERT Projects)
The USAID funded $ 10 million AIDS Prevention And Control
Project (APAC), currently operational in Chennai region, is being
implemented through a Tripartite Agreement between the
Government of India represented by NACO, United States Agency for
International development (USAID), and Voluntary Health Services
(VHS), Chennai. The Project has received wider acclaims from the
AIDS - watehers^both at National level as well as Internationally.
APAC Project, because of its excellence through technical
expertise, has become a model for the ongoing as well as upcoming
projects in the country, in the arena of creating awareness about
HIV/AIDS, and, taking preventive control measures on the spread of
this dreaded epidemic. This success has been achieved through
concerted efforts from a number of associated NGOs actively
involved in the process.
The AVERT project in Maharashtra is being implemented with
the objective to increase the use of effective and sustainable response
to reduce transmission of STD/HIV/AIDS and related infectious
diseases. An amount of USS 41.5 million is earmarked for this project.
The components are as follows-
)
Interventions in sex industry;
Improving Comprehensive health services;
Reduce High Risk Behaviour in Priority Population
Developing capacity of CBO/NGOs to respond
effectively to STD/HIV/AIDS;
(e)
To develop Communications Support Programmes;
(f)
Improve Access, Availability and Use of Condoms in
the Sex industry;
(g)
Establish Pilot/Demonstration Programmes for Outof-School Youth;
(h) To strengthen State and Municipal Capacity for
Planning, Monitoring, and Evaluation of State and
'*^Mmicipal Programme;
(i)
To develop longer term HIV/AIDS Programme of
Research and Surveillance in Maharashtra.
(a)
(b)
(c)
(d)
Department for International Development, UK. (DFID)
The DFID is providing financial support of 1.5 million pound
sterling for the "Sexual Health Project" in West Bengal, over a period
Estimated Burden—NACP-path meenakshi/
21
<
4/X
of three years. The focus of activity of this project is to control
programmes and develop appropriate IEC techniques.
DFlD^-is also supporting a national Healthy Highways
Project with the truckers’ community. The project is currently in the
pilot phase.
In addition DFID has formulated "partnership for sexual
health" project for the prevention and control of
HIV/AIDS and other sexually transmitted infections in
Andhra Pradesh, Gujarat, Kerala and Orissa for a five
years at a total cost of Rs. 154 crores (Pound Sterling28.08 million) of which Rs. 104.01 cores will be routed
through GOI as local cost and Rs. 50.29 cores as
technical cost to be managed by DFID for this project.
The objectives of the project are
•To ensure better sexual health for people vulnerable to sexually
transmitted disease (STDs) and HIV, especially the poor; and
•Lowering incidence of sexually transmitted infections in the
four States.
'
I *
k
The components of the project are:
• Surveillance;
•Condom programme by ensuring that quality condoms are
made available and acceptable to those who are vulnerable to
STD/HIV through effective social marketing;
•To strengthen service providers to respond to clients' needs
through targeted interventions for vulnerable groups, such as
sex workers and their clients, men having sex with men,
migrant workers and prison inmates.
The Europe^union is supporting ’the Lawyers' Collective ' for
a project that aims at advocacy and law and to protect the human
rights of those infected and/or affected by the AIDS epidemic.
Other Agencies
Norwegian Agency for Development (NORAD): In addition to
extensive support to NGO activities through out the country, NORAD
has contributed financial support for HIV/AIDS intervention
programme in the red light areas of Calcutta, West Bengal being
coordinated by All India Institute of Hygiene and Public Heath. A
contribution of US $ 500,000 has also been received for conducting
Estimated Burden—NACP-path meCTal^shi/^
22
I
I
National Physiciw&-Training programme, conducted by Christian
Medical Association of India in collaboration with NACO and WHO.
f
AUSAIDS and CIDA are other potential partners in HIV/AIDS
prevention and control in India.
Canadian International Developmental Agency
CIDA is providing financial and technical support in the
States of Rajasthan and Karnataka for some of the components of the
programme. An assistance of Canadian $ 12 million has been offered.
During the 10th plan period these agencies will be persuaded to
continue their support.
Improve communicable disease control
Primary/Secondary health care information:
services
through
the
National AIDS Control Programme has no infrastructure at
district level and below. The programme components are being
implemented through the existing primary, secondary and tertiary
health care infrastructure. The most important intervention in this
context has been the involvement of tertiary care in training and
capacity building and their linkage with that of secondary care health
institutions. The existing health care infrastructure is also being
utilized for management of HIV/AIDS cases.
During the 10th plan period, health care institutions will be
strengthened by-sproviding them equipment needed for diagnosis and
care and support to people living with HIV/AIDS. The infrastructural
supports in terms of their capacity to accommodate more HIV/AIDS
patients will also be required, particularly, in States with High HIV
prevalence. Budgetary support would be required by institutions at
secondary and tertiary care level for this purpose. Health care
functionaries would be utilized for dissemination of information.
Current Pattern of Monitoring and Evaluation:
For the effective monitoring evaluation to assess the
implementation of phase II of the National AIDS Control Programme,
the following mechanism is being adopted:
Estimated Burden—NACP-path meenakshi/
23
1
I
Jt
(i)
(ii)
(iii)
Conduction of base line, mid term and end valuation of the
programme components by an independent agency
Conducting the Annual Performance and Expenditure Review and
Conducting the National Performance Review (NPR) under the
National AIDS Control Board.
The same mechanism of monitoring and evaluation will
continue during the 10th plan period as well.
Current Status of HMIS, disease surveillance, its quality and
utilization/proposed:
For the. monitoring of the programme, a comprehensive
computerized HMIS system is in use, which include a set monthly and
quarterly progress reports on different components of the programme.
During the 10th plan period, the same computerized HMIS formats for
monthly and quarterly progress report would be used.
J-:--- r-tv-xr-'?'"
Estimated Burden—NACP-path meenakshi/
24
■f
I
TUBERCULOSIS
Persisfenf killer
I B is a social and environmental disease to <
' '
control which technical packages
-’1
J
o
alone are insufficient, says Thelma
Na.
arayan.
»•
a
I
i
r
f
i
!
[
I
/
I
I
r
(
*5
(IB), an ancient infectious
I disease,^ was known as the dreaded “white
plague,
pthisis or “consumption” in the
English terminology’ of the last century and as
Rajyaroga (king of diseases) in India.
Fifty years after Independence it is still India’s
biggest public health problem, claiming an estimat
ed 500,000 lives every year and causing disease in
about 17 million. Affecting adults in the prime of
their lives, it causes much suffering and economic
loss for patients, their families and the nation. It
results in 26 per cent of preventable adult deaths and
is one of the most important causes of death among
women of child-bearing age. What is tragic is that
the disease is curable.
Causah’ve agent
Though its infectious nature was recognised ear
lier, it was the German scientist Robert Koch, who
in 1882 isolated the bacteria Mycobacterium
Tuberculosis as the causative agent of tuberculosis,
fhe work of several researchers suggests that the
tubercle bacillus has co-existed with lfuman beings
since early times. Like othet organisms in the process
of evolution it has developed a complex ecological
relationship. This is indicated
presence of
indicated by
by the
the presence
of varvar-
Jous_strams of the organism with differing virulence.
The south Indian strain, for instance, is less virulent
than the British strain.
Animals harbour different species of mycobacte
ria. From the human disease point of view, cattle are
the most important. In Europe, Mycobacterium Boch
was a fairly important source of infection. It spreads
from cattle t^ people through drinking raw, unpas
Si
y‘\.'
te
teurised or unboiled milk. A major control pro
gramme, involving the detection and slaughter of
infected tuberculin positive animals, was successful
ly conducted.
Studies in Calcutta and Bombay in the Thirties
and Forties found that M. Bovis was not a cause of
human TB in those areas and was relatively uncom
mon in cattle. Studies in Madras around the same
period indicated the presence ofTB infection amo^
cattle herds.
- --More
recently
in
1995,
more recently m 199?, at a meeting in Madras,
JtS Br.
.ife ;
veterinary scientists and TB specialists expressed
concern about the occurrence ofTB in cattle. Cattle
pnmn' iiy suffer from pulmonary TB_(affecting the
lungs) with infection almost invariably progressing
to disease. It spreads through the air borne route'
The primary human disease due to M. Bovis is usually non-pulmonary.
i
r
■
! .
• ;
f'.'i
---.It
'
/
f
®
Id?
!
{Jnlaren at a 713 hospital tn Chennai
i
j
THE.i'2.'J2U- SURVEY OF THE ENVIRONMENT ’97
21'
r
'■.................................................
• I
'
There are a wide range of species of saprophytic
atypical environmental mycobacteria. While not
usually associated with human disease, more recent
ly, infections caused by M. Avium Intracellulare have
occurred in patients with AIDS.
J
Reasons for fhe spread
J
I
iI
Tuberculosis was probably only a disease striking
animals many years ago. Two important periods in
the epidemic spread of the disease among major
human population groups appear to have been:
(a) The period of transition to agriculture and
cattle domestication followed by population increas
es during the sixth and seventh millennia BC in the
eastern Mediterranean region and Europe. Some of
the earliest written records and ancient Indian med
ical treatises arc said to describe the disease in early
India. Tuberculosis probably existed among people
at low levels of endernicity. Most often, that is, small
proportions of the population suffered from the dis
ease at any time.
(b) The period of industralisation starting in
substantially in West Europe and the U.S. This wa; '
unrelated to medical intervention and before the dis
covery of anti-TB drugs, which occurred from' the
late Forties onwards.
It has been suggested that population incrca.se
and European migration, colonialism and war initi
ated “epidemic waves” in different regions of tk
world in the last century. In India, people faced
heavier taxes, landlessness increased and textile and
cottage industries were affected. The process of
impoverishment set the scene for repeated famine,
an increase in TB and other epidemics.
During the First World War (1914-1918), death ;
rates due to TB increased in all countries at war. The decline ofTB in the U.K. was halted for 10 yean”
from the start of the Second World War in 1939.
Recent studies show higher rates of TB during war.
conflict and among refugees.
India had a big problem of TB among the post--*’/
Partition refugees in 1947. Disrupted social condi-.
tions, undernutrition, poor housing and physical
and emotional stress are pre-disposing factors.
The association of socio-economic factors with TB was forgotten in the
“scientific optimism” generated by the discovery of chemotherapeutic drugs.
i
i
i
■
1
i
<
V- •
I. .
England and Europe in the 16th century. This
resulted in a process of urbanisation with over
crowded, unhygienic living conditions for the work
ing class in the new industrial and mining-towns.
People were paid low wages and had long hours of
hard working in appalling conditions. Research
indicated that this process was repeated in the U.S.
and Africa and that industrial and urban growth
were correlated with TB. There has been the obser
vation that TB was “perhaps the first penalty that
capitalistic society had to pay for the ruthless
exploitation of labour.” The environmental condi
tions during that .time were closely associated with
the poverty of people though the elites were-reaping
the benefits of empire and industry.
Observations in different countries, during the
past 150 years, have shown these conditions to form
the essential substrate in which TB and other infec. tious diseases thrive.
With improved housing,’working conditions and
nutrition of large proportions of their populations,
disease and death rates from TB started declining
Tibetan resettlements in India also report a high;.
prevalence ofTB.
The disease thus seems closely linked to the .
social history of humankind mediated through its;’
effect on the environment and on living standards. ■ The association of socio-economic and environ
mental factors with TB has been highlighted inter-nationally by researchers. It was explicitly recognised
in India at policy flaking levels by the Bhore
Committee (1946) and the Mudaliar Committee
(1961). It later was forgotten in the “scientific opti
mism” generated by the discovery of effective
chemotherapeutic drugs. These discoveries did accel
erate the tailend of the decline of the disease, in
countries that could afford and organise good'
nation-wide health services.
Physical environment
f
a\_HousingL Tuberculosis-primarily affects the
lung but can also spread within ^the body affecting
several other organs. It is an air-borne disease, trans
mitted to others by coughing, spitting and talking.
THE HINDU SURVEY OF THE ENVPR'CTTmENT '97
■
t
V
x
»
i
'■':-M
II
'j
11
’
fSWiOtl
a®
!
;
i -i
■i
’ t cause TB.
Poor working conditions
I
Only patients whose phlegm or sputum contain bac
teria (termed sputum positive) are sources of infec
> i
■
l{
i
11
1 1
Ia >.’ '
.1
-
■?
44 >
I k
1
|
‘
i
r
-
?
I
tion for other people.
The bacteria disperse in the atmosphere through
small droplets which survive for long periods in
dark, unventilated conditions. Sunshine, present
abundantly in India, effectively kills the bacteria.
Small tenements and houses with insufficient or
no ventilation provide the right environment for the
survival jmd spread of the organism. Poor, congested
housing, a result and indicator of poverty, is an
important element in disease transmission.
The National Sample Survey, a large multi-State
cross-sectional Indian study conducted in 1955-58,
found that people living in kutcha houses had a
higher prevalence of TB. Twenty to twenty five per
cent of patients with TB give a history of someone in
the family also having had TB. Government and
development groups working on housing schemes
and activist groups campaigning towards housing
rights thus indirectly contribute to TB control.
Smoky chulhas or cooking fires are shown by
studies and experience in India to increase the risk of
respirator}' problems. They could aggravate the dis
ease and make the life ofaTB patient more difficult
and perhaps be a risk factor in the spread of 1 B.
Similarly patients with TB are advised not to smoke
to prevent further lung damage. Promotion of
smokeless chulhas by development agencies, could
also be part of the anti-TB package.
b) Urbanisation'. There were attempts in the
Revised National Tuberculosis~Coiitrol Programme
•«
V;
1:
K
|i
11
I?'
I
in India in 1993 to give greater importance to urban
areas in the programme. The urban population has
increased over three decades from 20 per cent to 26
per cent in 1991, with some inter-State variations.
Before the Fifties, TB was considered to be largely an
urban problem, as it was in Europe. The National
Sample Survey findings, however, showed that it was
equally prevalent in urban and rural areas. Since 80
per cent of the population was then rural, the prob
lem was understood as being predominantly rural.
The National Tuberculosis Programme was the
first health programme to emphasise the need to
develop general health services in rural areas with
which TB services were to be integrated. Health ser
vices and health personnel were then predominantly
urban based. Thirtyfive years later, 74 per cent of the
population is still rural. Though some gains have,
beeirmade, there-continue to be urban-rural differ
entials in health budgets and health-care services.
I
I
fe
r;
h
:V
?-* ’
I
Vulnerable groups
The urban poor living in slums, jhuggi colonies
and shanty towns comprise about 60 per cent of the
urban population. The most deprived are homeless
migrants, rag pickers and street children. These vul
nerable groups are at greater risk of TB and.need
specific intervention, protection and social security.
The urban habitat has deteriorated greatly over
the years with basic infrastructure unable to cope
with the growing demands for housing, water, sani
tation, roads and transport. The growing presence of
small unregulated factories close to domestic-sites-or
THE HINDU SURVEY OF THE ENVIRONMENT '97
f
I
p
f
Z3_
»
S. t-.’ —t-. i-zr'‘
f
A TREE S NOT A RAW
FACE OF RUPEE
rr533
II
V
ii
iW
z
i
I
h
or
i
a
A"
i
0
i
1
’SaSorVERC^
aHO“'
------------------------------------------------------- .
1
For more details on pollution control contact “A
RO ARD
!
KARNATAKA STATE POLLUTION CONTROL B°AR°
1
6 7, 8 & 9th Floors, Public Utility Building,^M.G Bead Bang3 ore
Phone :: 5586782,
5586782, 5588270,
5588151 Fax
5588270, 5588151
Fax : 081- 5586321
Regional Officec :
* Mysore
• Raichur
1
• Davangere
•Mangalore
* Dharwad
* Gulbarga
AB e I a g a m
‘Hassan
?•
*
(
,1
i
1
j
I
I
I
I
(he j'jo'.vili o( shuns ncxr to large factories add (o the
prohlum. I he all-pervasive corruption prevents and
slows the development of basic services.
c) Indnstrnlisation and work environment'. The
work place provide.': an important environment
which could predispose to or facilitate the develop
ment of FB. Miners and quarry workers exposed to
various kinds of dust arc likely to develop silicosis,
and arc at greater risk of developing TB. In crowded
industrial sheds and factories, with poor ventilation,
inadequate sunshine and fresh air, if one or some
workers have untreated sputum positive TB there are
definite chances of spread of infection.
Rapid industralisation, driven by the present eco
nomic imperative, most often sacrifices safety proce
dures (dust control) and effluent treatment (air pol
lutants) which decrease profit margins. This is par
ticularly so when regulatory mechanisms are poorly
implemented and the workforce is inadequately
informed and poorly organised.
There arc several indicators that air pollution lev
els arc growing alarmingly in urban areas. Any addi-.
tional assault or damage to the respiratory system
(like smoking) could aggravate TB. The potential
occupational hazards faced by health personnel
working in close contact with TB patients has
received scant attention.
d) General causes'. In the early parts of the centu
ry there ware anti-spitting drives in several countries
such as rhe'U.K. Australia and South Africa among
oihcrs as a preventive measure against tuberculosis,
besides providing for better general environmental
hygiene and cleanliness. Given the rampant practice
of spitting in India, this would certainly be a chal
lenge.
Impact of political and economic policies
Moving beyond the physical environment co die
broader political and economic enviromments, it is
important to understand global changes that arc
occurring and their impact on nations and on peo
ple, especially the poor. Studies to understand the
immediate and short-term impact of structural
adjustment on poverty, unemployment and the envi
ronment are important for several reasons.
These programmes also have a direct relationship
with TB. Several countries across the world, for
instance in Africa and the Phillipi
k ines, have docu
mented the ill-effects of these programmes on the
health of the poor, including high TB rates.
The impact of explicit pro-privatisation policies
in State-run health services also needs careful study.
These services are used most often by the poor to
whom private care is inaccessible for long-term seri
ous illnesses likcTB.
Nationally, an important issue is the low priority
given to health in terms of Central and State
Government budgets. Equally important is the lack
I
‘B
z ■■
ii
«'■'■■
J
f
>■
H
■■IHiM
!
K:Y
■
/CSWilBBiOs
i
!
?
I
iiJilHIlCJ....
.
h/
’y
■
•
’
•••
r •
OSs
•■ W;IL....--,
r ‘
••
r
■.. • M<^<Ysei. - s
I
ssgs^aOsw;:'''cjY-r •> MS^ftetSKS.:
Congested and unhygienic housing: another cause of 1B
THE HINDU SURVEY OF THE ENVIRONMENT '97
75_
4.
■
..4.
;
*
of commitment to develop and maintain quality of
care and service in public sector institutions,
lolmcal interference, mismanagement and rampant
corruption have reached unacceptably high levels in
a v,tal area affecting the health and well-being of cit
izens. I his is an area for concern and social action.
E.'tpeclafions belied
tions. These events created a panic, raising tLw
^re of a new epidemic in which available look
treatment would not be effective. Newer dialnwj
and drugs were not available as basic research inT
had stopped three decades ago. The new cv;l
brought to mind the long period, from the late !■
to the early 20th centuries when TB was the lead
i here were great expectations that rhe National
luoc-rculosis Programme (NTP), articulated in 1962
after indigenous research, would be able to relieve
human suffering caused by TB and later reduce dis
ease transmission so that it no longer was a public
health problem.
evaluation studies in 1975 and 1988 and the
monitoring system of the NTP consistently report
gaps between expected performance and actual out
come. One study has indicated that only 8 to 16 per
cent or expected cases ofTB get complete treatment
from the public health services.
The past decade has seen a major change in the
cause of death in Europe and America, decimate
about 20 per cent of the population.
With international travel, global trade and ins
national capital being involved and potentisiaffected, the sense of urgency appeared great. IE
resulted in a new global policy environment i
which TB was once again high on the internatiom
public health agenda.
The World Health Organisation (WHO
declared TB a global emergency in 1993. There wi
increased intervention by the WHO, World Barj
and bilateral agencies in the control of TB in ‘V ’
and middle-income” countries.
Revised programme
Poor and congested housing, a
Jesuit and indicator of poverty, is an
iimportant environmental element
m disease transmission.
TB situation globally. In the mid-Eighties there was
a reversal of decline in the incidence of TB in the
b.S. This was followed between 1986-91 by an
increased number of cases up to 33 per cent.
Increased TB rates occurred in several West
Euiopean countries too. Reasons cited for the
increase in TB are association or co-infection with
HIV/AIDS, neglect of TB control programmes by
governments for over two decades and the develop
ment of Multi Drug Resistant (MDR) TB ( bacteria
chat are resistant or non-responsive to two co three
basic anti-TB drugs). A number of cases occur
among the inner city homeless, among drug users,
and among migrants and indigenous people.
Twen tv-five per cent of cases .in the U.S. were among
the “foreign born,” similar to West Europe.
■
MDR TB arises because of inadequate and
incomplete treatment for which the health services
and patients have been blamed, out of context to the
under-resourced and difficult circumstances in~
aIhcIi tiiev function and live respectively. The treat
ment or MDR TB is extremely expensive and often
has a unsuccessful outcome under the best condi7- .........
A revised National TB Control Prograniir:
(RN1CP) has been introduced in India. A
million soft loan from the World Bank has bee;
negotiated. This is conditional to using the XWO
package prescription. A key element in this is th
Direct Observation Treatment, Short Course.
Chemotherapy (DOTS) in which health worken
have to watch patients swallowing their pills three
times a week, for at least two months, out of a sir
month treatment.
There are doubts about the feasibility, sustain
ability and ethics of such an approach. Being entire
ly technical it does .nox-consider the social and. ejnvr
ronment roots of the disease. Hence it ignores sJcis!
and community dimensions within which any tech
nical package needs to be embedded. This would
include personal, social support to affected people
and their families. It would also recognise develop
ment organisations and activists as partners. xMosi
importantly, it would address the question of social
and economic inqeuality and injustice. It also
assumes that the health system will “deliver" rhe
RNTCP and does not tty to understand the reasons
why the NTP did not fulfil expectations.
1 helma Narayan is member-in-cliarge of research and
evaluation at the Society for Community Health, AwarenL,
Research and Action, Bangalore.
THE HINDU SURVEY Of THE ENVIRONMENT -9 7
J
Indian J Med Res 59, 11 November 1971 pp 1727-1736
Infection And Disease In a Group Of
South Indian Famiiies-Part XIII Skin Sensitivity
To Six Mycobacterial Antigens1
T. Jacob John J. Frimodt-Moller Roger A. Feldman2
P. Jayabal and
K. Ramanand Kamath3
Enterovirus Laboratory Christian Medical College and Hospital Vellore
Andhra^Pradenf
Madanapalle Tl,herculosis Research Unit Arogyavarani
Received for publication July 21 1971
A skin test survey using six mycobacterial antigens, namely PPD-S GA B F and
tested “th PPD S0GaaSdmtPle P“P“,ati°n °f VelIore’ Tamil Nadu- Each individual was
ested with PPD-S, G and two of the remaining four antigens. The frequency distri
Datt7mw7haCtlOn| t0 PP?’S aCCOrdin8 t0 the size of induration showed a bimodal
pattern wtth one cluster of small reactions (less than 8 mm) and another of large reac
tions (8 mm or more). The prevalence of large reactions increased with age from 2%
m pre-school ch.ldren to nearly 90% in adults of 35 years and over. Reasons to Vh'
other five antigens were also very common in the population. However, a majority of
such reactions meaured only 2-7 mm in diameter. The pattern of sensitivities to the
6 antigens suggests that M. tuberculosis infection is common in this community and that
reactions to the antigens from mycobacteria other than M. tuberculosis are to a large
extent cross-rcact.ons ansmg from tuberculin sensitivity. At the same time there were
a few mdmduals who had large reactions to one or more of these 5 antigens in the
,h“
Introduction
b
The‘uberculln skil1 test
been widely used in epidemiological studies of infection
by Mycobacterium tuberculosis. After the discovery of other mycobacteria that in
man and at times cause disease similar to tuberculosis, skin test antigens have been
tnd USSd iD epidemioI°Sical studies (Edwards et al 1968 Smith
and five ehCO
ted
Using antlgens de™ed from M SercXs
and five other species of mycobacteria on a sample population of Vellore Tamil
u, m an attempt to study the prevalence of infection by these organisms. The
N°- 01-326-01
the Center for
■ Present address : Chief Special Pathogens Secti,
ion Bacterial Diseases Programme Center for Disease
Control Atlanta Georgia USA
* Present address : Lecturer in Paediatrics
University of Malaya Kuala lumpur Malayasia
1
37
2
Infection and Disease in South Indian Families
methods of study, the overall results of the rate and range of reactions and an evaluation
of the role of cross-reactions are reported in this paper.
Material and Methods
The sample population : A continuing longitudinal study of infection and disease
in a group of families from three localities in Vellore town began in September, 1965.
The methods of selection and recruitment to the study and a description of the sample
population have previously been reported in detail (Feldman et al 1969). The sample
was representative of the population from which it was drawn in several demographic
characteristics. At the time of the skin test survey in January 1967 there were 90
families with 420 members under surveillance. Infants under 3 months of age were
not tested.
The antigens : Purified protein derivatives (PPD) from six species of mycobacteria
presented in Table I were used for skin tests. They were obtained from the Tuber
culosis Research Laboratory, Chamblee, Georgia, U.S.A., stored refrigerated and
transported and kept at the work spots in chipped ice. PPD-S, prepared from
M. tuberculosis was given in doses of 5 tuberculin units (TU), in 0 1 ml fluid. The other
antigens had equal protein nitrogen content as PPD-S and were also administered in
0-1 ml volume.
Table I. Six mycobacterial antigens used in the study
Antigen
Prepared from
1
!
PPD-S
PPD-G
PPD-A
PPD-B
PPD-F
PPD-Y
Mycobacterium tuberculosis
(prepared originally by Siebert)
Scotochromogen, Gause strain
Mycobacterium avium
Non-photochromogen, Battey bacilli
Rapid grower, Mycobacterium fortuitum
Photochromogen (yellow)
Mycobacterium kansasii
The tests : All tests were performed by a nurse who was unaware of the identity of
antigens which were coded along with similarly coded syringes. Each person was
given 4 antigens intradermally at proximal and distal sites of volar surface of the
forearms. PPD-S and PPD-G were given to everyone along with two of the remaining
4 antigens. Thus, there were six combinations of antigens (Table II). The first six z
persons tested on any day were given these combinations in a serial order and this cycle
repeated thereafter. Thus, the population was equally allocated to the six combinations.
The sites of injection were so rotated that each antigen had an equal chance of being
given at any of the four sites. Tests were given in morning and afternoon sessions on
Monday, Tuesday and Wednesday and the reactions read on Thursday, Friday and
T. Jacob John et al
3
Saturday. The horizontal and vertical diameters of induration were measured by the
same nurse and dictated to a recording clerk.
Table IL
Six combinations of PPD antigens used
Combination of antigens
PPD
PPD
PPD
PPD
PPD
PPD
S,
S,
S,
S,
S,
S,
G,
G,
G,
G,
G.
G,
A
A
A
B
B
Y
and
and
and
and
and
and
B
Y
F
Y
F
F
The greater diameter of induration was used for purposes of analysis. The
results on each individual were punched on an 80 column punch-card and all data
analysed with the help of a sorter and a tabulator.
Results
i
i
I
1
A total of 350 persons were tested and 347 returned to be read. They consisted
of 102 infants and pre-school children, 97 school-age children, 35 adolescents and
young adults and 113 adults of over 24 years (Table III). All persons were tested with
PPD-S and G, 174 were tested with PPD-A, 176 with PPD-B, 171 with PPD-F and 173
with PPD-Y.
i
Table HI.
I
Age distribution of persons tested and read
Age (years)
No. of persons
0— 4
5—14
15—24
25—34
35—44
>45
102
97
35
70
22
21
347~
All ages
i
i
Among the 347 persons tested 172 (50%) had no reaction to PPD-S and 175 (50%)
reacted with indurations measuring 2 to 30 mm. The distribution of those who reacted
according to the size of induration is shown in Graph 1. The distribution is clearly
bimodal and two visual-fit curves have been drawn to separate the clusters one of
small reactions and the other of large reactions, the point of separation being at 8 ram.
Reactions of 8 mm or more were recorded in 130 persons (38 %).
i
Measurable reactions of 2 mm and over occurred to PPD-G in 68 %, to PPD-A
in 63%, to PPD-B in 69%, to PPD-F in 28% and to PPD-Y in 53 % of those
• tested (Table IV). The frequency distributions of rections to these five antigens
!
I
Infection and Disease in South Indian Families
4
according to the diameter of induration are shown in Graph 2. Although all distribution
curves show a trough at 3-4 mm level, none of them is as clearly bimodal as that
of PPD-S.
Graph 1
The frequency distribution of reactors to PPD-S according to the size of induration
12
Cfl
10
©
Q
eJ
o
*0
I
6
&0
«
5CJ
4
<u
0-
2
0
4
2
6
8
10
12
16
14
18
20
22
24
26
28
30
Diameter of induration (mm)
Graph J2
The frequency distribution of reactors to PPD-F, A, G, B and Y according to size of induration]
50
46
0
UJ
< i6
PPO P
20
Z
16
12
PPD B
1
4
20
UJ
B '2
a
UJ
8
4
4
18 20 22
INDURATION ( m m |
INDURATION ( mm )
(
4^
■'J
■
J
Table IV.
A comparison of the frequency of reactions to 5 PPD antigens with PPD-S
Tested with
PPD-G & PPD-S
PPD-A & PPD-S
PPD-B & PPD-S
PPD-F & PPD-S
PPD-Y & PPD-S
Reactions to
PPD-G
PPD-S
PPD-A
PPD-S
PPD-B
PPD-S
PPD-F
PPD-S
PPD-Y
PPD-S
Induration
size
Ch
P
0
mm
112 (32) ♦ 172 (50)
65 (37)
82 (47)
55 Gl)
81 (46)
124 (73)
92 (54)
82 (47)
89 (51)
2
mm or more
235 (68)
175 (50)
109 (63)
92 (53)
121 (69)
95 (54)
47 (27)
79 (46)
91 (53)
84 (49)
4
mm or more
188 (54)
147 (42)
78 (45)
74 (43)
85 (48)
82 (47)
19(11)
66 (39)
68 (39)
72 (42)
8
mm or more
118 (34)
130 (38)
40 (23)
63 (36)
48 (27)
73 (42)
1 (1)
58 (34)
45 (26)
66 (38)
Number tested
•Percentage shown in parentheses
347
H
174
n
cr
Ch
O
EF
H
<6
176
171
173
I
6
Infection and Disease in South Indian Families
r<
S
When all reactions, irrespective of their size are considered they occurred more
frequently in response to PPD-G, A, B and Y than to PPD-S. However, reactions of
8 mm or more were more frequent to PPD-S than to others in most age groups (Graph 3).
Such reactions to PPD-S occurred in 2 % of children below 5, 17% of children of 5 to
14 years, 57% of those 15 to 24 years, 79% of those 25 to 34 years and 86% of those
35 to 44 years and 45 and avove.
t
Graph 3
The percentage of(reactors with indurations of 8 mm or more to the 6 PPD antigens in
different age groups
100]
90
)
SO
TO
r
60
k-
“50
i <'•>
CE
UJ
a- 40
r
30
20
INOUfUTlM »•»*
1
------e— —e—
0-4
5-14
15-24
25-34
35’44
>45
Age groups
Discussion
It is customary in clinical practice to classify tuberculin reactions as negative and
positive according to the size of induration. Usually, an arbitrary value such as 6, 8 or
10 mm is taken as the dividing line. It is assumed that reactions of smaller dimensions
are non-specific being caused by infection with mycobacteria other than M. tuberculosis.
In certain communities the results of skin test surveys may demonstrate a logical level
for such division. In the sample studied, approximately 50 % had no reaction. Those
who reacted fell in two groups, one with small reactions and the other with large reac
tions. The point of separation was at 8 mm (Graph 1). Therefore, indurations of 8 mm
and more were considered large reactions or positive in this sample. The frequency
!»
I
I
')
i
j —
I :*•-
T. Jacob John et al
jA
M
1
-<•5
"<4
t
■V
i .3
I
I
7
of such large reactions was low in pre-school children (2%) but rose steadily with in
creasing age until nearly 90% were positive after 35 years of age.
Recently, Wijsmuller et al (1968) reported on the pattern of tuberculin sensitivity
in the villages’of Mysore State. A bimodal pattern with the breaking point around
12-14 mm of induration was observed only among females, in response to 5 TU of
PPD-S. Reactions of 12-14 mm and over occurred in about 20% of the population
studied. They were found in 2% of pre-school children, 30% of those 35 to 40 years
and 40% of those over 55 years.
Frimodt-Moller (1966) studied the pattern of tuberculin sensitivity in six villages
around Madanapalle, Andhra Pradesh. There was no bimodal pattern. Among 402
persons tested, 100 (25%) had no reactions, 214(53%) had 1-7 mm of induration and
only 88 (22 %) had 8 mm or more of induration. Reactions of 8 mm or more were
seen in 5 to 10% of children at the age of 10 years, 20% of young adults of 20 and
40-50% of older adults.
These three areas with different prevalences of tuberculin sensitivity are located
within 200 miles from each other. It appears that infection by M. tuberculosis is more
common in Vellore town than in the rural areas of Mysore and Andhra Pradesh.
Mycobacteria other than M. tuberculosis have been classified mainly into four
groups namely scotochromogens, photochromogens, non-photochromogens and rapid
growers (Runyon 1959). The avian tuberculosis bacillus (M. avium) is closely related
to the non-photochromogen group (Wayne 1966). All four groups were represented
by skin test antigens in this study. Though a majority of individuals reacted to these
antigens, it appears that such reactions are, to a large extent, cross-reactions arising
from tuberculin sensitivity. There are three main reasons for this conclusion, as dis
cussed below.
Among guinea-pigs infected with M. tuberculosis and no other mycobacteria, a
significant proportion react to PPD-G, A, B, F and Y with indurations ranging from
5 to 15 mm (Edwards et al 1968). The same phenomenon occurs in Eskimos of Alaska
who are believed to harbour only M. tuberculosis and no other mycobacteria of the kind
considered here (Edwards et al 1968). Therefore, the demonstration of skin reactions
alone does not constitute evidence for the prevalence of mycobacteria other than
M. tuberculosis in the community.
A comparison of the PPD sensitivity profile of the Vellore population with that of
Eskimos in Alaska and of the village population near Madanapalle is shown in Graph 4
(Edwards et al 1968, Frimodt-MoUer 1966). The profile is a line joining the mean
indurations of reactions to PPD-S, Y, G, A, B and F in that order (Edwards et al 1965).
It is obvious that the vellore profile resembles more closely that of Eskimos than that
of the villagers near Madanapalle. The Eskimos profile is known to resemble that of
tuberculous guinea-pigs and of tuberculous patients (Edwards et al 1968). The
Madanapalle profile resembles that of residents of Georgia (USA) in whom infection
by scotochromogens is known to be common and also that of guinea-pigs infected
only with scotochromogens (Frimodt-Moller 1966).
I
8
Infection and Disease in South Indian Families
Graph 4
The PPD sensitivity profile
14
I
\\
12 -
I
1 0-
1
4
•o
•i
I
--------- VELLORE
--------- ALASKA
----------- MADANAPALL E
6
8 •
6 -
(
21
ol
(
S
Y
G
A
B
F
PPD antigens
The sample population can be divided into two groups, one with no or small
reactions to PPD-S (or PPD-S negative) and the other with large reactions (or PPD-S
positive). If the reactions to PPD-G, A, B and Y are studied in these two groups,
it will be seen that in the former group are the majority of individuals with no or small
reactions to these antigens and in the later group are the majority of individuals
with larger reactions (Graph 5). This pattern suggests that the tuberculin sensitivity
I
L
Graph
A comparison of reactions to PPD-G, A, B, ¥ and F between PPD-S negative and
PPD-S positive individuals
REACTION TO
PPD-
PPD-S ’NEGATIVE’
A
ppo-s ’positive ‘
l
G
i
i (
E
B
F
F
EH
80
E
A
r i
too
1
c
H
60
40
0 mm
20
J
■» 7. 0 0
2-7
mm
20
40
60
80
100
F
I
I
E33 8-23 mm
C
I
P
I
I
T. Jacob John et al
9
influences to a large extent the reaction to the other antigens, Reaction to PPD-F
alone differs from this pattern.
Though M. tuberculosis has thus been shown to be the predominant mycobacte
rium in Vellore population, the results of our study also show that it is not the only
agent causing sensitivity to various PPD antigens in this community. In a population
exposed only to M. tuberculosis as in the case of Eskimos or that of infected
guinea-pigs, the frequency distribution of the size of reactions to PPD-S forms a
bell-shaped curve. In them nearly all reactions are large. The frequency distribu
tion in our study shows a bimodal pattern with a cluster of small reactions ranging
from 2 to 7 mm. It is now well recognized that this pattern of tuberculin sensitivity
indicates the prevalence of some non-tuberculous mycobacteria in the population
(Palmer 1953. Edwards et al 1968). It is believed that these small reactions are the
result of cross-sensitivity to antigens of some other mycobacteria.
I
I
1
/:
1
s
/
In 63 individuals the reaction to at least one other PPD was greater by 4 mm or
more than that of PPD-S. The difference was 6 mm or more in 32 individuals and 8 mm
or more in 11 individuals. Though in most persons with no reaction to PPD-S there was
little or no reaction to the other antigens there were 39 persons with 0 reaction to PPD-S
and 4 mm or more reaction to at least one of the other antigens. Among these 39,
20 had reactions of 6 mm or more and 8 had reactions of 8 mm or more. These results
also indicate the prevalence of infection by some other mycobacteria in addition to
M. tuberculosis. Since the numbers and sizes of reaction to PPD-G were greater than
to any other PPD in this group it is likely that scotochromogens are present and
do contribute to this pattern of PPD sensitivity.
Acknowledgment
We are grateful to Miss Leela Thomas who performed the skin tests and to the
Vellore families for their participation in this study.
References
Edwards, L.B. and Smith, D.T. 1965. Community-wide tuberculin testing study in Pamlico Coun
ty, North Carolina. Am Rev Resp Dis 92, 43-54.
Edwards, L.B. Comstock, G.W. and Palmer, C.E. 1968. Contributions of nonhem populations to
the understanding of tuberculin sensitivity. Arch Environ Health 17, 507-516.
Feldman, R.A. Kamath, K.R. Sundar Rao, P.S.S. and Webb, J.K.G. 1969. Infection and disease
in a group of south Indian families. I. Introduction, methods, definitions and general observa
tions in a continuing study. Am J Epidem 89, 364-374.
Frimodt-Moller, J. 1966. Comparative mantoux tests with standard tuberculin and purified protein
derivatives obtained from mycobacteria different from M. Tuberculosis. Proceedings of the
21st Tuberculosis and Chest Diseases Workers’ Conference, Calcutta, 79-90.
Palmer, C.E. 1953. Tuberculin sensitivity and contact with tuberculosis : Funher evidence of non
specific sensitivity. Am Rev Tuber 68, 678-694.
Runyon, E.H. 1959. Anonymous mycobacteria in pulmonary disease. Med Clin N Am 43, 273290.
1
10
gft
Infection and Disease in South Indian Families
1967. Diagnostic and prognostic significance of the quantitative tuberculin tests. Ann
Smith, D.T.
Wa^to^fif’e^cation and identification of mycobacteria. HI. Species within group.
HI. ' Am R- Resp Dis 93, 919-928.
Wijsmuller, G. Raj Na5a;nv.Ma^rna'h’^es^Dis 97 «9-443
sensitivity in South India. Am Rev Resp Dis
,
Qf
t
!•
1
-•s.
. ; '1
iI
A'
r> -r
Could TuberculosiE be a Psvcho-eomatic disease ?
Robert Koch s discovery of M.tuberculosis in 1882 seemed to p r z . e
nature,
that the disease was infectious in nature.
The introduction of
skin
testing
in 1908 however showed that many more persons were
i n f e-z z ed
with
the bacillus than acutally had the disease,
disease.
In
India
it
is
generally held that one-third of all people are
infected with
~B
(primary tuberculosis?
and that this constitutes disease
immun1z >.
However a small proportion of people reactivate
the
disease
in
adulthood (secondary tuberculosis) for as yet unknown reasons,
How - 5
it
then that,
urbanisation, poor housing,
overcrowdina,
poverty,
exposure to occupational dusts and a variety of other
factors,
predispose to a reactivation of tuberculosis.
A
large amount of TB research today is focusses on
lasErure
the
response,
why it is that only some people develop
tuberculosis.
wny
tuberculosis,
some people respond differently to treatment.
An area of
research
which is ignored in current literature is psycho-somatic research
■- n
Tuberculosis,
Rene Dubos, and other early TB researchers
□ i scussed
that the
issue of stress was closely related
to the
concept
zf
resistance to tuberculosis.
But with the focus on
the
□ erm
=.z-d
therapeutics. this ideas has not been pursued.
Thomas Holmes was a pioneer in psychosomatic research credized
for showing the relationship between stressful life events ana disease
in general, and his work became the cornerstone of modern
‘mind-bocy'
research.
He started his work on psycho-somatic
research as
a
physician
in a United States TB sanatorium in
1949 exploring t.-.e
relationship between stress and tuberculosis,
Although he lacked
t-e
sophisticat ion of modern epidemiological technioues. several of
his
studies showed that persons who had experienced stressful
situations
such as divorce, death of spouse, a loss of a job were more likely
develop tuberculosis and less likely to recover from it.
He devised
numeric scales that quantified stressful events and did prospective
studies with control -groups tsome
(some of his work is
summarised below .
Although Holme’s work was rudimentary,
and open
to
scientific
criticism,
his basic contention may have been correct,
Holmes was
well regarded among his colleagues and his work received governmenz a1
funding.
necessari1y
His scientific
findings however were not
accepted.
This may have been because at that time the focus was :n
the
'germ‘ and that TB treatment was believed to be causing a major
decline in
the disease.
He
left TB research
in
19b2 and r.i =
subseauent work was all published in psychosomatic journals.
Holmes also emphasized the need
to understand - each □ at 1 E“: t - - holistically.
The following is a discussion of a 29 year old b 1 az k
woman at a case presentation:
At
this point in her life, the man she later married
rsturr.a*d
from war service to the small town in Louisiana, and after a month
superficial
acquainta’ncesh ip
they were married'..
The
mar11 a 1
ad justment was always poor
“nor
a
. .. She spoke of herself
herse1f
as
2
hotblooded woman"
who preferred church activities.
Her husband
preferred sports and parties.
6he resented the fact that he was not a
good provider and stated, "My father built a good home for my mother.
The husband chose Seattle
Here we are packed in; I need my own home",
The patient
as a place to live, and in 1946 they moved here..
always
resented the separation from her family and stated. “I
always
keep the far home available".
explained why this particular woman had Ibecome
then
Holmes
was
in the setting of unfulfilled deoendency ineeds,
tuberculous : "It
and
increasingly strained marital
adjustment
in a newi
and an
that the patient developed
pulmonary
unsympathetic environment,
This analysis recalled the holistic approach
to
tuberculosis " (33) .
that placed disease in the context of a patient's
psychosomatics
personal history.
The contention of this article is not that Holme s work proved
the link between stress and tuberculosis, but that the body of his
con t emp o r a ry tuberculosis
research has been completedly ignored in contemporary
liteature.
His research and his presentations sought to challenge the
standard model
of
the disease as a straightforward
infection.
"Although infection with the bacillus was necessary, tuoerculous could
not be understood without recognition of the etiologic role played by
underlying personalities and stressful
situations".
However
the
research spends that he initiated, has remained unworked on since that
time.
While Holmes developed a sophisticated model of tuberculosis as a
’psycho—somatic diseaseJ‘ he had little say about its prevention and
Of course one way to alleviate stress among the poor would
treatment.
to provide them with better jobs, housing and nutrition,
have been
like most contemporary medical people he advocated educating
However
people about their fears. anxieties—and mispercept ions. enabling them
to anticipate stressful events and thus adjust to them Petter.
Although we know that
The fate of Holme's work is not unusual.
our
tuberculosis
is a disease with social and environment root„s,
technical solutions ignore this.
Could the various factaors that we
are discussing, urbanisation, overcrowdmg, poverty mediate some □ f
their
influences
through stress Z
Is there a different way of
looking at the same problem ?
Summary of Thomas Holmes work of tuberculosis
*
In a study of 109 sanatorium residents Holmes found that pat ients
localised resolving tuberculosis, were anxious and had
who had
whereas those with
or high urinary ketosteroid levels.
normal
low urinary
and deteriorating TB were depressed and had
advanced
ketosteroid level.
. ,3
3
*
In a study of Seattle
residents ne
found
that
the disease
predominated m areas where those were1 poorer and non-wnite people.
*
To understand the effect of emotional stress and difficult
soc i al
relations,
he
interviewed 100 patients admitted
to
the Seatt1e
sanatorium.
He found that 71 7. had exoerienced financial hardship,
52 7. job dissat isf act ion,
31 7. met criteria for
alcohol ism and
that
these
factors clustered
in
the 2 years preceding
tne
admission.
*
He did
the
same
study using a control
group
and
a
interview
schedule which he devised to measure the previous
occurence
of
psychosocial
stress (Schedule of Recent Experience).
He compared
20 matched sanatorium employees who develooed TB and those who did
not and
found disturbing occurences more
frequently
in
the
preceding
two years of tuberculous
employees
alone,
He
al so
assessed
their "emotional integration" by a different
scale
and
found
that the TB employees were more
frequently
pathologically
i-iy
disturbed“.
h ■vecxtl Tit ->
*
In the next study he used a partially prospective design, comparing
patients who
redeveloped sputum positivity on treatment
after
3
months of sputum negativity (21 patients? to patients who
remained
sputum negtive (24 patients) . (jHe found that those who had '’“thrown
a positive" faced emotional problems during hosoitaiisation and had
long histories of
unstable lifes, lack
of
social.economic
and
family supports,
He then devised a new instrument to measure
tne
of
the
level
of
emotional
disturbance
which- bredisoosed
to
"throwing
a positive",
positive".
based on his data. HeHe re-analysed
10
controls
and assigned them scores on the basis of
their basel me
psychological
difficulties.
He predicted that of them, two wou i d
throw a positive
and his prediction
---- came true.
#
His final major study titled Experimental study of prognosis used
the Eerie Index, an instrument that identified
psychological
and
social factors characteristic of recovering patients,
A high Eerie
score predicted
recovery. He prospectively studied
41
randomly
selected patients. When 2b who had achieved a normal or high Eerie
Score were
located 5 years after the testing, none of
them were
treatment
failures.
However 5 of 15 patients who had
low Eerie
scores were treatment failures.
^hC\/<yv
' vu>fr'
ot-
gHQRT/ARTICLES
ADULT TUBERCULOUS MENINGITIS :
COMPARATIVE STUDY OF DIFFERENT CHEMOTHERAPEUTIC REGIMENS
VN Acharya*. BT Kudva**, VJ Retnamf, PJ Mehtat
5UMMARY
One hundred and two cases of adult tuberculous meningitis were studied over a period of 5 years.
According to treatment regimens, they were divided into four groups. Groups A and B were treated with
treptomycin, isoniazid and PAS/ ethambutol initially for 3 months followed by maintenance with INH +
pAC/ethambutol for 15 months (total 18 months). Groups C and D were treated with streptomycin,
soniazid + rifampicin or rifampicin + pyrazinamide for 2 months followed INH + rifampicin for 7 months
total 9 months). The response to therapy, which was quantitated by means of a scoring system on a
computer, was poor for Group A, fair for Group B and good for groups C and D.
INI
DUCTION
Tuberculous meningitis in adults has witnes
sed a drop in its mortality with the advent of anti
tuberculous drugs. Classically, long chemothera
peutic regimens have been advocated in its mana
gement. The disadvantage of this is frequent
patient dropouts due to non-compliance. The
advantage, however, is the low cost involved.
Grosser elucidated the bacteriological basis of
short-term chemotherapy in the treatment of tuber
culosis This regimen involves chemotherapy of 9
"nonths’ duration with essentially bactericidal
srugs. Many studies have been published using
the short-term regimen in pulmonary tuberculosis;
such a study has been lacking, however, in adult
tuberculous meningitis.
This prospective study was started in January
19
and the data on all cases was analysed in
December 1983.
Material and methods
i , ' One hundred and two consecutive cases of
I adult tuberculous meningitis were studied. Their
• failed neurological examination, investigations
at diagnosis and prognosis and a follow-up
complications and sequelae was done. The
^‘teria for selection of cases were as follows:
l^,essor and Head,** Lecturer, t Reader, t Reader (at present
Physic'an- TN Medical College and BYL Nair Hospital,
K^ment of Medicine, Seth GS Medical College and KEM
Bombay-400 012.
U^ed : 6.10.1984
13.7.1985
r ,985‘
Vol. 33.
No. 9.
1. ^CSF-nlnocyt^in:
2. RaJsed^CSE-^otcins-with low/low-normal CSF
sugar.
3. Positiye-CSF-srnear crrrehculture for acid fast
bacilji^4. Presence of other neural or extra-neural tuberculosis;
5.
response to anti-tuberculous
drugs in the presence of a high clinical index of
susgiciDn. At least four of these five criteria had to
be met with in each case.
The age range of these patients was 18 to 30
years (mean ± SEM 24.7 ± 1.1). The male to
female ratio was 3:1. These patients were random
ly assigned to one of four chemotherapeutic
regimens. In Groups A and B (long-term regimens),
the initial therapy was for 3 months and mainte
nance therapy for 15 months (total 18 months).
Groups C and D (short-term regimens) received
initial therapy for 2 months followed by mainte
nance therapy for 7 months (total 9 months). The
drug combinations used, their dosages and mecha
nism of action are listed in Tables 1 and 2.
Prednisolone was used in the dose of 30
mg/day initially and 5 mg/day as maintenance
therapy in all the 4 regimens. All cases were
studied on indoor basis for one month, and
followed up at approximately one, three, six, nine,
12 and 18 months. During follow up, clinical
appraisal, CSF examination, blood white cell count
and ESR were done to judge response. Both
clinical (percentage of amelioration of symptoms
and signs) and cytological (percentage of disap
pearance of lymphocytes in CSF) responses were
considered. The responses were quantitated after
assigning a point to each symptom and sign with
the aid of a computer. The response was
583
I
i
ABLE 1 : The dosages of various drugs used and their
mechanism of action.
Mechanism
of action
Drug
Dosage
jtreptomycin (SM)
soniazid (INH)
Ethambutol (EMB)
^ara-amino salicylic
icid (PAS)
Tifampicin (RFM)
°yrazinamide (PYZ)
25 mg/kg
10 mg/kg
15-25 mg/kg
C
C
150 mg/kg
10 mg/kg
35 mg/kg
s
c
c
s
S = Bacteriostatic
3 = Bactericidal
TABLE 2 . Details of the tour chemotherapeutic regimens
Groups
Initial therapy
Maintenance
therapy
Duration
A
(40)
18 months
3
(24)
C
(18)
INH, PAS
SM, INH, PAS
(15 months)
(3 months)
INH, EMB
SM. INH, EMB
(15 months)
(3 months)
INH, RFM
SM, INH, RFM
(7 months)
(2 months)
INH,
RFM
SM, INH, RFM, Pyz
(7
months)
(2 months)
D (20)
18 months
9 months
9 months
SM= Streptomycin; INH= Isoniazid; PAS = Para-amino
salicylic acid; RFM — Rifampicin; PYZ— Pyrizinamide.
TABLE 3: Response
regimens
Groups
A (40)
B (24)
C (18)
D (20)
to
anti-tuberculous
chemotherapy
A
(40)
B
(24)
C (18)
D (20)
2/
I
Clinical
Follow-up
Cytoiogical
response (weeks) response (weeks) after completlonoftherapy
(monthaO
Mean ± SEM Mean ± SEM
Mean ± SEM
24 2.2
24 ± 2.8
18— 1.5
8 ± 1.2
6± 0.16
5± 0.18
5 ± 0.14
3± 0.10
4 ± 0.20
4 ± 0.10
3 ± 0.05
2 ± 0.03
HC
OA
BPP
8 ± 1.0
(n = 10)
7 ± 1.0
(n= 5)
6 ± 0.0
(n = 1)
0
9 ± 1.0
5)
(n
7 ± 0.5
(n = 3)
0
0
12 ± 2
(n= 6)
11 ± 1
(n = 2)
0
0
MR
E
18 ± 1.0
(n= 8)
18 ± 0.0
(n= 1)
0
0
20— 0.0
(n= 1)
22 ±0.0
(n = 1)
0
0
Optic atrophy; BPP= Behaviour
HC= Hydrocephalus; OA
and personality problems; MR= Mental retardation; E =
Epilepsy
$84
*4 ■
1
RESULTS
The response to the four anti-tuberculoiy
regimens is shown in Table 3. The patients on ]
short-term therapy responded earlier and had
fewer neurological sequelae (Table 4) than those
on long-term regimens. The patients in Group d
had the best results. The patient from Group c.
one of the 5 from group B, and 3 of 10 from group
A who developed hydrocephalus required ventri- 1
culoatrial shunts. Hydrocephalous and optic 1
atrophy were early complications whereas beha
viour problems, mental retardation and epilepsy
came later. Based on computer analysis, the^
therapeutic response was judged as poor for i
Group A, fair for Group B, and good for Groups C <
i
and D.
DISCUSSION
In this study, short-term chemotherapy was
shown to give good results. This regimen has been
advocated for several reasons. It is known that
Mycobacterium tntierculosis is' an obligatory ri
aerobe grnwmcTontima 1 lyTaFa pH of 7 41.
It is hence not surprising that the organisms
invade the pia and arachnoid. Its replication t.me
is only 20 hours'. For optimum therapy, 3 populations of organisms need to be treated: (a)Jhe_ (
rapidly dividing extracellular nrganisms, (b)*e <• -~
slowly dividing extracellular organisms, and (cLffi0
slowly dividing intracellular organisms. The aim dW
short-term chemotherapy is to kill a large numb^ |
of organisms'in the rapidly.^idinfldaroup,
|
achieving rapid sterilization of the CSF a
I
lesions in the pia and arachnoid2. Furthermore 1
bactericidal drugs must be used in combinatio
|
prevent survival of mutant strains. Unlike in
j
nchymatous involvement of the lung and liver.
blood-brain barrier must be overcome.£hai£-£— I
logical studies inriirnta toat pyrazinamide dim1
1
into the meninges tn ahnut Rn% of the serum .e^ 1
Rifamnicinjoohas been shown to cross the
1
'i
— brain barrier and achieve a good
g"ood CSF e
.
j
Both these are superior to levels achiev
u.
isoniazid (about 50%) and ethambutol fabou g
The use of steroids in all patients was base
!■
TABLE 4 Time profile (in months) of the occurrence of neuro
logical sequelae.
Groups
said to be good with a score ot 9U7o and above
fair when 89 to 50% and poor when less than 50^'
The cut-off point of the response in time was com
sidered when maximum response could be judges
from the data collected. After completion of the
rapy. the patients were followed up for a period of
8 to 24 months (mean 18 ± 1.88) for neurological
sequelae.
I
'-i
IL
r
earlier results4,5.
S
JA PI
1985.
Vol-
..
-
as seen in Table 3 cytological response was
earlier than clinical response, but it was
^liest in Group D, followed by Groups 0, B and A
•^.hat order. The neurological sequelae were
*’ st in Groups C and D. While the four groups
^re age-matched, but not time, nutritionally or
*Lunologically matched, the patients did better
tf! short-term regimens than with long term*'imens. A longer term study is warranted to
^termine whether delayed response decides
Urological sequelae and the occurrence of re
pose if any in the short-term regimen.
REFERENCES
1.
2.
3.
4.
5.
Grosset J. Bacteriologic basis of short-course chemothe
rapy for tuberculosis. In: Clinics in Chest Medicine.
Philadelphia : WB Saunders Co 1980; 1 : 203-241.
Costello J. The modern treatment of pulmonary tubercu
losis. Postgrad Med J (suppl) 3: 181-184.
Mehta KP, Mahajan A,Umar Ali. Rifampicin in tuberculous
meningitis in children. Paper read at the 2nd Annual
Meeting of the Indian Society of Clinical Pharmacology,
Hyderabad, 1984.
Idriss ZH, Sinoo AA, Kronful MM. Tuberculous meningitis
in childhood. J DisChild 1976; 130:364-367.
Escobar JA, Belsey MA, Dueiias A. Medina P. Mortality
from tuberculous meningitis reduced by steroid therapy.
Paediatries 1975; 56: 1050-1055.
SERUM FERRITIN LEVELS IN APPARENTLY
HEALTHY ADULT POPULATION
I
Vasantha K Thavaraj, Vinodini Reddy*
SUMMARY
Serum ferritin concentration was estimated in 60 men and 60 women aged between 20 and 60 years.
They belonged to the high socio-economic strata and had normal haemoglobin levels. The mean level of
> arum ferritin was 50.4 ug/l and 24.0 ug/l In men and women respectively. Ferritin concentration was
i^nificantly low in women below 10 years of age, while the levels in post menopausal women were
J comparable to those of men. Thirty percent of them had values below 12 ug/l, indicating latent iron
H deficiency.
I
1 SfTRODUCTiON
Ferritin is the storage form of iron and only
25; wall quantities of it are present in the circulation1.
‘he development of a sensitive immunoradiomet-c assay has stimulated a number of studies on
'wriiin' status in different population groups2.
»a^6se studies suggest that the level of circulating
in is related to body stores of iron and,
gnl- before, it is a valuable method for assessment
status of iron stores3. In the present study,
^a*urn ferritin concentration was measured to
R]^ermine the iron stores and assess the extent of
iron deficiency in apparently healthy adult
and 13.7 g/dl respectively. Women had no mens
trual irregularities.
Venous blood samples were collected from all
the subjects for the estimation of haemoglobin
and serum ferritin concentration. Haemoglobin
was estimated by the cyanmethemoglobin method
and ferritin levels by immunoassay4.
RESULTS
Serum ferritin concentration ranged from 2115ug/l. The mean levels were 50.4 ug/l and
24.0 ug/l in men and women respectively (Table)
In women aged between 20 and 40 years, the
TABLE : Serum Ferritin Concentration in Men and Women
i^TERIAL AND METHODS
L A total of 120 adults belonging to the high
jj^O-economic group were investigated. There
L*® 60 men and 60 women and their ages ranged
F’’1 20-60 years They were apparently healthy
normai haemoglobin levels ( > 12 g/dl in
and >13 g/cll in men). The mean haemolevels in men and women were 15.4 g/dl
■k 091 Institute of Nutrition, Indian Council of Medical
Jamai Osmania PO, Hyderabad-500 007.
• 11.10.1984
13. 6.1985
No
Haemoglobin Serum Ferritin (ug/l)
Mean ± SE
Mean ± SE
Range
Men
60
15.4 ± 0.19
50.4 ± 3.2
9.2- 115.0
Women
60
13.7 ±0.11
24.0 ± 2.2
2.9 - 70.0
ferritin levels were significantly lower as compared
to those in men. Serum ferritin showed a rise with
age and in women above 40 years the mean level
was similar to that of men (Fig). 1 hirty percent of
the women had ferritin levels below 12 ug/l,
coc
k. k
| (ro K ■
H'
k + Fp : //nww^0 ov3
ewvAil : vko (a) pAlU ■ <rr^ •
L>k$ j'"'
'
2*, folle SKu4( A$U>^*
^boo^OSQ So MW
Social
KAT>
CGJ,*a/
bwokkh.
T
tut
Sol.
j).lv<A^v
«a
IUkK/-
Current Research
NTI Bulletin 1999, 35/1-4, 27-28
National Sample Survey to Estimate Annual Risk of
Tuberculosois Infection in Different Parts of India
A nationwide tuberculin survey with the National
x ^Tuberculosis Institute (NTI), Bangalore as the nodal
centre is currently under progress. The objective of the
survey is to assess the current epidemiological situation of
tuberculosis in different parts of the country by computing
Annual Risk of tuberculosis Infection (ARI)
The survey has been designed to estimate ARI in each
of the four zones of the country i.e., north, south, east and
west. In each zone 51,000 children aged 1-9 years are to
subjected to tuberculin testing. These children shall be
covered in 600 clusters spread over six districts, selected
by stratified systematic sampling. Therefore, a total of
about 2,04,000 children from 24 districts in the country
are to be included for the survey. The selected districts and
the number of rural and urban clusters to be covered in
each district are given in the Annexure.
A written consent for participation in the survey is
obtained from the parent/guardian of every child. Every
child is given ‘ITU of PPD RT23 with tween 80’
intradermally on the mid-volar aspect of the left forearm.
A suitable disposable tuberculin syringe and needle is used
for injection. A trained person reads the test results after
three days and the maximum transverse diameter of the
induration is recorded. The field work is conducted strictly
following the work instructions finalised after the pilot
studies.
Teams of Contractual Field Health Workers who have
dergone intensive training in tuberculin testing and
reading conduct the field work under direct supervision of
NTI and other collaborating centres. TRC, Chennai who
also have the responsibility of training all but one of the
eight survey teams is conducting the fieldwork in the
South zone. The fieldwork in rest of the country is being
carried out by NTI, which also has the responsibility of
conducting supportive studies.
'TG C-CvCtJ-vL, Delhi and Mahatma Gandhi Institute of
Medical Sciences, Wardha are collaborating for joint
supervision of the survey in 3 districts each of North and
West zones respectively. The support of district health
services and state TB cells is solicited for the conducting
the fieldwork.
The expected duration of the survey is about 3-4 years.
The analysis and reporting of the data shall be done
centrally at NTI.
Progress of the survey till 29.02.2000 is as under:
*
Consequent to acceptance of the survey protocol by
Central Steering Committee on operational research,
approval to conduct the survey was received from
Central TB division, Directorate General of Health
Services in April 1999. Subsequently, the finer
technical and operational details were re-examined
and the protocol was revised.
*
The work instructions were formulated and revised
following the experience gained in the pilot studies
conducted In Anekal, Bangalore (rural component)
and Mysore City (urban component).
*
The clearance from an Ethical Committee headed by a
retired Chief Justice
. has been obtained.
Recruitment of five teams of Contractual Field Health
Workers has been completed. The training of three
teams has been completed and another teams training
is under progress.
5b
*
The field work for the main survey which commenced
on 3.1.2000 in Dakshina Kannada and on 1.2.2000 in
Junagadh is under progress. About 30 clusters in
Dakshina Kannada and 19 clusters in Junagadh have
been completed so far.
The planning work including procurements has been
undertaken for initiation of the fieldwork in Rae
Bareilly, Medak and Delhi.
Dr V.K. Chadha
Epidemiologist
NTI, Bangalore
27
Annexure I
Details of Selected Districts for National ARI Survey
No of selected
clu sters
Total
Urban
Rural
Urban
Rural
197298
2270
3584
2694
3096
4654
1600
59100
329
842
762
282
2090
270
138198
1941
2742
1932
2814
2564
1330
180
13
33
30
11
82
11
420
61
86
61
89
81
42
Selected Districts!total)
17898
4576
13322
180
420
East Zone
Assam
Bihar
218346
828
3202
1496
2225
5523
2805
40291
49
387
91
345
679
233
178055
779
2815
1405
1880
4844
2572
110
3
24
6
21
42
14
489
27
96
48
64
166
88
Selected Districts (total)
16079
1785
14294
110
489
North Zone
Delhi
Himachal Pradesh
Punjab
Uttar Pradesh
198804
9421
1174
1757
2747
3215
2323
48632
8469
60
387
321
222
209
150172
952
1114
1370
2426
2993
2114
147
129
1
6
5
3
3
453
39
46
57
100
124
87
Selected Districts(total)
20637
9668
10969
147
453
West Zone
Gujarat
Maharashtra
229042
2395
3287
1544
5249
1130
2031
69508
778
2031
137
3391
98
739
159534
1617
1256
1407
1858
1032
1292
182
20
52
3
86
2
19
418
80
62
69
92
51
64
15636
7176
8460
182
418
South Zone
Andhra Pradesh
Medak
Belgaum
Dakshina Karnataka
Malapuram
Chingleput-M.G. *
Kanyakumari
Karnataka
Kerala
Tamilnadu
Golaghat
Bhagalpur
Dumka
Giridh
Cuttack
Bankura
Orissa
West Bengal
Madhya Pradesh
Rajathan
Delhi
Kangra
Gurdaspur
Hardoi
Jaunpur
Raebarcli
Junagadh
Nagpur
Ratnagiri
Thane
Jhabus
Kota
Selected Districts(total)
Total Urban Clusters 619
Total Rural Clusters 1780
Grand Total
2399
28
Population in 1000’s
(1991 census)
District
State
■3|
—■
:|^^^MliEffectiveness of BCG vaccination against tuberculous
^tt^S^neningitis: a case-control study in Sao Paulo, Brazil'
Filho,’ E.A. de Castilho,2 L.C. Rodrigues,3 & S.R.A. Huttly4
'csse-control study was carried out in the Metropolitan Region of Sao Paulo, Brazil, to determine the
^^^■-^^^^^^rotection against tuberculous meningitis conferred by BCG vaccination to children aged less then 5 years.
BCG vaccination coverage in the study area was about 88%. A total of 72 tuberculous meningitis
WGre studied as wel1 as 505 neighbourhood and 81 hospital controls.. Analysis of the deta using a
^^d'tiona, lo9'stlc regression for matched case-control studies indicated that the efficacy c4 BCG was
'?^riifar for b°th grouPs of controls, that for neighbourhood controls (84.5°/o) being slightly greater than that
controls (80.2°/o). No significant interactions were found between vaccination status and sex,
or socioeconomic status.
i, .
lb
fl introduction
that ranged from zero to 76% (2—5, 7-9,1Z, 17, IS, 24).
More recently, the case-control approach has been
advocated for the evaluation of the effeciiveness of
BCG vaccination (21, 22); however, such studies have
also reported a wide range of efficacies (13, 19. 23).
Notwithstanding doubts about the effectiveness
of BCG on the chain of transmission of tuberculosis,
and therefore on the incidence of and death rate from
the disease, the vaccine may still be usefd if it gives
protection against severe infantile forms of tuber
culosis. although such forms are not contagious (1,
15, 16). The results of a study in the United Kingdom
in 1950 already indicated that BCG vaccine offered
such protection (4); however, the resuits of a survey
conducted in Chingleput, India, were unciiear on this
matter (I, 12).
In Brazil, tuberci^nus meningitis is a notifiable
disease. For all cases of the disease a surveillance
form is completed, covering data from clinicai and
laboratory examinations and an epidemiculogicai his
tory (6). Despite the difficulties in diagnosing this
type of tuberculosis and the lack of confidence in the
data collection in countries with socioeconomic
characteristics such as Brazil, the seriousness of the
disease probably results in all cases being notined.
The hypothesis that BCG offers greater protec
tion against tuberculous meningitis than against pul
_________________
monary tuberculosis (4) is consistent wiri evidence
|y^^nivPar^rne 1 'J‘ ^Pidernio,°gy. Faculdade de Saude Publics,
from two Brazilian states with different schedules
for BCG vaccination. For example, in Rio Grande
de 830 Paul0, Av’ Dr Arnald0- 71S- SSo Paulo, 01255,
■ *■e^U0sts for reprints should be sent to this address.
do Sul, where children aged 7 years or older arei
O3wal<fo Cruz, Rio de Janeiro, Brazil.
vaccinated with BCG, the incidence of pulmonary
B-^^England0 ^C^00*
Hygiene and Tropical medicine, London,
tuberculosis in 1982 was about 12.5 times greater
than that of tuberculous meningitis. In contrast, in
liTSocial Medicine, Faculdade.de Medlcina, UniverSao Paulo, where children receive BCG vaccine dur
6 Federal de Pelotas, Pelotas, Brazil.
ing the first year of life, the incidence of pulmonary
K^Print No. 5047
H^t the beginning of the last decade of the 20th
■Century, tuberculosis still presents a public health
M^challenge, particularly in developing countries. In
jgr^’Brazil the death rate from tuberculosis has dropped
S^systematically since the introduction of specific
B^^S.e.motheraPy- However, the current death rate of 5.9
100000 Per
from tuberculosis indicates that,
B
frorn ’ntest’na^ infections and pneumonia, it is
S^still the infectious disease that causes most deaths in
|| ^the^country (14). According to official figures, the
H
of tuberculosis has shown a tendency to
*Q ®ra2^
annuai risk of infection is,
BS^owever’ not Prccisely^known because the policy of
S-^aSS and ^discriminate vaccination with BCG has
such estimates.
To reduce the global problem of tuberculosis,
fc®^ternadoriai health bodies have designated the
aw|L.endficahon and treatment of cases and vaccination
BCG as the principal components of control promrneS
lntradermal injection of BCG is contO
t^le ^est
immunizing against
s^^‘kiSeaSe' ^c^^cicss, the effectiveness of BCG
en Placed under doubt since several controlled
rePorted contradictory results, with efficacies
if
tln °f th° Wor,d Health Organization. M (1): 69-74 (1990)
IIB
<© World Health Organization 1&90
Si'
69
--- -------------------------------------------------------------
V. Wiinsch Fllho et al.
£
tuberculosi5 to that of tuberculous meningitis was
52:1; regional differences may, however, confound
this apparent relationship.
BCG vaccine coverage varies markedly in dif
ferent regions of Brazil. Vaccine coverage in Sao
. Paulo city is reported to be very high, exceeding
100% according to official estimates; a survey con
ducted m 1982-3, however, estimated that the
coverage was 88% for children less than 18 months
of age/*
In view of these facts and of the necessity to
better define the role of BCG in tuberculosis control
programmes, we carried out a case—control study in
the Metropolitan Region of Sao Paulo to evaluate
the effectiyene?s_of BCG^ainst tuberculous meningitis. Our findings are reported here.
Materials and methods
Study area
The study was conducted in the Metropolitan Region
of Sao Paulo (MRSP), which comprises 37 muni
cipalities within an area of 8053 km2. The population
of MRSP recorded in the 1980 census was 12.5
milhon, which corresponded to 50.3% of that ofthe
State of Sao Paulo and 10.6% of the entire Brazilian
population. A wide range of living standards prevail
in the study region, with some areas having only very
basic environmental and social amenities. MRSP is
very urbanized with only a few rural areas in the
outskirts of the city (20)?
Tuberculous meningitis
Between 1979'and 1983, about 150 cases of tuber
culous meningitis per annum were notified in the
State of Sao Paulo, and of these a third were among
children under 5 years of age (incidence, approx
imately 1 7 per 100 000 per annum). The majonty of
cases of the disease notified in the State of Sao Paulo
are admitted to two hospitals: Emilio Ribas or Mandaqui.
BCG vaccination
Routine BCG vaccination (Moreau-Rio de Jandl
strain)0 is compulsory in Brazil as part Of^
programme established by the Ministry of Health $
is recommended that the vaccine be given tO'childr^l
without a previous tuberculin test, between birth aS)
the end of their first year of life. Although BCG 1
widely used the protection afforded by it has nevi
been assessed in Brazil
S
Selection of cases
All notified cases of tuberculous meningitis whjS
were admitted to the Emilio Ribas or Mandaoiil
Hospital from 1 January 1981 to 31 December I92T
that involved patients who were bom after 197?
(coverage with intradermal BCG became high
Brazil only from 1979) were ascertained. Because 0?
the difficulties in diagnosing tuberculous meningitf
unequivocally, criteria were defined based on clinic^and epidemiological findings, as well as on theTSM
of bacilloscopy, culture of cerebrospinal fluid (CSFi
MRS1^01"0135^
CaSCS seIected wcre residents
Data on cases were collected in two questiol
naires one based on the hospital records and the
other on household interviews to chaffy,,-and com-plement the hospital record data, particularly ii
relation to the BCG vaccination status of the chi
dren. The vaccination status of children was deter
mined from their vaccination cards and, if possible!
by the presence of vaccination scars. Cases wer:J
classified as BCG-positive only i:'their date of vaccine
ation preceded that of diagnosis of tuberculous men^
ingitis. In situations where a case died, the child!
concerned was still included in the study and at
household visit w-as made. The mother was thee?
questioned about the dead chile’s BCG vaccinatioJ
status and was asked to prod ice the vaccinatioo®
card. If the mother had lost the child’s card, health®
centre archives were searched. The mother’s worej
was accepted only if she affirmed that her child hajj
never been vaccinated, and such children were ciasS
sified as unvaccinated. Cases were excluded from tbl
analysis if they could not be located or information^’
could not be obtained about their vaccination stati^
Selection of controls
2'llStrat
Vuaccination c^age in the municipaiity o( Sao
5 o Pal" Un bi
'n,Ormation- Government of the State of
Sao Paulo. Unpublished document, 1982 (in Portuguese).
6 [Health care in the Metropolitan Region of Sao Paulo']
Paper
presented at the WHO/PAHO Regional Meeting of the Technical
Consultation on Primary Health Care and Development Services
pLn035 and Lar90 CitieS1 Washin9‘on, DC, 20 November
1981. PAHO unpublished document (In Portuguese).
Neighbourhood controls. Cases and controls weitf
matched by hpiQe_area and socioeconomic stratun^
In order to obtain a minimum of four suitably
controls per case, eight potential controls wef3
FundaqSo Ataulpho de Paiva, Rio de Janeiro, Brazil.
70
WHO Bulletin OMS. Vol 68
*
Effectiveness of BCG vaccination against tuberculous meningitis
children in the neighbourhood of each
i ^^eOJhefrommother
of the index child was asked to
Ignommate two children of neighbours, each of whose
^toothers ^nominated two more. This process was
^^ntmued until eight controls were identified. The
^0nljnfequisite was that the children nominated
should have been born after 1978. Information on
Ifecoiitrois was collected during home visits, using a
wSvery'similar questionnaire to that used for cases. To
gteSetermine the BCG vaccination status of the neigh^fefjourhood controls, the same procedure was used as
^R'iop^ses’ aQd only those children whose vaccination
f^TSatus was known with certainty were accepted. Chib
dren who had a past or present history of tuberculosis
M ^were rejected as controls.
hfe''
■
s ^(Hospital controls. In order to detect any biases that
have ^een introduced by the neighbourhood
S gSntrols, a second series of controls was selected from
?
patients in Emilio Ribas Hospital. In order to ensure
S.that there was one suitable hospital control for each
| Scase, attempts were made to identify three potential
| ^controls per case.
~
I ^^V^Hospital controls who were suspected to have
I SrSatuberculosis were excluded. Also, children with
I
wh’ch couJd bi ve been prevented by vaccine-8-’ measles or d.phtheria, were also excluded,
^^cause it was considered likely that those not vaca8ainst thesc diseases would have had less
Ig^Kana: of having received BCG vaccine. The hospital
0^^-r°1S Were selected rrom patients admitted with
caused by Streptococcus pneumoniae or
influenzae and who also satisfied the
recluirements of sex, home area, date of
date of admission to hospital ±6 months
< ’■iX<VTeSpeCt ^Jba.t of the index case.
K?.p’Y‘T.W0 q^stionnaires were used for hospital cont0 °blain ho:iPitaI Acords data and the
which was similar to that for the neighbour®-|M2?4?.controls, for the home visit. The BCG vaccinaof the hospital controls was determined in
way that of the cases and neighbourhood
li^SUtus
an& estimation of vaccine efficacy
?Umb€r
Ifeverat 8
cases that were expected to be
StUdy p€riod and the level of BCG
' above 5no/ Wa?j concluded ^at vaccine efficacy
0 C°U d be detected at the 5% level of
I Bff’-significan/
‘e at a power of 80%-
f ship Udd /?naCy Was estimated from the relationfPtuberculoid
■ Where RR is the relative risk of
B Pared to th meninglds among the vaccinated comt -the odds raet Un^fC’nated Mildren, estimated from
U0‘ The odds ratio was calculated by
1 ■■WHO Bulletin
conditional logistic regression analysis for matched
case-control studies, using the EGRET software
package (1,0).
Results
Characteristics of the study population
During the study period, a total of 474 cases of
tuberculous meningitis were notified in Sao Paulo
State, 196 (41.4%) of which involved 0-4-year-olds.
In the two hospitals where the investigation was
carried out, 271 diagnosed cases of tuberculous
meningitis were admitted, Hl (42.4%) of which were
children aged less than 5 years. Any case that trans
ferred between the two hospitals or which was re
admitted was counted only once, using the initial
admission to the first hospital in the respective year.
A total of 94 eligible cases remained, 19 of which
were omitted from the study (11 were born outside
the study area, while eight could not be located). Of
these 75 cases that were located and visited, there was
uncertainty about the true vaccination status of two
children, who were therefore also eliminated from the
study. Also a control could not be obtained for one
case, which was therefore removed) Altogether, 72
cases fulfilled the study criteria (25 cases from 1981,
24 from 1982, and 23 from 1983).
For these 72 cases, which formed the basis of the
study, the (history of contact with tuberculosis
patients was traced for 46 of them (63.9%j) Chest Xrays were available for only 50 of the 72 cases, and 46
(92%) were positive for tuberculosis. The results of
CSF cultures were available for all cases, and Myco
bacterium tuberculosis was isolated from 10 children
(13.9%). The results of CSF smears for acid-fast
bacilli were available for all cases and were positive
for four children (5.5%). The overall case fatality
rate was 50%, but was higher among those aged less
than 1 year (60.5%). Of children who survived,
many suffered neurological sequelae from which they
recuperated with difficulty. Only 11 of the children
investigated recovered without exhibiting apparent
neurological abnormalities (Table 1).
A total of 520 neighbourhood controls and 83
hospital controls were visited. Of the neighbourhood
controls, 15 were omitted (12 were bom outside the
defined study area, one was ^ng treated for-pulmonary tuberculosis during the period in which he was
visited, and for two others doubts remained as to
their true vaccination status). Neighbourhood con 9 .
trols could not be located for four cases. C.
__ _83
Of the
hospital controls, two were omitted (one was bom
boi
outside MRSP and there were doubts about the
vaccination status of the other), leaving 81. Hospital
controls could not be obtained for 12 cases.
°^S. Vol 68 1990.
71
i
1
j> j
*
V. Wiinsch Fllho et al.
Table 1: Clinical outcome of the 72 cases of tuberculous
meningitis that formed the basis of the study, Sao Paulo,
Brazil
No. of cases
Outcome
36 (50.0)'
Died in hospital
Died after discharge
from hospital*
2 (2.8)
Neurological sequelae
23 (31.9)
No apparent
neurological sequelae
11 (15.3)
Total
72 (100.0)
Table 2 shows the distribution of cases and
controls by age, sex and socioeconomic status
(defined in terms of family income, area of residence,
degree of domestic crowding, and mother’s education
level), and vaccination status. For ease of presenta
tion, unmatched data arc shown; however, matching
Table 2: Characteristics of the study cases and controls,
Sao Paulo, Brazil
No. of
nei( hbourhood
controls
No. Of
hospital
controls
20 (24.7)
34 (42.0)
7 (9.7)
35 (6.9)
36 (7.1)
W4 (20.6)
330 (65.3)
Sex
Male
Female
49 (68.1)
23 (31.9)
25 0 (49.5)
255 (50.5)
51 (63.0)
30 (37.0)
Socioeconomic
status’’
1
2
Unknown
58 (80.6)
13(18.1)
1 (1-4)
431 (85.3)
74 (14.7)
55 (67.9)
No. of
cases
Age
<6 months
6-11 months
12-23 months
14 (19.4)*
30 (41.7)
21 (29.2)
>24 monthi ,
Vaccination
status
Vaccinated
Unvaccinated
Total
was- preserved to estimate vaccine efficacy. T^J
thirds of subjects came from the municipality of Sa0
Paulo, most from the peripheral poorer areas, while
the remainder were from the other 36 municipaliti^
of MRSP. Hospital ^controls were matched by
and by age (within 6 months) but differed from
with respect to their socioeconomic status, mortg
frequently coming from higher status groups. Neigk®
bourhood controls, as expected, had the sarnjw
socioeconomic status as cases, but were older and®
had a higher proportion of females.
Efficacy of BCG vaccination
' Figures in parentheses are percentages.
* Death was verified at a home visit.
Characteristic
I »
------- ---------------- H
18 (22.2)
9(11.1)
4
Vaccine efficacy was calculated separately for each’
group of controls using a conditional logistic r?
gression analysis, and the results are shown in Table
3. The efficacies obtained were similar and high, the
efficacy for neighbourhood controls (84.5%) bei^
slightly greater than that for hospital control
(80.2%). No significant interactions were fourS
between vaccination status and sex, age, or soaoeconomic status.
Discussion
--J
To the best of our knowledge, this is. the first largil
study to quantify the effect of ^BCG vaccinatioij
against tuberculous meningitis. Previously, Miceli dj
al. in a case-control study in Argentina reported thz!J
BCG vaccination had an efficacy of 100% againsl
this form of tuberculosis, although the sample sid
was small (13). Our results indicate that in the studjl
community in Sao Paulo, BCG vaccination wan
highly effective against tuberculous meningitis in chM
dren below 5 years of age. This finding is verj
encouraging for the prevention of a disease that has!
high fatality rate and serious neurological sequelae,
among many of those who survive. Tuberculoig
meningitis predominantly affects children fro|
26 (32.1)
Table 3: Efficacy of BCG vaccination against tubercukw
meningitis for matched pairs of cases and controls,
Paulo, Brazil
•
42 (58.3)
30(41.7)
^63 (91.7)...
42 (8.3)
72(88.9)
9(11.1)
72
505
81
* Figures in parentheses are percentage:;.
* Children were classified as socioeconomic status 1 if they
satisfied at least three of the following conditions: per capita
household income less than one minimum wage (about USS 58.26
in May 1986); greater than four persons per bedroom; resident in
the peripheral area of Sao Paulo municipality or other mun
icipalities of the Metropolitan Region of C.So Paulo; and mother
illiterate or only partly literate.
Vaccine
efficacy (%)
Cases and neighbourhood
controls*
84.5 (66.7-92.3%)*
Cases and hospital
controls'
80.2 (40.6^-93.4%)
.
•-'?
* Adjusted for age, sex, and socioeconomic status.
* Figures in parentheses are the 950/o confidence intervals. ^
' Adjusted for socioeconomic status.
§
WHO Bulletin OMS. Vol 68,9j
72
h ^3
Jo
u *•
‘ ■ bW
fe’? —
•
Effectiveness of BCG vaccination
|^x)or^r socioeconomic and environmental back^grouncis, who may also be less likely to have been
VSyaccinated with BCG. Since the majority of cases
J ^occur among children aged 3-11 months, we recoml^'gmend that BCG vaccine be administered within the
^prst .'3 months of life and that efforts are made to
achieve a high coverage across all socioeconomic
R6sum6
Efficacite de la vaccination par le BCG centre
la mdnlnglte tuberculeuse: 6tude cas-temolris
3 S3o Paulo, Br6sll
01 the «se-«ontrol approach has been
Igrecommended for studies of the effectiveness of BCG
^^vaccination (15, 21, 22). Compared with controlled
^trials; the case-control approach is both quicker to
r;...carry out and cheaper. In accord with the reports of
pother workers (13, 19, 23), our findings indicate that
si^the case-control method is useful for eval rating the
UjCffecttveness of BCG vaccination. It is particularly
^encouraging that similar results were obtained with
psthe two groups of controls, hospital controls
^although easy to locate, differed from cases with
f ••.respect to socioeconomic status.[The neightourhood .
controls, although they had the kme socioeconomic I
j
against tuberculous meningitis
A I'approche de la fin du XXe siecle. la tuberculose
pose encore un probleme de sante publique
notamment dans les pays en developpement. Pour
y remedier, les organismes sanitaires internationaux ont decide d’axer les programmes de lutte
sur I identification et le traitement des cas de
maladie et sur la vaccination par le BCG.
Bien que le vaccin BCG soit largement utilise
i
6 m°nde a titre de mesure preventive centre
la tuberculose, sa valeur a ete remise en question,
L’article presente iune discussion
"
du role du BCG
en tant que mesure de lutte cc.,;,
J centre la tuberculose
et etudie la protection qu’il confere
gite tuberculeuse.
--------------- centre la menin-
;ame socioeconomic HoV
F’^t'■ !:es Pollt|gues de vaccination par le BCG dans
ige and sex d^bmmSXXwe'X , a
ightly higher for neighbourhood controls but this
h. -ay have arisen because of further confounding No
fMS“usaandUnd f°r interactions
'accmfatlJs and a£e’ sex, or socioeconomic sta us but
In most? Wat rathen Smal‘,0StUdy these
sa=~:s
■ doe not ,
agains' tut*™losis, althoigh i°
^se^Ou^u^'^^ 56
of the
•
of tube "
enCoura8in8
'he prevcn•^stimulate further cn^6011181^15 3nd s^ould hopefully
Ration and childhooH^k0^0 Studies of B(~G vaccin•_
hildhood tuberculosis in other countries.
'
,j. es Etats bresiliens de Rio Grande do Sul et de Sao
Paulo, qui ont differents calendriers vaccinaux, ont
ete evaluees; on a pour cela examine I'incidence
de a meningite tuberculeuse dans chacun de ces
Etats. Dans I Etat de Rio Grande do Sul, oil les
enfants sont vaccines a I'age de sept ans I’lncidence de la meningite tuberculeuse est environ
quatre fois plus elevee qu'a Sao Paulo ou les
enfants sont vaccines avant I'age d'un an La
couverture vaccinale du BCG varie sensiblement
d une region d I'autre-du Bresil. Par exemple en
ville de Sao Paulo, elle est tres elevee, ddpassant
meme tOO /. selon les chiffres officiels; toutefois, une
Faav f
S6e e" 1982-19S3 ^nduit a estimer
mrnnl0 n C°UVerture vaccinale chez les enfants de
Tune
n mo's. L'article rapporte les resultats
dune etude cas-temoins menee dans la zone
urba.ne de Sao Paulo (Bresil) afin de determiner la
tuberr 7"
le BCG conlre la meningite
tuberculeuse chez les enfants de moins de cinq
<
g^nowledgementS
g' sm°rhA LPi0 ,and Dr H-G- 'en Dam. both of'
i icTropicai Medicine and n r n^00' Of Hy9iene and
i^eral de Pe|otas' (or. °.r
Victora- Universidade
;^Cal assistance This J?! he P Ul comments and tech: .•^hout the ful( cooperabon
P°Ssib,e
™bas and Mandaoui Hnf » hG med'Cal staff of Emilio
: , L- Walsh for his conthh' * S'
are Qrateful also to
^nuscript.
tnbution to the preparation of this
Tv^This invpctic^.
Who.
Ion was partiy supported by a grant from
r-,,1 L’®tUdcnf POrt6 Sur 72 cas de meningite tuber-
hosoi^’nmOinS de voisinage et 81 temoins
ho,spitaliers. On a calcule I'efficacite du vaccin
selon la formule 1-flfl, dans laquelle flfl est SZ
suiets6 vt31" de m®nin9ite tuberculeuse chez les
Cin!l vaccines Par raPPort aux sujets non vac
cines expnme par le odds ratio. Avec une analyse
de regression logistique conditionnelle nous
avons calcule que I'efficacite du vaccin Ttaib
ldqdre9mZntdT
9r°UpeS de tamoins,
legerement plus grande toutefois chez les temoins
nd I OIS'n^e (84'5%) qUe chez les temoins hosP taliers (80,2 Zo). On n'a observe aucune relation
'OOMS. Vol 68 1990.
73
&
I
V. Wiinsch Fllho et al.
significative entre I'etat vaccinal et le sexe, I’age
ou le niveau socio-economique.
Bien que des reserves aient ete exprimees
quant a I’efficacite du BCG sur la chaine de trans
mission de la tuberculose et par consequent sur
I’incidence et la mortalite generale dues a cette
'maladie, nos observations montrent qu'il peut etre
utile s’il protege centre les formes infantiles
graves de tuberculose. Comme la plupart des cas
de meningite tuberculeuse chez I'enfant surviennent chez les nourrissons de 3 a 11 mois, nous
recommandons d'administrer le BCG au cours des
trois premiers mois de la vie.
References
i!
i
1. Aneja, K.S. Immunoprophylaxis in tuberculosis. Jour
nal of communicable diseases, 16(1): 4&-53 (1983).
2. Aronson, J.D. et al. A twenty-year appraisal of BCG
vaccination in the control of tuberculosis. Archives of
internal medicine, 101: 881-893 (1958).
3. Medical Research Council Tuberculosis Vaccines
Clinical Trials Committee. BCG and vole bacillus
vaccines in the prevention of tuberculosis in adoles
cents. First Report. British medical journal. 1: 413427 (1956).
4. Medical Research Council Tuberculosis Vaccines
Clinical Trials Committee. BCG and vole bacillus
vaccines in the prevention of tuberculosis in adoles
cence and early adult life. Third Report. British
medical journal, 1: 973-978 (1963).
5. Medical Research Council Tuberculosis Vaccines
Clinical Trials Committee. BCG and vole bacillus
vaccines in the prevention of tuberculosis in adoles
cence and early adult life. Fourth Report. Bulletin of
the World Health Organization, 46: 371-385 (1972).
6. Centre for Health
Information.
[Manual for
epidemiological surveys. Bules and instructions.']
Sao Paulo. Secretary of State for Health, Government
of the State of Sao Paulo, 1978 (in Portuguese).
7. Clemens, J.D. et al. The BCG controversy. Journal of
the American Medical Association,. 17: 2362-2369
(1983).
8. Comstock, G.W. & Palmer, C.E. Longterm results
of BCG vaccination in the southern United States.
American review of respiratory disease, 93: 171-183
(1966).
.
9. Comstock, G.W. & Webster, R.G. Tuberculosis stitaies
in Muscogee County, Georgia. American review of
respiratory disease, 100: 839-884 (1969).
74
10. EGRET epidemiological graphics, estimation
testing package. Statistics and
EpidemiolonyS
Research Corporation, Seattle, WA, 1988.
11. Brazilian Institute for Geography and- StatlstTj^l
[Annual statistics for Brazil]. Brasilia, Secretariat fofa
Planning of the Office of the President of the’W
Republic, 1983 (in Portuguese).
12. Indian Medical Research Council. Tuberculosis B
Prevention Trial, Madras. Trial of BCG vaccine in'S
South India. Indian journal uTiitedical research,
(suppl.): 1-74 (1980).
13. Miceli, I. et al. Evaluation of the effectiveness of BCQ^
vaccination using the case-control method in Buenc-JP
Aires, Argentina. International 'Journal" of eppfc-j
demiology, 17: 629-634 (1988).
14. [Mortality statistics. Brazil 1980.] Brasilia, National
Secretariat for Basic Health Care, National DivisionW
of Epidemiology, 1983 (in Portuguese).
15. WHO Technical Report Series No. 651, 1980. (Vacc^g
nation against tuberculosis:, report of an ICMR/WHolgl
Scientific Group).
16. WHO Technical Report Series No. 652, 1980 (SCGW
vaccination policies: report of a WHO Study Group)J|i
17. Palmer, C.E. et al. Community trials on BCG vaccina’®
tion. American review of tuberculous and pulmonary^
diseases. 77 : 877-907 (1958).
,
18. Rosenthal, S.R. et al. BCG vabcination against tuber-'^
culosis in Chicago. A twenty-year study statistically®
analyzed. Pediatrics, 28: 622-641 (1961).
19. Shapiro, C. et al. A case-control study of BCG ano^'j
childhood tuberculosis in Cali, Colombia.
national journal of epidemiology, 14: 441-446 (1985)-^B
20. [Studies and investigations—Part 73.] Sao PaulolS|
Coordinator for Analysis of Data, Secretary for Ecciv®;
omics and Planning, Government of the State of Sad®
13
Paulo, 1977 (in Portuguese).
21. Smith, P.G. Retrospective assessment of the effec-^'
tiveness of BCG vaccination against tuberculosis®:,.'
using the case-control method. Tubercle, 63: 23-35^.'.
(1982).
22. Smith, P.G. et al. Assessment of the protectiveg^;
protectiveg%
efficacy of vaccines against common diseases usinjW.?
case-control and cohort studies. International
nal of epidemiology, 13: 87-93 (1984).
23. Smith, P.G. Case-control studies of the efficacy
BCG vaccine in children. Bulletin of the Internation^^^
Union against Tuberculosis and Lung Disease, 62(3«B
70-76 (1987).
24. ten Dam, H.G. Research on BCG vaccination.
ces in tuberculosis research, 21: 79-106 (1984).
25. WHO Technical Report Series’No. 671, 1982
culosis control: report of a joint IUAT/WH0
Group).
WHO Bulletin QMS. Vol 68
Women and TB
A.M. Kudale & D.C. Deshmukh traces the methodological problems in field-based research on
gender differentials in Tuberculosis and offers some practical solutions
Introduction
Centre, Pune were each to be interviewed by one male and
one female investigator.
Despite decades of research and effective treatment
strategies being available, the problem of tuberculosis remains
n) Community based patients: Mapping of all patients
unsolved, taking its toll on society even today. The Government
diagnosed as suffering from TB 3-4 months before the start
of India is rehauling its National Tuberculosis Programme in
of data collection or currently on treatment in the Parinche
an effort to ensure the cure of TB patients by providing the
Valley, Pune District (FRCH's action-research area consisting
most effective medicines and by direct monitoring of treatment
of 20,000 population) would be done by contacting all
follow-up of patients. However, as with other such projects,
available formal and informal sources. The patients would
successful implementation of this programme would require
be interviewed by an FRCH researcher working at Parinche.
community involvement for which it
Information about these patients
is mandatory to study the disease in
TB Takes its Toll
would be collected by community
its socio-cultural aspect.
* Approximately one third of the world's population is infected based women known as "tais" as
with Mycobacterium Tuberculosis.
per specific guidelines provided to
While much has been written * Tuberculosis kills over one million women each year.
about the influence of socio * About 80% of deaths due to tuberculosis occur in the them by FRCH staff. Tais would
economic and cultural factors on age group of 5-49 yrs. - the most productive years of life. prepare case studies of the
compliance with tuberculosis * The risk and prevalence of tuberculosis infection are similar patients and do a follow-up of
in males and females until adolescence, after which, they
their treatment behaviour for six
treatment, there is hardly any research become higher in males.
done on gender differentials in * Tuberculosis is as prevalent in the rural areas as it is months. This would also allow us
to study the advantages/
tuberculosis and its control. Hence, in the urban concentrations.
disadvantages
of involving
FRCH has undertaken a study on the gender specific problems
community
based
female
workers
in
a
research
process.
of help-seeking behaviour, of women TB patients,
simultaneously providing them with useful information to
iii) Non-affected persons : Non-affected persons would be
help improve their access and adherence to treatment.
selected from 20 villages picked out from a list of villages
from where come the patients registered at the DTC. These
Methodology
20 villages are within a radius of 50-60 kms from Pune city,
The study methodology includes conducting interviews
for the purpose of logistics and convenience of data collection.
of TB patients and non-affected people with the Exploratory
. iv) Health care providers : All health care providers in the
Model Interview Catalogue (EMIC)* method, which is the-first
same 20 villages would be identified through discussions
time such a method is used in data collection in studying TB.
with key informants from the villages and interviewed with
The purpose of using EMIC in the present study is to
the help of a semi-structured interview schedule. This would
identify patterns of distress, perceived causes and the
include all providers identified and visited by villagers including
preferences as well as past history of help seeking and
traditional and folk healers for health and medical care.
treatment. A section of the EMIC will be focussing on social
v) It is proposed to conduct a focus group discussion (FGD)
stigma attached to the disease, which has an important effect
in the evening after completing the interviews in each of
on TB-felated illness behaviour, to produce a stigma scale.
the 20 villages. In each village emphasis would be laid on
Selection of Patients and Interview Strategy
Initially, the following strategy for selection of patients
and their interviews was decided upon:
i) Clinic based patients : 80 rural clinic-based patients (40
male and 40 female) who have been registered for treatment
4-6 weeks prior to the date of interview at the District TB
participation by a different socio-economic or gender strata
in the FGD so that all stratas would be covered. The Sarpanch
or members of the Mahila mandal were to be contacted to
help gather the kind of people we wanted for the discussion.
It has been our experience so far, that besides the target
group many others take part in the discussion and it is quite
difficult to stop them from interfering in the proceedings.
5
FRCH NEWSLETTER, XIII NO. 1, JAN.-FEB. ’99
Village Health Ethnographic Study
To understand the community's perceptions about health
it was decided to conduct a detailed study emphasising health
and related aspects of the village concentrating on stigma
related diseases such as TB. leprosy, HIV/AIDS etc. with
a focus on gender differentials. This study was to be conducted
by a village based local women researcher. This is an attempt
at developing methodology for collecting health information
using qualitative research methods with the help of a village
based local woman as an investigator by giving her appropriate
inputs.
Altering the Methodology
During pretesting we faced certain difficulties regarding
our methodology hence, after due discussion with members
of our research advisory committee, certain changes have
been made in selection criteria of patients as follows:
Clinic based patients: As Directly Observed Therapy
(DOTS) has been implemented in the whole of Pune district,
we had to take patients who came under DOTS treatment.
Not a single rural woman patient who attended the clinic
at DTC and who fulfilled our criteria could be interviewed
by us. The explanations given by medical staff at the DTC
and as understood by the researchers are:
* Even if a woman came to the DTC she is always accompanied,
by her husband, brother, mother-in-law etc. and hence, is unable
to talk on these sensitive issues in their presence.
* She is in a hurry to go back home.
* Most of the patients, female as well as male, are transferred
to the nearest PHC for further treatment.
Hence, the criteria for interviewing patients was changed and
it was decided to visit rural hospitals and PHCs in rural areas
of radius 50-60 kms from Pune to cover the target.
Moreover, it was observed during pretesting that females
feel more comfortable with and talk more openly to female
researchers. So instead of one male and one female researcher
it was decided that female researchers would interview female
patients and male researchers, male patients. The same
strategy was followed in interviewing non-affected persons.
The- qualitative information gathered in the interview is
immediately coded by the interrater method. The code is
mutually decided upon by the researchers.
Initially, we had planned to conduct FGDs in the evening
when all the women would be at home and had some time
to spare for us but it was suggested by village women and
others in the community to conduct FGDs in the morning
on a weekly market day. The reasons given were: firstly, it
being the market day nobody would go for work and hence,
everybody would be available for group discussions and
secondly, in the evening most of the men folk in the
community came home absolutely drunk making it difficult
for women to come for group discussions. Incidentally, the
problem of alcoholism was prominent in 2 of the 4 villages
where FGDs were conducted.
After due discussion it was decided that as it would be
very hectic for the community based patient if s/he was visited
by FRCH researchers and again by tais for in-depth study
and follow up, instead, only tais would visit the patient and
take in-depth information form patients. When it was observed
that tais give more information orally than in writing, as
writing seems to be a constraint for them, we decided that
the one of the FRCH researcher would write whatever relevant
information the tais had gathered about the patients. The data
collected through this method has been purely qualitative
to date and will be analysed by the computer package popularly
used for qualitative data 'Text Base Beta’.
These were a few methodological problems faced by the
researchers in the present study and the necessary
modifications made to our strategy. We hope our experience
will benefit other researchers carrying out similar work.
Note : • EMIC: The Explanatory Model Interview Catalogue is a highly
appropriate method for acquiring information about social and cultural
determinants of illness, help seeking, treatment adherence and other aspects
of illness behaviour. Though rather lengthy and exacting, it is able to
simultaneously collect qualitative as well as quantitative information in depth
about any particular illness.
EMIC has been developed in studies of tropical diseases and mental
health in India and used in Africa, East Asia, Europe and North America.
It has already been usca for operational research studies of leprosy,
onchoceriasis and childhood diarrhoea.
Contd. from page 4
ensure proper utilisation of the existing public and private
health systems through health education, awareness of
individual rights, more economic resources to help them buy
better services, timely referrals to public/private providers
by the village health workers in complicated cases etc.
Empowerment of the People - Bottom-up Approach:
It has been realised that no development initiative including
FRCH NEWSLETTER, XIII NO. 1, JAN.-FEB. ’99
basic health care can succeed unless people's participation
in the scheme is ensured. The 73rd amendment to the
constitution now offers people an opportunity to initiate
and plan schemes for their own development. The 33%
reservation provided to women and schedule castes in
Panchayats should ensure the participation of the weaker
sections of the society in the political process and give
them socio-economic empowerment.
6
rlu^wAri^»vx.~t.u: 4Zj',4t.>'4j.i7S.^XM.'i.
•■< 'w,7^-k
(
TB Control
NTI Bulletin 1999, 35/1-4, 3-7
Tuberculosis Control and Economic Issues
2
VK Chadha 1 , Preetish S Vaidyanathan", Sanjay Singh
SUMMARY
The health of its people is reflected in the economy
of a nation - healthy people produce healthy
economies. It is unfortunate that in our country the
effects of ill health on economy have not been fully
appreciated. The burden imposed on individuals,
families and the community by diseases like
tuberculosis (TB) contains an economic dimension. TB
extracts costs - invariably in an economic sense - at all
levels of the society, either directly through
e* mditure incurred in providing health and social
c^_- and support, or indirectly in terms of lost
opportunities such as loss of employment. Other
intangible costs include the anguish and anxiety
experienced by the patients and their families. The
havoc wrought by TB on individuals, families, whole
communities and economies is enormous and the
ensuing discussion shall focus on the economic issues
related to the problem of TB and its control.
KEY
WORDS-.
HEALTH
ECONOMICS;
CONSTRAINTS, SCC, DOTS, FINANCIAL
SUPPORT
THE BURDEN OF TUBERCULOSIS
•
The magnitude of TB can be appreciated from the fact
that it is now the world’s foremost cause of death from a
sin°le infectious agent killing more people than AIDS,
n ria and other infectious diseases1. There are about 3
million deaths from TB annually all over the world with
someone dying of TB every 10 seconds2’3. In India alone,
there is one death due to TB every minute4. Though many
children die of tubercular meningitis and miliary TB, the
brunt of the disease is borne by those in the age group of
15-59 years5'8. Approximately 6.7% of all deaths and
18.5% of deaths in the age group 15-59 year age group in
the developing world are attributable to TB2’8. Though
TB deaths are more common amongst males, it is
pertinent to mention that TB kills more women than all
'Epidemiologist, "Sr. Medical Officer, ^Field Investigator
National TB Institute, 8, Bellary Road, Bangalore 560 003
3
other infectious diseases and maternal deaths combined9.
Among more than 900 million people in India today, every
second adult is infected with the tubercle bacilli and more
than 2 million people develop active TB each year. At any
point of time in our country there are an estimated 14-15
million TB patients, which is nearly one third of the global
burden of this disease. About 3 million of these patients
are highly infectious and spread the disease in the
community10.
There are few studies, if any, on the actual cost or
consequences of TB on the family, community and the
overall economy of our country. However, the special
burden on the society caused by TB is evident from the
age and sex distribution of this disease. Although
morbidity and mortality in any age group have significant
economic and social consequences, no community can
afford to lose its citizens in prime years of life which are
not only the productive years in terms of wage earning
but also a period of shouldering family and social
responsibilities. Further, the prevalence of the disease
increases with age, and in those over 35 years, the disease
is 2 to 3 times more common among males as compared to
females5. The economic consequences as a result of TB
are enormous since the contribution to the economy is
generally higher by older adults as compared to young
adults and by males as compared to females11. In a high
proportion of households, the patients suffering from TB
arc the sole breadwinners. Several gainfully employed
persons lose time from work due to their illness and many
even stop working12. The time off from work prior to
diagnosis and during treatment is an economic loss to the
families and the nation. Moreover, many of the caregivers
have to take time off from work to assist the patients. The
magnitude of the economic losses to the nation can be
gauged from the fact that TB is the singleJargesLcausg of
loss of Disability Adjusted Life Years (DALY’s) among
adults of the developing countries . In our country, TB
accounts for ^37% of total DALY’s lost because of
disability and premature death1 (One DALY is equal to
one lost year of healthy productive life).
The deaths in prime aged adults who are parents,
community leaders and breadwinners have a particularly
onerous burden and its consequences on children and
3
other dependants can be great. It is commonly observed
that when a mother dies, her children are more prone to
suffer from malnutrition, disease and death. This assumes
great significance in TB as the maximum prevalence of
disease in females occur in the age group of 20-35 years5.
Similar are consequences arising out of paternal deaths.
socio-demographic changes and increase in multi drug
resistant cases shall further exacerbate the problem. It is
estimated that the worldwide incidence of TB would
increase by more than 50% by the year 2005 and at least
30 million people would die of TB in the next 10 years2.
Consequently, TB has been declared as a global
emergency by the World Health Organization.
TB begets poverty and poverty begets TB
Most of the communicable diseases are associated with
poverty,
under-nourishment,
over-crowding
and
unhygienic living conditions. In the National Sample
Survey (1955-58) it was found that the prevalence of TB
in urban areas was higher among those living in ‘kutcha’
houses as compared to those in ‘pucca’ houses5. This
indicated a possible association between economic and
hygienic conditions and the prevalence of disease. Similar
observations have also been made in the only study
conducted in the country with the specific objective of
comparing the disease prevalence in different
socio-economic groups1^. In the same study, prevalence of
pulmonary TB was found to decrease with increase in
education level and per-capita family income. Nearly 70%
of the cases were accounted in non-workers and
agricultural workers13. Studies from other countries have
also shown that TB is concentrated in lower
socio-economic groups as these people are most
vulnerable to contract the disease. It has been observed
that the poorest households lose more time from work than
better off households through chronic diseases like TB.
Thus, the effects of the disease on patients can be
devastating both financially and emotionally.
Worsening of TB situation has been observed in recent
years all over the world. In 1970’s and 1980’s, TB was
ignored by much of the International health community
and priority was not accorded to the control or research of
this disease. This neglect of TB control programmes and
advent of HIV epidemic has already facilitated the return
of TB in wealthy nations4. In the developing countries
where the disease was never controlled the situation is
expected to worsen in future as a result of the increasing
HIV sero-prevalence rates since HIV infection is the
single most important risk factor for developing TB4. Due
to HIV/TB interface, there has been a resurgence of TB in
developed countries and the incidence of TB has been
recorded to have increased 3-4 times in some of the
African countries4. The impact of HIV epidemic on the
burden of TB is expected to be equally pronounced in our
country since about half of the population in the age group
susceptible for HIV infection is already infected with
tubercle bacilli. Thisupsurge in HIV related TB incidence
would make the economic burden of TB much greater in
addition to the human catastrophe. In addition, the
4
CONSTRAINTS IN TB CONTROL
The National Tuberculosis Programme (NTP) was
formulated by NTI after carrying out many path breaking
Operational Research studies in the field of TB. It has a
strong sociological basis as it was the first felt-need
programme based on scientific observations'4. Though
formulated in 1962 and subjected to periodical revisions,
the NTP, which is one of the largest disease control
programmes in our country, has not achieved the desired
results. Nearly half of the patients seek relief due to
suffering from the government health facilities14 and
others from private health agencies15. Among those
approaching general health facilities, only 16% are offered
sputum examination14. These patients incur sizeable
expenditure on general antibiotics, cost of transportation
and loss of wages before they are diagnosed as suffering
from TB. X- ray, as a case finding tool has limitations of
over diagnosing and is about 7-10 times costlier than
sputum microscopy which is also a more reliable
diagnostic tool17. The priority to sputum microscopy has
been given in the programme from its inception.
Nevertheless, there has been an overemphasis on using
X-rays by the physicians for diagnosing TB. This leads to
economic losses, which is augmented by overestimation
of cases and avoidable operational problems.
Under the NTP, antitubercular drugs are provided free
of cost to the patients. However, there is a perpetual
shortage of drugs in the government pharmacies while
there are adequate supplies of drugs in the market at high
prices. This leads to irregular and inadequate intake of
anti-TB drugs leading to treatment failure. At the same
time, the patients may have to incur the high cost of
procuring drugs at the market price. Even in a programme
offering free service, there are direct and indirect costs to
the patients, which may encourage drug defaulting in the
long run. The private health agencies, in addition to high
service charges, do not adhere to the standard drug
regimens15. This leads to financial losses for the patient
and increased possibility of drug resistance. In a recent
study by NTI, it was found that 50% of the TB patients
raised money during their illness by way of debt and
mortgage18. Economic insecurity also intensifies the
existing emotional problems like anxiety and fear about
the future.
Effective TB treatment not only cures prevalent cases
but also prevents future cases, which arc indirect benefits
of chemotherapy. When the treatment is not completed,
the disease has a greater likelihood of becoming multi
drug resistant. The cost of treating such patients is so
enormous that it is beyond the scope of any health
programme and the patient continues to infect others in the
community. Secondly, the efficacy of failure regimen is
poor.
One of the alternatives adopted to overcome the
problem of drug default has been the gradual replacement
of the 12 month long course treatment with a shorter and
more effective six month short course chemotherapy
(SCC). Though the cost of long course therapy is initially
less than the short course, it is ultimately higher when one
considers the cost incurred due to treatment failure. Since
SCC leads to better compliance resulting in higher cure
rates, the cost per death averted is lower when compared
to the long course treatment. Even though SCC has been
proven to be more cost-effective than long course regimen
both from provider perspective and household
perspective, it has so far been introduced only in 294
districts. Within these districts only a small proportion of
smear positive cases actually receive SCC drugs. Even
this situation is because of constraints of resources and not
according TB the priority it deserves.
Quality control is another area, which has not received
sufficient attention. India is poorly equipped in terms of
quality contr 1 laboratories. Studies reveal that the
bioavailability of Rifampicin varies from one
manufacturer to another and even from batch to batch.
Monitoring of drug quality is not easy in a country of
India’s size and more so because of involvement of the
large number of small pharmaceutical manufacturers in
addition to the major ones. Expensive drugs like
Rifampicin and Ethambutol have limited shelf lives in the
hot and humid weather conditions prevalent in India. Poor
packaging, lack of proper cold storage and transportation
facilities compounds the problem.
Globally speaking TB receives the least amount of
funding amongst all infectious diseases. In 1990, only 16
million dollars of foreign aid from all donors was available
to control the TB epidemic and this amounted to less than
one-tenth of all external aid to developing countries
—F-------~ —
~
••'’()
making TB the worlds most neglected health crisis" .
Though it has increased to 40 million dollars in 1996, still
the funds are grossly inadequate considering that only a
fraction of all health expenditure is directed towards the
process of controlling activities in the developing
countries.
SUGGESTED ACTIONS
Early diagnosis and treatment of TB is important.
Therefore, promoting the awareness of the disease through
health education and the importance of prescribing the
appropriate antitubercular drug regimens and preventing
treatment default cannot be over- emphasized.
One of the major determinants for successful cure of
TB cases is the level and intensity of supervision provided
by the health care delivery system. The approach that is
being gradually adopted by TB programmes all over the
world is to ensure that each dose is administered to the
patient under the supervision of a health worker or a
dedicated health volunteer. This strategy is called the
Directly Observed Treatment Short Course (DOTS) and
has yielded high cure rates of 85-90% in many countries
and in pilot areas of our country21,22. In 1993 after
Government of India, WHO review the programme has
been revised with DOTS as the main theme.
Implementation of RNTCP - Though implementation
of RNTCP involves additional expenditure for creating the
infrastructure for maintenance of the programme and
giving DOTS, it is the only way to ensure high cure rates.
The successful implementation of DOTS minimizes
deaths due to TB and averts hospitalization of patients
releasing many beds in hospitals that require large
resources to be maintained. Reduction in the prevalence of
TB among workers reduces their absenteeism on account
of poor health. There are additional savings due to lower
number of relapses and preventing development of
resistance to anti-TB drugs and in both of these situations
the treatment is much costlier. Reduction in transmission
of infection by rendering infectious cases non- infectious
also leads to future benefits. It has been estimated that for
every 1% of GDP spent on DOTS in India there will be a
return of 8% per annum11. The government of India has
sought World Bank assistance to extend DOTS to about
330 million people in the course of the next five years11.
However, it has so far been introduced in only a miniscule
of the population. DOTS should be implemented as early
as possible to rapidly cut down the chain of disease
transmission.
Increased financial support - The analysis of TB control
programmes in some African countries has shown that
treating smear positive TB costs around 20-57 dollars per
death averted23. The cost per discounted year of life saved
is less than US $ 10. There are few interventions that are
as cost effective as anti-TB treatment^. Thus, the worst
infectious killer of adults is also a very cost-effective
disease to combat and by curing even one person, the
disease can be prevented from spreading to dozens of
other people in the community.
5
Appropriate use of funds - Although few public health
expenditures provide so much value for so little money yet
many nations are spending more than half of their health
budget on expensive hospital services that by comparison
save very few lives. The role of Sanatoria and TB
hospitals is very limited and needs evaluation as it is
expensive to maintain beds for TB patients in developing
countries.
the modest infrastructure so that these programmes work
efficiently. The resources will also be needed to fund and
coordinate practical research projects to build the basis for
effective long-term TB control. Implementation of DOTS
strategy under RNTCP is the step in the right direction but
rapid expansion of the programme and long term financial
support is the need of the hour.
Deaths due to TB could only decline if increased
financial support is made available each year to TB
control programmes in developing countries. Additional
resources are also required to cater to the increasing
number of patients having HIV and TB as these patients
may also require expenses due to hospitalization.
ACKNOWLEDGEMENTS
Occupational resettlement - Some of the patients may
be physically incapacitated due to destroyed lung even il
the patient is cured and ruined financially during the course
of treatment. So, occupational resettlement of the
individual is of great importance. It is essential that
consideration be given to after care and rehabilitation,
taking into consideration the person’s ability to work. One
who was employed prior to his illness should go back to his
previous job provided it is not unsuitable for the altered
state of his health. Vocational training may be imparted to
others so that they can be suitably employed and do not
become a burden on the society.
REFERENCES
Research - Strengthening the operation research,
improving the functioning of the existing health care
systems and ropin in of NGOs to assist control
programmes arc other essential ingredients to successfully
combat the menace of TB.
CONCLUSION
Considering the enormous burden imposed by TB both
in terms of suffering and its socio- economic impact on our
country, TB control interventions must be intensified with
provision of all necessary support. Even though the health
systems in our country is given lower priority and budget,
there is sufficient justification for enhancing investments
in TB control. Since TB control activities are also one of
the most cost-effective interventions, appropriate actions
aimed at reducing the enormous burden of TB must be
accorded the highest priority. For this a strong political will
and advocacy is required to appreciate the enormity of the
problems due to TB and to allocate appropriate budgets for
TB control programmes. Enhanced finances are needed to
enable the TB programmes to undertake training
programmes, improve registration systems and monitoring
tools, to finance medicines, microscopes and to improve
6
The authors are indebted to Dr (Mrs) P Jagota, Director,
National TB Institute, Bangalore for her encouragement in
writing this article and to Mrs Vijayalakshmi B for
secretarial assistance.
1. World development report, 1993: Investing in health:
The global burden of disease in 1990, 213.
2. Dollin RJ. Raviglionc MC and Kochi A: A review of
current epidemiological data and estimation of future
tuberculosis incidence and mortality; WHO/TB/93,
173.
3. World Health Organization: Tuberculosis Advocacy:
Apractical guide: 1999, WHO/TB/98.239.
4. Chadha VK: Global trends of tuberculosis - An
epidemiological review: NTI Bulletin 1997, 33,
11-18.
5. Tuberculosis in India - A National Sample Survey,
1955-58, special report series No.34, Indian Council
of Medical Research, New Delhi.
6. Raj Narain, Geser A, Jambunathan MV &
Subramanian M: Tuberculosis prevalence survey in
Tumkur district: Indian J TB 1963,10, 85- 116.
7. National
Tuberculosis
Institute,
Bangalore:
Tuberculosis in a rural population of south India - A
five year epidemiological study: Bull WHO 1974,51,
473-88.
8. Murray CJI, Styblo K & Rouillon A: Tuberculosis in
developing countries: burden, intervention and cost:
Bull IUAT & LD, March 1990, 65, 2-19.
9. Maire Connolly & Paul Nunn: Women and
Tuberculosis; WHO Stat Qtly 1996, 49,115-119.
10. Tuberculosis control in India, Developing role of
NGOs; ACTIONAID INDIA 1996,10, 9.
11. World Health Organization: The potential economic
benefits of the DOTS strategy against tuberculosis in
India 1996, WHO/TB/96.218.
12. Needham DM, Godfrey, Faussett P & Foster SD:
Barriers to tuberculosis control in urban Zambia, Int J
TB & Lung Dis, 1998, 2, 811-817.
13. Field Trial of short-term intermittent chemotherapy
against tuberculosis - an ICMR project. Project Report
1989. Department of community medicine and
Department of microbiology, Mahatma Gandhi
Institute of medical sciences, Sevagram Wardha.
14. Banerji D & Andersen S: A sociological study of the
awareness of symptoms among persons with
pulmonary tuberculosis. Bull WHO 1963, 29,
665-683.
15. Upleker M & Sheela Rangan: Tackling tuberculosis The search for solutions; Bombay: Foundations for
research in Community Health, 1996.
16. Jagota
P,
Mahadev
B,
Srikantaramu
N,
Balasangameshwara VH & Sreenivas TR: Case
finding in District Tuberculosis Programme: potential
and performance: Indian J TB 1998, 45,39-46.
18. Sophia
Vijay,
Balasangameshwara
VH
&
Srikantaramu N: Treatment dynamics and profile ol
tuberculosis patients under the District TB
Programme (DTP)-A Prospective cohort study,
Indian J TB 1999, 46, 239-250.
19. National Tuberculosis Institute: Performance of the
National TB Programme for the year 1996: NTI
Bulletin 1997,33/19-20.
20. World Health Organization - Report on the
tuberculosis epidemic, 1997: WHO/TB/97.,224.
21. World Health
Organization:
Treatment oi
tuberculosis: Guidelines for national programme,
Ed-2, 1997, Geneva, WHO/TB/97, 220.
22. Khatri GR: The Revised National Tuberculosis
Control Programme: A status report on first 1,00,000
patients: Indian J TB 1999,46,157.
23. Murray C, Styblo K & Annik Rouillon: Tuberculosis
disease control priorities in developing countries:
London; Oxford University Press, 1993, p 233-259.
17. Naganathan N, Padmanabha Rao K & Rajalakshmi
R: Cost of establishing and operating a tuberculosis
bacteriological laboratory: Indian J TB 1974. 21.
HEALTH CARE REFORM
.......... clear statements of policy objectives in health are often lacking, and governments
have an important responsibility in this area. The roles of the principal actors in health care
delivery and consumption are in a state of flux. Government responsibilities are being
scaled down, but public regulation is essential. Patients, rather than administrative units,
are being more commonly regarded as the proper focus for planning resource utilization.
New methods of paying providers have to accommodate cost-containment as well as the
professional needs of providers and equitable access for patients.
A transfer of responsibility to private financing sources exaccerbates inequities, and on its
own will do nothing to improve performance in the public sector. Public sector
improvements are being explored through decentralization of fund holding, contracted
purchasing,overall budget control mechanisms and in some cases a clearer
purchaser-provider separation. In choosing the principal policy instruments, clear policy
objectives are necessary. Formulating these is a government task, and cannot be privatized.
Source : Crease A: Global trends in health care reform:
Wld Hlth Forum 1994, 15/4, 317-322.
7
.
INT J TUBERC LUNG DIS 4(3):189-190
, © 2000 IUATLD
EDITORIAL
Good news from India
ALL THOSE CONCERNED with the grim global
problem of tuberculosis will be greatly encouraged to
read the report on ‘The status and prospects of tuber
culosis control in India’ by Khatri and Frieden in the
present issue of the Journal.1 India, with its vast pop
ulation and diversity, has long been regarded as one
of the greatest global challenges for tuberculosis con
trol. It is thought to harbour nearly 30% of the
world’s tuberculosis.2 A national sample survey in
the mid-1950s documented the high prevalence; ’ as a
result, a National Tuberculosis Programme was initi
ated in 1962, but over the years encountered many
difficulties in implementation.4-5 Reviews revealed
treatment completion rates of less than 30%,6 and
little effect on incidence.7
Following these reviews, a Revised National Tuber
culosis Programme (RNTP), based on the WHO
‘DOTS’ programmes,8 was launched in late 1993.
Starting with pilot projects in five sites in different
States covering 2.5 million population, the programme
expanded to cover 130 million by mid-1999. Alto
gether 146 000 patients had been put on treatment.
Results of treatment have been most encouraging.
In smear-positive new patients, cure rates rose from
70.7% in 1993 to 83.9% in 1998, with an overall
average of 79.4%. The overall failure rate was 3.5%,
mortality only 3.7% and default rate 8.7%. Among
those who had remained smear-positive after earlier
treatment and received retreatment (Category II),
63% were cured and 3% completed treatment; only
8% remained smear-positive at the end of treatment.
This suggests that most of the ‘failures’ of initial treat
ment were not due to multiple drug resistance (MDR),
but to failure to observe treatment meticulously.
This has been an impressive start, but of course
many challenges lie ahead. By late 1998 the RNTP
covered only 2% of India’s population. At that stage
large scale expansion was initiated, aiming at cover
ing a quarter of the whole population by early 2000.
This must be putting great strain on the organisation
of new resources, including training.^: had been rec
ognised that the quality of supervision, national, State
and local, would be crucial. Supervisors must have
the status and training to stimulate, encourage and
monitor the work. This has now been initiated.
Two formidable difficulties lie ahead. Most pa
tients with health problems in India first consult pri
vate practitioners, many of whom may not even be
Western trained. Treatment is often chaotic and illsustained.10 There must be resultant risks of MDR
development, though this has not yet been systemati
cally reviewed. There is obvious need for extensive
undergraduate and post-graduate training, and for
schemes to develop cooperation between private
practitioners and the national programme.I()’11 HIV
infection is rapidly expanding in India. Though many
patients are still in the incubation phase, there is a
real threat of a future explosion of case load of tuber
culosis, perhaps with a grim component of MDR. It
may be a race against time to prevent, or at least mit
igate, MDR TB by widespread good treatment.
The clear global threat of HIV, and the as yet incom
pletely defined threat of MDR,12 still fully justifies
WHO’s 1993 claim of tuberculosis as a global emer
gency. It becomes even more urgent to increase the
number of countries fully implementing the DOTS pro
gramme. The addition of a ‘DOTS Plus’ programme to
deal with MDR is at present under discussion.1 ’
Have these excellent initial results of the revised
Indian national programme any lessons for the rest of
the world? One of the problems causing concern is
the increasing advocacy of health sector reform,
whereby responsibility and budgets are delegated to
peripheral areas.14 In many developing countries
peripheral areas are thought to have been neglected
by the centre. There are obvious theoretical advan
tages in the proposed reform, but certain disadvantages.
These include loss of the economies of scale in such
matters as drug purchase. Not all peripheral areas
may give tuberculosis the degree of priority it deserves.
Resultant high prevalence in one area can soon spill
over into others. There may be loss of central leader
ship and expertise. In India much responsibility for
tuberculosis is being decentralised to local ‘coali
tions’ which have to fulfill certain conditions to
receive free drug supplies from the centre. The centre
also provides much technical support and leader
ship. Establishing coalitions is proving more com
plex in big cities where many diverse bodies tend to
be involved. Experience in India may prove of value
to other countries.
Indonesia and the Philippines are two other coun
tries with large and diverse populations, and an
immense tuberculosis problem. They are only in the
earliest stages of establishing a practicable national
programme, and could benefit from some of the lessons
being learnt in India. Although the DOTS programme,
aided by a substantial loan from the World Bank, has
been brilliantly successful in about half of China,15 it
is still uncertain whether the costs will be taken over
by the Chinese Government when the loan runs out in
2001. There is a danger that the tendency towards
health sector reform may lead to exclusively periph
eral responsibility and to patients having to pay for
their treatment. The USA, the richest country in the
world, has demonstrated in New York the sort of
♦
190
The International Journal of Tuberculosis and Lung Disease
disaster that can occur when a previously successful
service is subsequently neglected.16
John Crofton
13 Spylaw Bank Road
Colinton
Edinburgh EH 13 OJW
References
1 Khatri G R, Frieden T R. The status and prospects of tubercu
losis control in India. Int J Tuberc Lung Dis 2000; 4: 193-200.
2 World Health Organization. Report on the tuberculosis epi
demic, 1998. WHO/TB/98.237. Geneva: WHO, 1998.
3 Indian Council of Medical Research. Tuberculosis in India. A
sample survey 1955-58. Special report series No. 34. ICMR:
New Delhi, 1-21.
4 Pathania V, Almeida J, Nunn P, Kochi A. The behaviour and
interaction of TB patients and private for-profit health care
providers in India: a review. Geneva: WHO, undated.
5 Ogden j, Rangan S, Lewin S. Tuberculosis control in India. A
state-of-the-art review. London: The Foundation for Research
in Community Health, Bombay; London School of Hygiene
and Tropical Medicine, 1999.
6 World Health Organization. Tuberculosis programme review,
India. Geneva: WHO, 1992.
World Health Organization. Prevalence and incidence of tu
berculosis in India. A comprehensive review, 1997. WHO/TB/
97.231. Geneva: WHO, 1998.
8 Maher D, Chaulct P, Spinaci S, Harries A. Treatment of tuber
culosis: Guidelines for national programmes. 2nd ed. Geneva:
WHO, 1997.
9 Uplekar M W, Shepard D S. Treatment of tuberculosis by pri
vate practitioners in India. Tubercle 1991; 72: 284-290.
10 Uplekar M. Involving the private medical sector in tuberculosis
control: practical aspects. In: Porter J D H, Grange J M, eds.
Tuberculosis. An interdisciplinary perspective. London: Impe
rial College Press, 1999.
11 Brugha R, Zwi A B. Tuberculosis treatment in the public and
private sectors—potential for collaboration. In: Porter J D H.
Grange J M, eds. Tuberculosis. An interdisciplinary perspec
tive. London: Imperial College Press, 1999.
12 World Health Organization/lnternational Union against Tuber
culosis and Lung Disease. Anti-tuberculosis drug resistance in the
world. The WHO/IUATLD global project on anti-tuberculosis
drug resistance surveillance. Geneva: WHO, 1997.
13 Espinal M A. Multidrug resistant tuberculosis. MDR TB. Basis
for the development of an evidence-based case-management
strategy for MDR TB within the WHO's DOTS strategy. Pro
ceedings of 1998 meetings and protocol recommendations.
Geneva: WHO, 1999.
14 Tayler E. Tuberculosis and health sector reform. In: Porter J D H,
Grange J M, eds. Tuberculosis. An interdisciplinary perspec
tive. London: Imperial College Press, 1999.
15 World Health Organization. Global tuberculosis control.
WHO report 1999. Geneva: Communicable Diseases, WHO,
1999.
16 Frieden T, Fujiwara P I, Washko R M, Hamburg M A. Tuber
culosis in New York City—turning the tide. New Engl J Med
1995; 333: 229-233.
*
INT J TUBERC LUNG DIS 4(3):193-200
© 2000 IUATLD
The status and prospects of tuberculosis control in India
G. R. Khatri,* T. R. Friedenh
* Directorate General of Health Services, Ministry of Health and Family Welfare, New Delhi,
f Regional Office for South-East Asia, World Health Organization, New Delhi, India
------------------------------------------------------------------ SUMMARY
»•
SETTING: India, where much of the global strategy for
tuberculosis control was established, but where, every
year, there are an estimated 2 million cases of tuberculosis.
OBJECTIVE: To describe the policies, initial result, and
lessons learned from implementation of a Revised National
Tuberculosis Control Programme using the principles of
DOTS (Directly Observed Treatment, Short-course).
DESIGN: A Revised National Tuberculosis Control Pro
gramme (RNTCP) was designed and implemented start
ing in 1993. With funding from the Government of India,
State Governments, the World Bank and bilateral donors,
regular supply of drugs and logistics was ensured. Persons
with chest symptoms who attend health faeft^s are
referred to microscopy centres for diagnosis. Diagnosed
cases are categorized as per World Health Organization
guidelines, and treatment is given by direct observation.
Systematic recording and cohort reporting is done.
RESULTS: From October 1993 through mid-1999,
146 012 patients were put on treatment in the pro
gramme. The quality of diagnosis was improved, with
the ratio of smear-positive to smear-negative patients
being maintained at 1:1. Case detection rates varied
greatly between project sites and correlated with the per
centage of patients who were smear-positive among
those examined for diagnosis, suggesting heterogeneous
disease rates. Treatment success was achieved in 81% of
new smear-positive patients, 82% of new smear-negative
patients, 89% of patients with extra-pulmonary tuber
culosis, and 70% of re-treatment patients.
CONCLUSION: The RNTCP has successfully treated
approximately 80% of patients in 20 districts of 15
states of India. Treatment success rates are more than
double and death rates are less than a seventh those of
the previous programme. Starting in late 1998, the
programme began to scale up and now covers more
than 130 million people. Maintaining the quality of
implementation during the expansion phase is the next
challenge.
KEY WORDS: tuberculosis; India; direct observation of
treatment; short-course chemotherapy; microscopy;
supervision
MANY OF THE PRINCIPLES of tuberculosis con
trol were established in India, but these principles were
not widely applied within the country until recently.
In the mid-1950s, India conducted a national survey
of tuberculosis prevalence which documented the
enormous burden of suffering caused by the disease
and indicated the need for political commitment to
its control.1 The Tuberculosis Chemotherapy Cen
tre in Madras (now the Tuberculosis Research Cen
tre, Chennai) first documented the efficacy and safety
of domiciliary treatment of tuberculosis,2 the neces
sity and feasibility of treatment supervision in rhe
community—now called directly observed treat
ment,’ and the efficacy of intermittent chemotherapy
for tuberculosis as a means to simplify treatment obser
vation for patients and providers.4 The National
Tuberculosis Institute (NTI), Bangalore, demonstrated
in the early 1960s that even with limited health ser
vices most tuberculosis patients seek care at health
facilities, indicating that active case finding is not nec
essary.-' The NTI also demonstrated that technicians
in the periphery can perform sputum smear microscopy effectively if they are given minimal training but
regular supervision, indicating the feasibility of the
use of sputum microscopy as the primary tool for
diagnosis of tuberculosis.6 Hence many of the princi
ples which are now globally recommended as the
Directly Observed Treatment, Short-course (DOTS)
strategy7 were documented in India.
India has nearly 30% of the global burden of tuber
culosis disease.8 With an estimated incidence of new
smear-positive tuberculosis of 85 per 100 000 popu
lation, there are nearly 1 million new smear-positive
cases of tuberculosis a year and about 2 million total
new cases. A National Tuberculosis Programme has
been implemented since 1962 and has established
tuberculosis centres in more than 440 of the more
than 520 districts in India. In addition, there are 330
Correspondence to: Dr G R Khatri, Deputy Director General (Tuberculosis), Directorate General of Health Services Min
istry of Health and Family Welfare, Ninnan Bhavan 523C, New Delhi 110 01 1, India. Fax: ( + 91) I 1-301-8126 J-mailddgtb@nb.nic.in
“
*■ *
Article submitted 25 May 1999. Final version accepted 5 November 1999.
194
The International Journal of Tuberculosis and Lung Disease
chest clinics, located mostly in urban areas, and more
than 47 000 beds exclusively for TB patients. In spite
of the fact that the programme had been in operation
for three decades, no significant epidemiological impact
on disease prevalence was observed.9 With this back
ground, in 1992, the Government of India, together
with the World Health Organization (WHO) and
Swedish International Development Agency (SIDA),
reviewed the national programme and concluded that
it suffered from managerial weakness, inadequate
funding, over-reliance on X-ray, non-standard treat
ment regimens, low rates of treatment completion,
and lack of systematic information on treatment out
comes (Government of India/WHO, unpublished). As
a result, a Revised National Tuberculosis Control
Programme (RNTCP) was designed with input from
the Government of India, State governments in India,
the World Bank, and bilateral donors. In this article
we describe the evolution, experience, and results of
this programme, and the prospects and challenges for
the future.
METHODS
As of 1999, India had a population of more than 980
million people in 32 States/Union Territories. The
country has an infant mortality rate of 72 per 1000
live births, ranging from 78 in rural areas to 46 in
urban areas. Between States infant mortality rates
range from 13 to 97 per 100 000 live births.10 There
are 19 official languages in the country.
Starting in October 1993, the RNTCP was imple
mented in a population of 2.35 million in five sites in
different states of India (Delhi, Kerala, West Bengal,
Maharashtra, and Gujarat). The programme was
expanded to a population of 13.85 million in 1995
and 20 million in 1996. Rapid scale-up began in late
1998, when another 100 million population were
covered under RNTCP.
Table 1
The basic principles of the RNTCP are: 1) political
commitment to ensure adequate funds, staff, and
other key input; 2) diagnosis primarily by microscopy
of patients presenting to health facilities; 3) regular
and uninterrupted supply of anti-tuberculosis medi
cations including the use of a patient-wise box which
contains the entire course of treatment for an individ
ual patient; 4) direct observation of every dose of
treatment in the intensive phase and at least the first
of three doses each week in the continuation phase of
treatment; 5) systematic monitoring, supervision and
cohort analysis; one additional paramedical staff for
organization of uninterrupted treatment and one for
ensuring quality laboratory service for every 500 000
population. Additional staff are provided for difficult
mountainous and tribal areas.
The diagnostic algorithm recommended by the
WHO11 is used, except that X-ray is taken if all three
acid-fast bacilli (AFB) smears are negative and there is
no response to 1-2 weeks of antibiotics, without
waiting for three more smears to be done. If only one
of three specimens is positive, an X-ray is taken and
the patient is evaluated. Treatment regimens are as
given in Table 1. Patients treated in Categories I and II
whose smears are positive at the end of the intensive
phase receive another month of intensive phase treat
ment. All treatment is given thrice weekly on alter
nate days. During the intensive phase, every dose is
directly observed; medications for the continuation
phase are packaged into a weekly blister pack, and at
least the first dose each week is directly observed.
Funding is available based on a 5-year ‘soft’ loan
of US$142 million from the World Bank. The loan is
structured as a ‘time slice’ so that an increasing pro
portion of the costs are to be borne by the State and
National governments of India. Policy direction, super
vision, drugs and microscopes are provided by the
central government. The districts, each of which has
an average population of about 2 million, have a Dis-
Treatment regimens used in the Revised National Tuberculosis Control Programme, India
Category of
treatment
______________ Type of patient______________
__________ Regimen*
Category I
New sputum smear-positive
Seriously ill sputum smear-negative
Seriously ill extra-pulmonary +
2H3R3Z3E3/4H3R3
Category II
Sputum smear-positive relapse*
Sputum smear-positive failure*
Sputum smear-positive treatment after default
2H3R3Z3E3S3/IH3R3Z3E3/5H3R3E3
Category III
Sputum smear-negative, not seriously ill
Extra-pulmonary, not seriously ill
2H3R3Z3/4H3R3
* The number before the letters refers to the number of months of treatment. The subscript after the letters refers to the number of doses per week. Patients
who weigh more than 60 kg receive additional rifampicin 150 mg. Patients over 50 years old receive streptomycin 500 mg. Patients in Categories I and II who
have a positive sputum smear at the end of the initial intensive phase receive an additional month of intensive phase treatment.
* Examples of seriously ill patients are those suffering from meningitis, disseminated TB, tuberculosis pericarditis, peritonitis, bilateral or extensive pleurisy, spinal
TB with neurological complications, smear-negative pulmonary TB with extensive parenchymal involvement, intestinal and genitourinary TB.
* In rare and exceptional cases, patients who are sputum smear-negative or who have extra-pulmonary disease can have relapse or failure. This diagnosis should be
supported by culture or histological evidence of active tuberculosis. In these cases, the patient should be registered as 'Other' and given Category II treatment.
H = isoniazid (600 mg); R’ = rifampicin (450 mg); Z = pyrazinamide (1500 mg); E = ethambutol (1200 mg); S = streptomycin (750 mg).
The status and prospects of tuberculosis control in India
trict Tuberculosis Control Society which receives funds
directly from the Central Government out of the
World Bank assistance and which can hire contrac
tual staff, purchase necessary items, and perform
other functions more efficiently than through the usual
government procedures. State governments provide
the health infrastructure and staff. In rural areas,
India has an established health infrastructure, with
LAfc larger health centres for each 100 000 population and
smaller centres for each 30 000 population.
Printed technical and operational guidelines have
been provided to all health centres, laboratory guide
lines have been provided to all microscopy centres,
and illustrated guidelines for health providers giving
observed treatment have been provided in local lan
guages. Modular training has been used for all staff,
from medical officers to health workers. Smaller cen
tres are assisted by sub-centres each covering a popu
lation of 5000 and staffed by a male and a female
health worker. These individuals are responsible for
treatment observation; where they are not available,
treatment observation is done by community volun
teers including child survival workers, traditional
midwives, and community and religious leaders. Ob
servation by a family member is not acceptable in the
programme. In urban areas the institutional arrange
ments are more variable. In some larger cities with
limited health infrastructure, the RNTCP has funded
specialised full-time staff for microscopy and for
treatment obser. tion. In these urban areas, approximately one Iab( . atory technician and one treatment
observer (TB HeaFh Visitor) are funded and hired to
provide services for each 100 000 population cov
ered. In some rural areas with large numbers of va
cancies, laboratory technicians have been funded by
the programme as a temporary measure.
Outcomes are analysed and presented as per stan
dard definitions.11 Only patients who were started on
treatment were registered; outcomes are presented
only for evaluated patients. Whereas virtually all
(99.4%) new smear-positive patients registered have
been evaluated, in the early years of the programme
some patients with smear-negative and extra-pulmo
nary tuberculosis who were placed on treatment were
not evaluated. This situation improved in 1996, when
more than 90% of all patients’ outcomes were evalu
ated, and improved further in 1997, when more than
99% of all patients’ outcomes were evaluated.
RESULTS
From 1993 through mid-1999, more than 146 000
patients were put on treatment in the RNTCP. In
terms of case detection activities, detailed informa
tion on microscopy activities is available for 1998; of
145 906 patients examined for diagnosis, including
both new and previously diagnosed patients, and
including some patients living outside the areas cov-
195 *
ered, 19 698 (13.5%) were found to be sputum
smear-positive.
Rates of sputum positivity have been quite consis
tent within individual project sites, but vary markedly
in different project areas. For example, positivity
rates ranged from 3-4% (average 3%) over five quar
ters in Pathanamthitta District of Kerala, which had
the lowest proportion, to 15-25% (average 21%)
over six quarters in Mehsana District of Gujarat,
which had the highest proportion positive of all dis
tricts in 1998. Project areas with a higher proportion
of patients with chest symptoms who had positive
smears also had higher total and smear-positive detec
tion rates (e.g., 33 vs 99 per 100 000 per year in
Pathanamthitta and Mehsana, respectively). This re
lationship is an increase of approximately 20 cases/
100 000 population/year in detection of new and re
treatment smear-positive cases for every 5% increase
in sputum positivity rates, beginning at 30 smear-pos
itive patients per 100 000 population per year at a
positivity rate of 5% (Figure 1, r2 = 0.8). There is an
analogous, although weaker, inverse correlation be
tween median age and detection rate (r2 = 0.4); areas
with a higher median age have lower detection rates,
with a relationship of a decrease of about 30 new
smear-positive cases per year per 100 000 population
for every 5-year increase in median age.
Quality of diagnosis has improved, as indicated by
a ratio of smear-positive to smear-negative patients of
approximately 1:1 (Table 2). The guidelines to put
only seriously ill patients with smear-negative and
extra-pulmonary tuberculosis on Category I treat
ment have generally been followed: in 1998, of 14477
patients with smear-negative or extra-pulmonary
tuberculosis for whom information is available, 2034
(14%) were placed on Category I treatment. Detection rates of new smear-positive patients show sub
stantial variability between sites (range, for example,
Annualised from the second quarter of 1999, from a
low of 60 total and 24 new smear-positive cases per.
100 000 population in Kannoor District of Kerala, to
269 total and 104 new smear-positive cases per
100 000 in Jaipur district of Rajasthan). Different
rates may reflect varying incidence, but detection
rates in virtually all areas have increased over time,
particularly after widespread expansion. For all
RNTCP’ areas, rates increased from 87 total and 36
new smear-positive cases per 100 000 to 120 total
and 47 new smear-positive cases per 100 000 popula
tion, annualised, between the first and second quar^ters of 1999.
There have been large differences in the rates of
smear-positive tuberculosis for males and females.
While the ratio of male to female newjni£a£j?psitiye
cases is 2:1, the male to female ratio among extrapulmonary tuberculosis is 1:1.4. Among smear-positive
patients, the male:female differences increase mark
edly with age (Figure 2).
f
I
'j
i.
r
I
fI
!■
f-
-?
CJJ?”
I
J:
196
The International Journal of Tuberculosis and Lung Disease
100
90
S 80
Q. 70
o
o
o
o
60
2 50
M
o(0
$ 40
I? 30
ISQ>
W 20
10
oL
o
5
20
15
10
25
Positivity among patients examined for diagnosis
Figure 1 Correlation between per cent of patients with chest symptoms who are AFB smear
positive and smear-positive detection rate per 100 000, Revised National Tuberculosis Control
Programme, India, 1993-1998. For every 5% increase in the percentage of patients who are spu
tum smear-positive, there is an increase of approximately 20 smear-positive patients per 100 000
population, starting at 34/100 000 with a positivity rate of 5%. Two project areas with small total
populations and two areas with a large floating population have been excluded.
Treatment outcomes have been consistently good,
with about eight out of 10 patients being successfully
treated (Table 3). Treatment success has been high
among all types of patients and in all but one of the
20 areas implementing the RNTCP (range in 19
areas, 72% to 90%). Treatment success has increased
for all types of patients between 1995 and the first
two quarters of 1998, the most recent period for
which information is available. In the ‘other retreat
ment’ category, the proportion of smear-negative
patients has increased, resulting in an increased pro
portion of this type of patient being reported as hav
ing completed treatment rather than being cured.
Among the 1458 patients who had remained smear
positive after earlier treatment and who were treated
with the Category II regimen, 63% were cured, 3%
completed treatment, 8% died during treatment,
15% interrupted treatment, 8% remained smear-
positive at the end of treatment, and 2% were trans
ferred to other jurisdictions.
DISCUSSION
Since 1993, India has successfully implemented a TB
control programme using principles for diagnosis and
treatment which were documented in India, and
which are now recommended by the WHO as DOTS.
Because of India’s diversity, inadequate resources,
huge population, and problems of implementation,
ensuring effective tuberculosis control is not easy.
Effective diagnostic policies have been difficult to
implement because of long-standing over-reliance on
X-ray as the primary means of diagnosis. This prac
tice has continued in the former programme despite
the documentation that 50-70% of smear-negative
patients placed on treatment did not have tuberculo-
Table 2 Patients placed on treatment under the Revised National Tuberculosis Control
Programme, India, 1993-1999 (first half)
Year
1993
1994
1995
1996
1997
1998
1999 (Ist half)
Total
New
smear
positive
New
smearnegative
Extrapulmonary
Relapse
smearpositive
Other
retreatment
cases
Total
patients
392
1 060
2 144
6 365
7 747
12 354
24 959
55 021
603
1 179
1 945
6198
7129
11 268
20 020
48 342
13
288
606
1 814
2 186
4015
7426
16 348
44
410
679
868
1 035
1 937
3 487
8460
2
311
669
1 724
2 644
5 687
6 804
17 841
1 054
3 248
6 043
16 969
20 741
35261
62 696
146012
The status and prospects of tuberculosis control in India
197
6000
1 : 0.5
5000
1 : 0.7
1 : 0.3
4000
1 : 0.3
3000
1 : 0.2
2000
II
1 : 0.2
1000
0
Figure 2
1 : 2.1
MF
0-14
MF
M F
15-24
15-24
MF
25-34
M F
35-44
M F
45-54
M F
M F
55-64 65 OR MORE
Cases by age and sex, Revised National Tuberculosis Control Programme, India, 1993-
1998.
sis at all.12 Close monitoring of the ratio of positive to
negative patients diagnosed, and intensive supervi
sion to ensure th'at three AFB smears are correctly
performed and a trial of antibiotics is given prior to
X-ray and treatment of smear-negative tuberculosis,
have been essential. In quite a few areas, ensuring
accurate history-taking has also required intensive
supervision; initially some areas placed previouslytreated patients on Category I regimens, which resulted
in higher rates of relapse (Ministry of Health and Fam
ily Welfare, unpul shed data, January, 1999).
Wide variations (5-32%) in the proportion of
smear-positive cases among patients examined for
diagnosis have been reported earlier.13 The correla
tion seen between detection rate and smear positivity
rate, if confirmed in other areas, may provide indirect
evidence to help evaluate the proportion of all cases
that are detected in the programme by providing a
rough estimate of the number of total cases. Specifi
cally, if confirmed and if taken in conjunction with
other data, this ratio, by providing a rough estimate
of total cases, might help provide a denominator to
assess whether a programme is meeting the global tar| get of detection of 70% of new smear-positive cases?
Similarly, trends in the proportion of patients found
to be symptomatic among those examined could pro
vide data on trends of disease, but only if the rate of
sputum examination is constant or if changes in this
rate are taken into account.
In some areas, there have been concerns that
patients are being selected to be placed on the
RNTCP. This situation has arisen primarily, although
not entirely, because of the partial geographic cover
age of the programme, as some of the patients who
attend RNTCP facilities for diagnosis may live out
side of RNTCP areas of coverage. The goal is that,
among diagnosed patients living in RNTCP areas, at
least 90% should be treated under the RNTCP.
Patients who are unable or unwilling to participate in
the programme are to receive conventional treatment
with isoniazid, streptomycin, and ethambutol. Until
recently, the reporting system did not allow accurate
determination of the number of patients living in the
covered area who were diagnosed but not started on
RNTCP treatment. The reporting formats have
recently been changed so that this important indica
tor can be monitored in the future.
To maintain high rates of sputum conversion and
cure has also required frequent supervision. Evalua
tion of areas with poor treatment outcomes has
nearly always revealed that the primary problem has
been failure to ensure direct observation of treatment
at a time and place convenient to patients.
Although the Category II regimen has been contro
versial, even among patients who remained smear
positive after 5 months of Category I treatment and
who were then placed on Category II treatment, two
thirds were successfully treated and only 8% re
mained smear-positive after 5 months of Category II
treatment.
In this initial phase, the results of India’s RNTCP
have been comparable to those of other countries
implementing DOTS. Case detection rates are higher
than those reported from Bangladesh14 and China,15
while rates of successful treatment are comparable to
those from Africa and South-East Asia and lower
than those from Peru, China, and Vietnam.8
The high rates of treatment success and low rates
of death among patients treated are particularly
remarkable when compared with the previous pro
gramme. A systematic analysis of the outcomes in
smear-positive patients treated with short-course, non
observed chemotherapy in a relatively well-functioning
district revealed success rates of only 40% (compared
with 80% in the current programme) and death rates
of 29% (compared with 4% in the current pro
gramme).16’17 The population coverage of the RNTCP
until late 1998 was only 2% of the total population
b
i
C
■
198
Table 3
The International Journal of Tuberculosis and Lung Disease
Outcomes of treatment under the Revised National Tuberculosis Control Programme, India, 1993-1999 (lirst half)
Cured
n (%)
Completed
treatment
n (%)
Died
n (%)
Remained
smear
positive
(failed)
n (%)
277 (70.7)
860 (81.3)
1 606 (75.1)
4 855 (76.5)
6176 (80.5)
3 910(83.9)
17 684 (79.4)
17(4.3)
9 (.9)
68 (3.2)
170 (2.7)
108(1.4)
47(1.0)
419(1.9)
32 (8.2)
41 (3.9)
71 (3.3)
228 (3.6)
270 (3.5)
172 (3.7)
814(3.7)
2 (.5)
32 (3.0)
77 (3.6)
262 (4.1)
259 (3.4)
139 (3.0)
771 (3.5)
43(11.0)
68 (6.4)
167 (7.8)
649(10.2)
668 (8.7)
336(7.2)
1931 (8.7)
21 (5.4)
48 (4.5)
149 (7.0)
181 (2.9)
191 (2.5)
56(1.2)
646 (2.9)
New smear-negative patients evaluated
1993
603
1994
651
1995
1 264
1996
5 538
1997
7 099
1998 (1st half)
4451
Total
19 606
327 (54.2)
483 (74.2)
947 (74.9)
4379 (79.1)
5 990 (84.4)
3 894 (87.5)
16 020 (81.7)
28 (4.6)
15(2.3)
45 (3.6)
181 (3.3)
207 (2.9)
145 (3.3)
621 (3.2)
10(1.7)
7(1.1)
11 (.9)
88(1.6)
115(1.6)
66(1.5)
297(1.5)
215(35.7)
132 (20.3)
201 (15.9)
725 (13.1)
691 (9.7)
310(7.0)
2 274(11.6)
23(3.8)
14(2.2)
60 (4.7)
165 (3.0)
96(1.4)
36 (.8)
394 (2.0)
New extra-pulmonary patients evaluated
1993
13
1994
169
1995
467
1996
1 669
1997
2 181
1 550
1998 (1st half)
Total
6049
8(61.5)
156 (92.3)
402 (86.1)
1 438 (86.2)
1951 (89.5)
1 427 (92.1)
5 382 (89.0)
0(0)
2(1.2)
11 (2.4)
23(1.4)
47 (2.2)
25(1.6)
108(1.8)
0(0)
0(0)
0(0)
6 (0.4)
2(0.1)
5 (0.3)
13 (.2)
5 (38.5)
7(4.1)
27 (5.8)
163 (9.8)
155(7.1)
79 (5.1)
436 (7.2)
0(0)
4(2.4)
27 (5.8)
39 (2.3)
26(1.2)
14 (.9)
110(1.8)
Relapsed smear-positive patients evaluated
44
1993
23 (52.3)
1994
376
284 (75.5)
1995
433
257 (59.4)
1996
838
521 (62.2)
1997
1 024
714(69.7)
1998 (1st half)
687
499 (72.6)
3402
Total
2 298 (67.5)
3 (6.8)
4(1.1)
32 (7.4)
24 (2.9)
30 (2.9)
16(2.3)
109 (3.2)
4(9.1)
17(4.5)
22 (5.1)
36 (4.3)
72 (7.0)
41 (6.0)
192 (5.6)
1 (2.3)
11 (2.9)
15(3.5)
92 (11.0)
57 (5.6)
40 (5.8)
216(6.3)
10(22.7)
21 (5.6)
72(16.6)
117(14.0)
118(11.5)
76(11.1)
414(12.2)
3 (6.8)
39(10.4)
35(8.1)
48 (5.7)
32 (3.1)
15(2.2)
172 (5.1)
Other retreatment patients (includes smear-negative cases)
1993
2
2(100.0)
217
1994
95 (43.8)
563
1995
337 (59.9)
1 598
1996
1 023 (64.0)
1997
2511
1 585 (63.1)
1298
765 (58.9)
1998 (1st half)
6189
3 807 (61.5)
Total
0(0)
45 (20.7)
49 (8.7)
68 (4.3)
136(5.4)
167 (12.9)
465 (7.5)
0(0)
18(8.3)
31 (5.5)
105 (6.6)
172 (6.8)
83 (6.4)
409 (6.6)
0(0)
19(8.8)
28(5.0)
114(7.1)
135(5.4)
60 (4.6)
356 (5.8)
0(0)
32(14.7)
76(13.5)
211 (13.2)
377 (15.0)
200 (15.4)
896(14.5)
0(0)
8(3.7)
42 (7.5)
77 (4.8)
106 (4.2)
23(1.8)
256 (4.1)
Year
Patients
New smear-positive patients evaluated
392
1 058
2 138
1993
1994
1995
1996
1997
1998 (1st half)
Total
6 345
7 672
4 660
22 265
of India. Starting in the third and fourth quarters of
1998, large-scale expansion was undertaken, and as
of mid-1999 about 130 million people were covered
by the RNTCP. Case detection rates have been high,
with more than 60 000 patients placed on RNTCP
treatment in the first half of 1999. Treatment out
comes in the expansion areas are not yet available.
Expansion to these areas required the finalisation and
printing of several hundred thousand copies of tech
nical documents, hiring of more than 500 staff, train
ing of more than 4000 doctors, 1000 laboratory
technicians, and 20 000 allied health staff, and pro
curement of nearly 3000 microscopes as well as drugs
for more than 100 000 patients.
The population which will be covered by mid2000, if it were a single country, would be the fifth
largest country in the world, and the DOTS pro
Defaulted
n (%)
Transferred
out
n (%)
gramme is now second only to China’s in size glo
bally. In 1999, more than 140 000 patients were
placed on DOTS in India. Current plans are for con
tinued expansion to a population of approximately
250 million by the end of 2000, and to the entire
country as soon as technically and operationally fea
sible. This may entail the additional coverage of 100150 million population or more, per year, in a phased
manner so that drug supply, training, supervision,
and monitoring can all be ensured. Sustainability of
the programme will require creation of new supervi
sory posts by state governments, which was a condi
tion of the World Bank credit and which has already
taken place in some states and is underway in others,
and continued financial and technical support from
the national level. Costs of key inputs such as drugs
and microscopes have decreased in recent years, and
The status and prospects of tuberculosis control in India
199
I
the total incremental recurrent cost of coverage of the
entire country may be of the order of $50 million or
less, an expense which may not be unrealistic for a
population of one billion people. The key challenge in
the years ahead will be to balance the urgent need for
rapid expansion with the paramount importance of
ensuring quality of implementation.
Acknowledgements
We are grateful to the staff of the Central TB Division, State govern
ments, District Tuberculosis Officers, and medical and paramedical
staff for their hard work and dedication to the programme, to the
Ministry of Health and Family Welfare for support, and especially
to Dr S P Agarwal, Director General of Health Services. Some of the
data presented in this article has been published in the Indian Jour
nal of Tuberculosis (reference 17).
References
1 Indian Council of Medical Research. Tuberculosis in India—A
sample survey 1955-58, Special report series No. 34, ICMR;
New Delhi: 1-21.
2 Tuberculosis Chemotherapy Centre, Madras. A concurrent
comparison of home and sanatorium treatment of pulmonary
tuberculosis in South India. Bull World Health Organ 1959;
21:51-144.
3 Fox W. Self-administration of medicaments. A review of pub
lished work and a study of the problems. Bull Int Union Tuberc
1961; 31: 307-331.
4 Tuberculosis Chemotherapy Centre, Madras. A concurrent
comparison of intermittent (twice-weekly) isoniazid plus strep
tomycin and daily isoniazid plus PAS in the domiciliary treat
ment of pulmon . tuberculosis. Bull World Health Organ
1964;31:247-2’ I.
5 Banerji D, Anderson S. A sociological study of awareness of
symptoms among persons with pulmonary tuberculosis. Bull
World Health Organ 1963; 29: 665-683.
6 Baily G V J, Savic D, Gothi G D, Naidu V B, Nair S S. Potential
yield of pulmonary tuberculosis cases by direct microscopy of
sputum in a district of South India. Bull World Health Organ
1967; 37: 875-892.
7 WHO. Framework for effective tuberculosis control. WHO/
TB/94.179. Geneva: WHO, 1994.
8 WHO. Report on the tuberculosis epidemic, 1998. WHO/TB/
98.237. Geneva: WHO, 1998.
9 WHO. Prevalence and incidence of tuberculosis in India: a
comprehensive review. 1997. WHO/TB/97.231. Geneva:
WHO, 1998.
10 Ministry of Health and Family Welfare. Annual Report 19971998, Government of India, New Delhi, 1998.
11 World Health Organization. Treatment of tuberculosis. Guide
lines for national programmes. 2nd ed, 1997. WHO/TB/
97.220. Geneva: WHO, 1997.
12 Nair S S. Significance of patients with X-ray evidence of active
tuberculosis not bacteriologically confirmed. Indian J Tuberc
1974;21:3-5.
13 Rieder H L, Arnadottir T, Tardencilla Gutierrez A A, et al.
Evaluation of a standardized recording tool for sputum smear
microscopy for acid-fast bacilli under routine conditions in low
income countries. Int J Tuberc Lung Dis 1997; 1: 339-345.
14 Chowdhury AMR, Chowdhury S, Islam M N, Islam A,
Vaughan J P. Control of tuberculosis by community health
workers in Bangladesh. Lancet 1997; 350: 169-172.
15 China Tuberculosis Control Collaboration. Results of directly ob
served short-course chemotherapy by 112842 Chinese patients
with smear-positive tuberculosis. Lancet 1996; 347: 358-362.
16 Datta M, Radhmani M P, Selvaraj R, et al. Critical assessment
of smear-positive pulmonary tuberculosis patients after chemo
therapy under the district tuberculosis programme. Tubercle
Lung Dis 1993; 74: 180-186.
17 Khatri G R. The Revised National Tuberculosis Control Pro
gramme: a status report on the first 100 000 patients. Indian J
Tuberc 1999; 46: 157-166.
r
r
L-
t
I
$
s
I
RESUME
CADRE : LTnde, ou une grande partie de la strategic
mondiale de lutte centre la tuberculose a ete elaboree,
mais on chaque annee, on estime a 2 millions le nombre
de cas de tuberculose.
OBJ ECT IF : Decrire les tactiques, les resultats initiaux et
les lemons tirees de la mise en oeuvre du Programme
National Revise de Lutte centre la Tuberculose appliquant les principes du DOTS (traitement directement
observe et de courte duree).
SCHEMA : Un programme National Revise de Lutte contre la Tuberculose (RNTCP) a ete elabore et mis en
oeuvre depuis 1993. La fourniture reguliere de medica
ments et 1’appui logistique ont ete assures par le financement du Gouverncmcnt de ITndc, des Gouvcrncmcnts
des Etats, de la Banque Mondiale et par des donateurs
bilateraux. Les sujets accusant des symptomes thoraciques et se rendant dans les services de sante sont referes
dans les centres de microscopic pour le diagnostic. Les
cas diagnostiques sont classes selon les directives de
I’OMS et le traitement est donne sous observation
directe. L’on assure un enregistrement systematique et
des rapports par cohorte.
resultats : D’octobre
1993 a la mi-1999, 146012
patients ont ete mis sous traitement dans le cadre du pro
gramme. La qualite du diagnostic etait bonne, le ratio des
patients a bacilloscopie positive sur les patients a bacilloscopie negative etant maintenu a 1:1. Les taux de detec
tion des cas varient largement selon les sites de projet et
sont en correlation avec le rapport du nombre de patients
a bacilloscopie positive sur 1’ensemble des consultants, ce
qui suggere une heterogeneite des taux de maladie. Le succes apres traitement fut obtenu chez 81% des nouveaux
patients a bacilloscopie positive, 82% des patients a bacil
loscopie negative, 89% des patients atteints de tubercu
lose extra-pulmonaire et 70% des cas de retraitement.
conclusion : Le RNTCP a traite avec succes environ
80% des patients dans 20 districts de 15 etats de 1’Inde.
Les taux de succes du traitement ont plus que double et
les taux de deces sont reduits a moins d’un septieme de
ceux du programme anterieur. Le nouveau programme
demarre a la fin 1998 commence a s’etendre et couvre
actuellement plus 130 millions de personnes. Le
prochain defi est de maintenir la qualite de sa mise en
oeuvre au cours de sa phase d’expansion.
I
5;
I’
y.
I
200
The International Journal of Tuberculosis and Lung Disease
RESU M EN
MARCO DE REFERENCIA : India, donde fuera establecido gran parte de la estrategia mundial de control de la
tuberculosis, pero donde cada ano hay 2 millones de
casos de tuberculosis.
objetivo : Describir las politicas, los resultados iniciales y las lecciones aprendidas de la implementacion de
un Programa de Control Nacional de Tuberculosis basado en el DOTS (tratamiento directamente observado
de corta duracion).
METODO : Se designo e implemento un Programa de
Control Nacional Revisado de Tuberculosis (RNTCP)
en 1993. Con fondos del Gobierno de la India, de los
Gobiernos Estatales, del Banco Mundial y de donantes
privados se aseguraron drogas y logistica. Las personas
con sintomas respiratorios que concurren a los ccntros
medicos son reteridos a centres para estudio microscopico
para diagnostico. Los casos diagnosticados son clasificados segun las normas de la OMS y se les suministra trata
miento con observacion directa. Se efectua un registro
sistcmatico y un informe de cohortes.
RESULTADOS: Desde octubre de 1993 hasta mediados
|?vi or
k
1^2-
de 1999, sc incluycron 148.012 pacientcs en el programa de tratamiento. La calidad del diagnostico fue
buena, con una relacion de pacicntes esputo positivo a
esputo negativo mantenida en 1:1. La tasa de deteccion
de casos vario grandemente entre los lugares seleccionados y esta en relacion con el porcentaje de pacicntes que
eran esputo positivos entre los examinados para diag
nostico, sugiriendo tasas de enfermedad heterogeneas.
Se alcanzo un exito terapeutico en el 81% de los pacientes nuevos con esputo positivo, en el 82% de los
pacicntes nuevos con esputo negativo, en el 89% de los pacientes con tuberculosis extrapulmonar y en el 70% de
los pacientes en retratamiento.
CONCLUSION: El RNTCP trato con exito alredcdor del
80% de los pacientes en 20 distritos de 15 estados de la
India. Las tasas de exito terapeutico son mas del doble y
las tasas de mortalidad son menos de una septima parte de
aquellas de los programas previos. El programa se inicio
en 1998, ha crecido y ahora cubre mas de 130 millones de
habitantes. El proximo desafio es mantener la calidad de la
implementacion durante la fase de expansion.
INT J TUBERC LUNG DIS 4(3):201-207
© 2000 IUATLD
Modeling the epidemiology and economics of
directly observed therapy in Baltimore
:
■
i.
C. P. Chaulk,*n M. FriedmaiV R. Dunning^
* Annie E Casey Foundation, tCity Chest Clinic, Baltimore City Health Department, *The Johns Hopkins Schools
of Hygiene & Public Health, and Medicine, § Fiscal Policy Studies Institute, n Bureau of Disease Control, Department
of Epidemiology and Preventive Medicine, Baltimore City Health Department, Baltimore, Maryland, USA
<
_______________________________________________________________________________________________ SUMMARY
SETTING: From 1958 to 1978, Baltimore maintained
one of the highest pulmonary tuberculosis (TB) rates in
the US. But, from 1978 to 1992 its TB rate declined by
64.3% and its ranking for TB fell from second highest
among large US cites to twenty-eighth. This TB trend
coincided with the implementation of an aggressive
directly observed therapy (DOT) program by Balti
more’s Health Department.
OBJECTIVES: We used modeling to estimate the range of
TB cases prevented in Baltimore under DOT. Case esti
mates equal the difference between the observed number
of TB cases in Baltimore versus the expected number if
Baltimore’s TB trend was replaced by the TB trend for
the US (low estimate) or the TB trend for all US cities
with over 250000 residents (high estimate). Economic
savings are estimated.
RESULTS: Without DOT we estimate there would have
been between 1577 (53.6%) and 2233 (75.9%) more
TB cases in Baltimore, costing SI8.8 million to $27.1
million. Cases prevented and expenditures saved in
creased with increased DOT participation.
CONCLUSION: Our model predicts that Baltimore’s TB
decline accompanying DOT resulted in health care sav
ings equal to twice the city’s total TB control budget for
this period. These results arc most plausibly due to DOT,
since it was the only major change in Baltimore’s TB
control program, and rising TB risk factors—AIDS,
injection drug use, poverty—in a city where TB had
been epidemic should have triggered a TB increase as in
comparable US cities, rather than the observed decline.
As national TB rates continue to decline it will be impor
tant to identify ways to capture and reinvest these sav
ings to support effective TB control programs.
KEY WORDS: tuberculosis; directly observed therapy;
economics; prevention; public health; health care expen
ditures; community-based treatment
FROM 1958, when the Centers for Disease Control
and Prevention .(CDC) began reporting pulmonary
tuberculosis (TB) incidence by city as part of its mod
ern TB surveillance program, to 1978, Baltimore
maintained one of the highest TB rates in the US, gen
erally ranking among the highest two or three US cities
for TB (Figure 1).
However, between 1978 and 1981, the Baltimore
City Health Department (BCHD) launched a clinic
based program of directly observed therapy (DOT)
for city-defined ‘high risk’ TB patients, people who
were unemployed, homeless, or alcoholic. These pa
tients were recruited with free anti-TB medications
and free transportation to attend one of the city’s five
chest clinics for DOT. Each year during this period,
roughly 25% of all TB cases in Baltimore were treated
using this case management strategy.
In late 1981, a community-based arm was added
to the DOT program whereby any TB patient in Bal
timore was eligible to receive city-managed DOT.
Community outreach, including home-based TB ther
apy by public health nurses, provided additional in
centives for program participation.
Over time, BCHD expanded its outreach to in
clude workplace and school-based DOT, hospital vis
its by public health nurses, community and provider
education, to recruit a higher proportion of newly di
agnosed cases, and city nursing staff assigned to help
manage high risk patients such as prisoners, sub
stance abuse treatment facility patients, participants
of a clinical study tracking the natural history of
human immunodeficiency virus (HIV) infection among
injection drug users,1 and clients of a community
based AIDS service facility.
By 1992, DOT was being used to treat nearly 80%
of Baltimore’s TB cases. During this same time, TB
declined by 64.3% in Baltimore and its ranking for
TB fell from second highest among large US cities to
twenty-eighth.2 This trend contrasts sharply with the
unexpected rise in TB in the US between 1985 and
iy
I.
r
F
k.
r
I:’
I
I
Correspondence to: C Patrick Chaulk, MD, MPH, Annie E Casey Foundation, 701 St. Paul Street, Baltimore, MD 21202,
USA. Tel: (+410) 547 6600. Fax: (+410) 547 6624. e-mail: patc@aecf.org
Article submitted 1 March 1999. Final version accepted 21 October 1999.
i
202
The International Journal of Tuberculosis and Lung Disease
1st
Communtty-based DOT
Sth
7
Cllnlc-baaeO DOT
10th -
15th
20th
25th
30th
I 1959 I 1961 I 1963 I 1965 I 1967 I 1969 I 1971 I 1973 I 1975 I 1977 I 1979
1958
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
| 1981 I 1983 ' 1985 ! 1987 I 1989
1980
1982
1984
1986
1988
1991 I
1990 1992
Figure 1 Baltimore's ranking for tuberculosis incidence among all US cities with over 250 000
residents between 1958 and 1992.
1992, when some 64 000 more TB cases were
reported than predicted by earlier trends.3
Although many putative factors have been cited
for this resurgence, including AIDS,4 immigration
from countries endemic with TB,5 and clinical mis
management,6 the failure of TB patients to complete
a recommended course of antibiotic therapy has
been cited as the biggest obstacle to effective TB
control.7
The CDC estimates that between 1987 and 1991
roughly 30% of reported TB cases in the largest US
cities did not complete their anti-TB therapy within
the recommended 12-month period.8 Analysis of 1993
TB surveillance data suggests that only 66% of patients
completed therapy by 12 months, while many others
took nearly two years to meet this goal.9 These trends
contrast with the recommendations of the Advisory
Council for the Elimination of Tuberculosis (ACET)
and the CDC that at least 90% of active cases com
plete therapy within 12 months.10’11
DOT has been recommended by virtually all lead
ing US TB control organizations and by the WHO in
order to maximize treatment completion.10’12-14 DOT
is predicated on the theory that watching patients
ingest all their TB medications results in greater treat
ment completion. When recommended antibiotic reg
imens are completed, over 95% of patients infected
with susceptible TB organisms should be cured, and
the spread of infection to others reduced.15
A growing body of evidence supports the relative
effectiveness of a community-based approach to DOT
It has been associated with higher treatment comple
tion rates than any other strategy, particularly self
administered therapy,16 TB reductions in US settings
where the prevalence of TB risk factors such as AIDS,
poverty, substance abuse, and immigration are
rising,2’17 and the prevention of multidrug-resistant
TB (MDR-TB).18 DOT also appears cost-effective
compared to self-administered therapy.19-21
We re-examined Baltimore’s TB trend under DOT
between 1978 and 1992 to estimate the epidemio
logic and economic impact of not using DOT in Bal
timore. In doing so, we provide what we believe is the
first quantitative measure of the preventive effects of
DOT in a high incidence city. These findings, when
coupled with existing cost-effectiveness studies, may be
instructive at a time when national TB rates are declin
ing and the renewed focus on TB control instituted dur
ing the 1990s may seem less compelling today.
Program selection
Baltimore’s TB control program presents a unique
opportunity to assess DOT since BCHD has compre
hensive epidemiologic, economic and programmatic
data available. Moreover, the decline in TB and fall in
city ranking for TB following adoption of DOT
occurred in a city where TB historically had been
problematic, suggesting that this TB decline was other
wise unexpected.
Baltimore also experienced a rise in TB risk factors
such as unemployment, injection drug use, poverty,
and especially AIDS, similar to other large US cities
where TB rates rose.2 Between 1985 and 1992, for
example, Baltimore’s AIDS case rate rose by 171.5%,
making it the leading cause of death in Baltimore for
young men and women, and Baltimore one of the
highest ranking cities for AIDS-related mortality.22
Finally, an expanding, community-based DOT strat
egy was the only major change in the city’s TB control
effort. During this period there were no major alter
ations in hospital control practices, case finding,
screening or prevention activities related to TB which
Modeling DOT in Baltimore
might otherwise provide a programmatic explanation
for the decline of TB in Baltimore.
203
60
S 50
Expected Baltimore trend
Q
£ 40
METHODS
o
o
We assumed that without DOT between 1978 and
1992, Baltimore’s TB trend would have been different.
We hypothesized that this alternative trend would
most likely have fallen within a range defined by two
other actual TB trends during this period: the TB
trend for the US (US TB trend), and the TB trend for
all US cities with over 250 000 residents (large US city
TB trend).
The actual Baltimore, US, and large US city TB
trends were identified based on annual TB incidence
reported in the CDC Annual Tuberculosis Statistics:
States and Cities (Figure 2).23 We terminated these
trends at 1992 because the CDC ceased reporting
large city data in 1993, and instead began reporting TB
incidence for large cities by metropolitan statistical
area (MSA).
Based on these trends we used Lotus 1,2,3 Version
5 to model a range of additional TB cases and health
care costs if there been no DOT in Baltimore.24 This
general approach has been used to model other US TB
trends.25’26
Using the 1978 incidence of TB in Baltimore (48.3/
100 000) as the starting point for our model, we then
compared the observed number of TB cases for the
study period (the actual number of reported TB cases
in Baltimore between 1978 and 1992) with the
expected number of TB cases derived by applying
the US and large US city TB trends to Baltimore’s
1978 starting point.
We subtracted the hypothetical increase in the TB
rate under the US TB trend (low estimate), and the
hypothetical increase in the TB rate under the large
US city TB trend (high estimate) from the actual Bal
timore rate. This rate difference was converted to
cases by back-calculating the annual TB rates reported
by the CDC with Baltimore’s annual population which
30
o
s
I 20
Observed Baltimore trend
o
m
10
0
78 79 80 81 82 83 84 85 86 87 88 89 90 91 92
Year
Figure 3 Estimated excess tuberculosis cases (TB) in Baltimore
if Baltimore's observed TB trend for 1978-1992 were replaced
by the trend for the US.
then served as our estimate of the range of additional
cases.
Finally, health care expenditures for treating these
hypothetical cases were calculated using cost data
from the published literature. We used rounded treat
ment estimates of $13 500 per case for DOT and
$12 700 per case for non-supervised therapy from a
prior study of Baltimore’s DOT program that assumed
full drug susceptibility (i.e., no mono- or multidrug
resistant cases).7
j.?*■
RESULTS
Our low estimate, replacing Baltimore’s TB trend under
DOT with the US TB trend at the Baltimore starting
point, predicts 1577 more cases without DOT than
observed in Baltimore (Figure 3). Instead of the 2940
cases reported in Baltimore between 1978 and 1992,
there would have been 4517, or 53.6%, more cases.
Our high estimate, replacing Baltimore’s TB trend
under DOT with the large US city TB trend at the Bal
timore starting point, predicts 2233 more cases with
out DOT then actually observed in Baltimore (Figure 4).
Thus, instead of the 2940 cases reported between
I- r
i,j.
60
60
I 50
Expected Baltimore trend
5
50
1
1 40
a 40
s
Baltimore
o'
| 30
a
V)
g;
Cities with 250,000
8.
Si 20
o
2.233 excess cases
30
o
20
Observed Baltimore trend
CD
United States
10
p 10
0
0
78 79 80 81 82 83 84 85 86 87 88 89 90 91 92
Year
Figure 2 Observed tuberculosis rates 1978-1992 for Balti
more, all US cities with over 250 000 residents, and the US.
78 79 80 81 82 83 84 85 86 87 88 89 90 91 92
Year
Figure 4 Estimated excess tuberculosis cases (TB) in Baltimore
if Baltimore's observed TB trend for 1978-1992 were replaced
by the trend for all US cities with over 250 000 residents.
t
L
204
The International Journal of Tuberculosis and Lung Disease
Table Annual proportion of all tuberculosis (TB) cases in Baltimore managed through directly observed therapy, estimated cases
prevented and treatment costs saved if the TB trend in Baltimore was replaced by the TB trend for large US cities, 1978-1992
Year
% DOT cases1
Cases prevented per year*
Costs (thousands, US$)
saved per year§
1978* 1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
40
59
198
58
244
55
234
76
268
3089 2961
3391
25
—
26
78
26
109
29
106
32
128
176
54
127
55
117
54
160
60
154
60
134
—
988
1387
1344
1626
2234
1605
1482
2035
1946
1694 2511
* Starting year for directly observed therapy (DOT); assumes no cases prevented, therefore no costs saved.
1 Proportion of all pulmonary tuberculosis cases in Baltimore managed under DOT.
‘ Estimated number of cases prevented each year based on large US city trend replacing observed Baltimore trend for pulmonary tuberculosis.
s Estimated gross annual treatment costs saved based on cases prevented due to directly observed therapy. This totals $28 292 800 less offsetting cost increases
of $ 1 122 000, for net savings of $27 170 000.
1978 and 1992, there would have been 5173, or
75.9% more cases.
The Table displays the annual proportion of all TB
cases in Baltimore managed under DOT with the esti
mated annual number of additional cases and con
comitant costs based on the large US city TB trend
analysis. Total additional treatment costs would
range from $18.8 million (US TB trend) to $27.1 mil
lion (large US city TB trend).
DISCUSSION AND CONCLUSION
Effective TB control is anchored by several critical
public health activities—timely case finding and
reporting, targeted screening and preventive therapy,
and effective management of active cases.27 Among
these activities, treating infectious cases first is the
most important public health goal.
Our analysis focused on Baltimore’s TB case man
agement strategy, where an expanding urban DOT
program emphasizing aggressive community out
reach and patient-centered therapy was accompanied
by a 64% reduction in TB at a time when comparable
US cities were experiencing a rise in TB. The relative
significance of this decline in TB is demonstrated by
Baltimore’s drop in city TB ranking from second
highest to twenty-eighth.
Although modeling can be vulnerable to impreci
sion, it is commonly used to predict expected from
observed TB trends. Here we used TB cases reported
to the CDC to estimate a possible range in TB cases
prevented due to DOT in Baltimore. Our approach
benefited from using actual Baltimore, US, and large
US city rates, rather than hypothetical TB rates.
Our findings suggest that in the absence of DOT
case management there would have been substan
tially more TB cases in Baltimore—as many as 2233
more—than observed, with additional health care
expenditures of roughly $27 million. The number of
cases prevented per year appears to increase pari
passu with DOT participation. We note that the esti
mated annual expenditures to treat these additional
cases would have been twice Baltimore City’s total TB
control budget during the study period.
Since Baltimore often ranked highest among large
US cities prior to implementing DOT, we believe that,
of the two alternative trends, the one based on the
large US city cohort may be most appropriate.
There may be limitations to our analysis. Funda
mentally, we are only simulating the assumed interac
tion between programs (DOT) and outcomes (TB inci
dence). For example, our hypothesis may be incorrect:
the TB trend in Baltimore between 1978 and 1992
might have been essentially the same even without
DOT. However, this is unlikely since the observed TB
decline would be counterfactual to Baltimore’s docu
mented high 20-year TB trend prior to DOT. More
over, implementation of comprehensive DOT else
where has also been followed by declines in TB.18’28
Even if our findings are valid for Baltimore, they
may have limited generalizability to other cities, as
local TB control programs vary in resources, priori
ties, and staffing. Overall program capacity and per
formance, quite apart from DOT, also influence local
TB trends. For example, poor (late) case finding and
reporting would counter even 100% DOT participa
tion since undiagnosed infectious cases would con
tinue to spread TB throughout the community.
On the other hand, our model underestimates the
impact of DOT. For one thing, our large city trend is
based on an average incidence rate for these cities
which should yield lower estimated savings given Bal
timore’s high ranking among these cities prior to
DOT. Optional trends for our analysis might be con
structed based on rates of a ‘comparable’ city or on
the city ranking second highest for TB (since Balti
more ranked second at the start of our trend). How
ever, the length of our analysis (nearly 15 years) pre
cludes the use of a single city for our analysis. Over
such a time frame, city demographics and populations
at risk for TB vary, while TB programs, resources, and
clinical expertise change.
Second, the US and large US city trends include the
observed Baltimore trend which should, if anything,
dampen the effect of DOT. Also, a few of the large US
cities employed DOT at least during the latter portion
of our study period, although in a more limited fash
ion.29-31 Removing any DOT effect in these cities and
the Baltimore trend from this cohort should increase
our savings estimates.
Modeling DOT in Baltimore
Significantly, between 1985 and 1992, AIDS rose
by 172% in Baltimore, making Baltimore one of the
highest cities for AIDS-related mortality. Thus, this
relative growth in AIDS cases should have been
accompanied by a growth in TB similar to many
other large US cities impacted by AIDS during this
period rather than the observed decline in TB.
Also, expansion of DOT from the clinic-based
through the community-based phase was done incre
mentally, taking nearly 14 years. We believe these TB
control results would have been accelerated if a
higher proportion of patients had been recruited into
DOT therapy at the program’s beginning. Applying
DOT to a larger proportion of cases at the start should
prevent even more cases, as failure to treat even a single
case of TB can produce mini-epidemics.32-35
We terminated our analysis at 1992 because of
changes in surveillance practices and did not extend
our analysis to capture savings beyond 1992 in Balti
more. Between 1993 and 1996, for example, annual
DOT participation rates exceeded 90%, while oneyear treatment completion rates were 92%, 94% and
96% for the years 1993, 1994 and 1995. For the
larger period from 1978-1996, Baltimore’s TB rate
declined by 69.1%. An extended trend would there
fore support additional savings.
Our cost assumptions are also conservative since
we excluded indirect costs and did not estimate losses
(economic) from TB deaths prevented.36 Also, we
could not account tor the economic impact of an in
creasing proportion of cases nationally that were
multidrug-resistant and which were more prevalent
during the latter portion of these trends, particularly
in large cities. Estimated cases prevented since the
mid-1980s, when TB was undergoing its greatest up
swing, are more likely to involve MDR-TB and would
be considerably more expensive to treat (averaging as
much as $225 000 per case, or over 10 times the per
case estimate in our model).6 Conversely, in Balti
more, DOT appears to have been helpful in avoiding
significant drug resistance problems, as the rate of
MDR-TB was below 0.06% during the study period.2
Finally, we used our model to examine the implica
tions of modeling Baltimore’s trend nationally, i.e.,
what effect similar (incremental) implementation of
DOT might have had if adopted nationally or by the
largest US cities where reported treatment completion
rates have been the lowest.
We therefore reversed our analyses for this period,
replacing the large city and US TB trends with the
Baltimore TB trend at their respective incidence rates
of 1978. The number of cases prevented nationally
would have been 134 521, and the number prevented
in the largest cities would have been 62 272. Treat
ment costs saved would be $1500M and $728M,
respectively. Assuming, arguendo, that the Baltimore
experience could be reasonably replicated, these pro
jections may be conservative since our model assumes
205
applying DOT to approximately 25% of all cases ini
tially, and only then incrementally expanding it to a
large proportion of cases over the next 14 years.
Nonetheless, these reverse scenarios may be instruc
tive for discussions over the feasibility of TB elimina
tion, while similar trend analysis might guide TB pro
gram evaluation in other urban settings.
Baltimore’s TB trend under DOT suggests that
making DOT available to as many patients as pos
sible had significant public health and economic ben
efits, dramatically reducing both the incidence and
costs of tuberculosis in Baltimore. However, as noted
in a previous study,2 Baltimore’s TB incidence trend in
recent years may be assuming an asymptotic shape,
suggesting that more focused screening and preven
tion activities and case-finding must supplement DOT
if substantial reductions in TB are to continue to
occur in Baltimore. Thus, although DOT appears to be
a key management tool for active cases, it is but one
component of a comprehensive TB control program.
Other essential components include effective associate
and contact investigation, screening and preventive
therapy for high risk populations, and comprehensive
community surveillance, case finding, and disease
tracking.27’37
Further research should focus on developing more
cost-effective delivery mechanisms, particularly in com
munities containing diverse populations of foreignborn and new immigrants. Research should also iden
tify ways to capture treatment savings which can then
be used to sustain community-based interventions.
Acknowledgements
The authors thank Ms Doris Austin for her assistance in preparing
this manuscript.
t
iv.
!
I.
b
References
1 Graham N H M, Nelson K E, Solomon L, et al. Prevalence of
tuberculin positivity and skin test anergy in HIV-1-seropositive
and -seronegative intravenous drug users. JAMA 1992; 267:
369-373.
2 Chaulk C P, Moore-Rice K, Rosetta R, Chaisson R E. Eleven
years of community-based directly observed therapy for tuber
culosis. JAMA 1995; 274: 945-951.
3 Centers for Disease Control and Elimination. Expanded tuber
culosis surveillance and tuberculosis morbidity—United States,
1993. MMWR Morb Mortal Wkly Rep 1994; 43: 361-365.
4 Brudney K, Dobkin J. Resurgent tuberculosis in New York
City: HIV, homelessness and the decline of TB control pro
grams. Am Rev Respir Dis 1991; 144: 745-749
5 McKenna M T, McCray E, Onorato I. The epidemiology of tu
berculosis among foreign-born persons in the United States,
1986-1993. N Engl J Med 1995; 332: 1071-1076.
6 Mahmoudi A, Iseman M D. Pitfalls in the care of patients with
tuberculosis; common errors and their association with the ac
quisition of drug resistance. JAMA 1993; 270: 65-68.
7 Addington W W. Patient compliance: the most serious remain
ing problem in the control of tuberculosis in the United States.
Chest 1979; 76: S741-S743.
8 Centers for Disease Control and Prevention. Tuberculosis pro-
r-
r'.
206
The International Journal of Tuberculosis and Lung Disease
gram management in the United States 1986-1991. Atlanta,
GA: Centers for Disease Control and Prevention, 1993.
9 Bloch A B, Simone P M, Cauthen G M, Kelly G D, Dansbury
K G, Castro K G. Completion of tuberculosis therapy for pa
tients reported in the United States in 1993. Int J Tuberc Lung
Dis 1999; 3: 273-280.
10 Centers for Disease Control and Prevention. Improving patient
adherence to tuberculosis treatment. Revised ed. Atlanta, GA:
Centers for Disease Control and Prevention, US Dept of Health
and Human Services, Public Health Service, 1994.
11 Centers for Disease Control and Prevention. Initial therapy for
tuberculosis in the era of multi-drug resistance: recommenda
tions of the Advisory Council for the Elimination of Tubercu
losis. MMWR Morb Mortal Wkly Rep 1993; 112: 397-406.
12 American Thoracic Society. Treatment of tuberculosis and tu
berculosis infection in adults and children. Am J Respir Crit
Care Med 1994; 149: 1359-1374.
13 Centers for Disease Control and Prevention. Initial therapy for
tuberculosis in the era of multi-drug resistance: recommenda
tions of the Advisory Council for the Elimination of Tubercu
losis. MMWR Morb Mortal Wkly Rep 1993; 42(No. RR-7):
1-8.
14 World Health Organization. WHO report on the tuberculosis
epidemic 1997. Geneva, Switzerland: WHO, 1997.
15 Combs DL, O’Brien R J, Getter L J. USPHS tuberculosis short
course chemotherapy trial 21: effectiveness, toxicity, and ac
ceptability. The report of final results. Ann Intern Med 1990;
112: 397-406.
16 Chaulk C P, Kazandkian V A. Directly observed therapy for
treatment completion of pulmonary tuberculosis: consensus
statement of the public health tuberculosis guidelines panel.
JAMA. 1998; 279: 943-948.
17 Frieden T R. Fujiwara P I, Washko R M, Hamburg M A. Tu
berculosis in \ew York City—turning the tide. N Engl J Med
1995;333:229-233.
18 Weis S E, Slocum P C, Blais F X, et al. The effect of directly ob
served therapy on the rates of drug resistance and relapse in
tuberculosis. N Engl J Med 1994; 330: 1179-1184.
19 Moore R D, Chaulk C P, Griffiths R, Cavalcante S, Chaisson R E.
Cost-effectiveness of directly observed versus self-administered
therapy for tuberculosis. Am J Respir Crit Care Med 1996;
154: 1013-1019.
20 Burman W J, Dalton C B, Cohn D L, Butler J R G, Reves R R.
A cost-effectiveness analysis of directly observed therapy vs
self-administered therapy for treatment of tuberculosis. Chest
1997; 112: 63-70.
21 Palmer C S, Miller B, Halpern M T, Geiter L J. A model of cost
effectiveness of directly observed therapy for treatment of
tuberculosis. J Public Health Management Practice 1998; 4: 113.
22 Selik R M, Chu S Y, Buehler L J W. HIV infection as leading
cause of death among young adults in US cities and states.
JAMA 1993;263:2991-2994.
23 Centers for Disease Control and Prevention. Annual tubercu
losis statistics: States and Cities. Atlanta, GA: Centers for Dis
ease Control and Prevention. Annual volumes 1978-1996.
24 Lotus 1,2,3, Spreadsheet for Windows, release 4. Cambridge,
MA: Lotus Development Corporation, 1993.
25 Centers for Disease Control and Prevention. Tuberculosis mor
bidity—United States, 1992. MMWR Morb Mortal Wkly Rep
1993; 42: 696-704.
26 Centers for Disease Control and Prevention. Expanded tuber
culosis surveillance and tuberculosis morbidity—United States,
1993. MMWR Morb Mortal Wkly Rep 1994; 43: 361-366.
27 Centers for Disease Control and Prevention. Essential compo
nents of a tuberculosis prevention and control program.
MMWR Morb Mortal Wkly Rep 1995; 44(RR-11): 1-16.
28 Pozsik C, Kinney J, Breeden D, Nivin B, Davis T. Approaches
to improving adherence to antituberculosis therapy—South
Carolina and New York, 1986-1992. MMWR Morb Mortal
Wkly Rep 1993; 42: 75-81.
29 Fujiwara P 1, Larkin C, Frieden T R. Directly observed therapy
in New York City. Clin Chest Med 1997; 18: 135-148.
30 Weis S E. Universal directly observed therapy: a treatment
strategy for tuberculosis. Clin Chest Med 1997; 18: 155-163.
31 Schecter G F. Supervised therapy in San Francisco. Clin Chest
Med 1997; 18: 165-168.
32 Braden C R. Infectiousness of a university student with laryn
geal and cavitary tuberculosis. Clin Infect Dis 1995; 21: 565570.
33 Arneil A S, Batterstill J. A mini-epidemic of tuberculosis in the
Upper Fraser Valley Health Unit. Can J Public Health 1973;
64: 497-499.
34 Kenyon T A, Valway S E, Ihle W W, Onorato I M, Castro K G.
Transmission of multidrug-resistant Myocobacterium tubercu
losis during a long airplane flight. N Engl J Med 1996; 334:
933-938.
35 Sacks J J, Brenner E R, Breeden D C, Anders H M, Parker R L.
Epidemiology of a tuberculosis outbreak in a South Carolina
junior high school. Am J Public Health 1985; 75: 361-365.
36 Viscusi W K. The value of risks to life and health. J Economic
Literature 1993; 3: 1912-1946.
37 Bishai W R, Graham N M H, Harrington S, et al. Molecular
and geographic patterns of tuberculosis transmission after 15
years of directly observed therapy. JAMA 1998; 280: 1679—
1684.
RESUME
CADRE : Entre 1958 et 1978, Baltimore a conserve les
taux les plus eleves de tuberculose (TB) pulmonaire aux
Etats Unis. Mais de 1978 a 1992, son taux de TB a
decline de 64,3% et en matiere de TB parmi les grandes
villes des Etats Unis, elle est passee de la deuxieme a la
vingt-huitieme place. Cette tendance de la TB a coincide
avec la mise en oeuvre par le Departement de la Same de
Baltimore d’un programme agressif de traitement directement observe (DOT).
OBJECTIF: Nous avons utilise la modelisation pour
estimer les marges de cas de TB prevenus a Baltimore
pendant le DOT. Les estimations de cas representent la
difference entre le nombre observe de cas de TB a Balti
more et le nombre attendu si la tendance de la TB a Bal
timore avait ete remplacee par la tendance de la TB aux
Etats Unis (estimation faible) ou par la tendance de la
TB pour 1’ensemble des villes de plus de 250000 resi
dents aux Etats Unis (estimation elevee). Les economies
qui en ont resulte ont ete estimees.
RESULTATS : Sans DOT, nous estimons qu’il y aurait eu
entre 1577 (53,6%) et 2223 (75,9%) de cas supplementaires de TB a Baltimore, representant un cout de 18,8 a
27,1 millions de dollars. Le nombre de cas prevenus
ainsi que les depenses evitees ont augmente avec une
augmentation de la participation au DOT.
CONCLUSION : Notre modele predit que le declin de la
TB a Baltimore accompagnant le DOT a entraine des
economies en soins de sante egalant a deux fois le budget
Modeling DOT in Baltimore
total de lutte antituberculeuse a Baltimore pendant cette
periode. Il est le plus plausible que ces resultats soient
dus au DOT, puisque celui-ci a constitue la seule modi
fication majeure du programme de lutte antituber
culeuse a Baltimore et puisque 1’augmentation des facteurs de risque de la TB (SIDA, drogues par voie
intraveineuse et pauvrete) dans une ville ou la TB avait
207
ete epidemique aurait du entrainer une augmentation de
la TB comme dans les villes comparables des Etats Unis
plutot que le declin qui a ete observe. Comme les taux
nationaux de TB continuent a decliner, il sera important
d’identifier les fa^ons de recuperer ces economies et de
reinvestir dans le soutien a des programmes efficaces de
lutte antituberculeuse.
RESUMEN
MARCO DE REFERENCIA : Entre 1958 y 1978, Baltimore
poseia una de las tasas mas altas de tuberculosis pulmonar (TB) en los EEUU. Pero desde 1978 a 1992 su tasa
de TB declino en 64,3% y en la escala jerarquica
descendio del segundo puesto entre las ciudades mas
pobladas de los EEUU al puesto 28. Esta tendencia de la
TB coincidio con la implementacion de un programa
agresivo de tratamiento directamente observado (DOT)
por parte del Departamento de Salud de Baltimore.
OBJETIVO : Utilizamos la modelizacion para estimar el
rango de casos de TB que se previnieron en Baltimore
con el DOT. Los casos estimados resultan de la diferencia entre el numero de casos de TB observados en Balti
more y el numero esperado si la tendencia de TB en Bal
timore fuera reemplazada por la tendencia de la TB en
los EEUU (estimacion baja) o la tendencia de la TB para
todas las ciudades de los EEUU con mas de 250000
habitantes (estimacion alta). Se estimaron los ahorros
economicos.
RESULTADOS : Memos estimado que sin DOT hubiera
habido entre 1.577 (53,6%) y 2.233 (75,9%) casos
suplementarios de TB en Baltimore con un costo de 18,8
a 27,1 millones de dolares. Los casos evitados y los gastos ahorrados ascendieron con el aumento de la participacion en el DOT.
CONCLUSION : Nuestro-modelo predice que la decli
nacion de la TB en Baltimore que acompano al DOT
resulto en ahorros en salud iguales al doble del presupuesto del control total de la TB en este periodo. Es mas
plausible que estos resultados scan debidos al DOT ya
que fue el unico cambio en el programa de control de la
TB en Baltimore y a que los factores de aumento de riesgos de la TB (SIDA, uso de drogas inyectables, probreza)
en una ciudad donde la TB ha sido epidemica hubieran
desencadenado un aumento de la TB como en otras
ciudades de los EEUU en lugar de la declinacion observada. A medida que las tasas nacionales de TB continuen declinando sera importante identificar las formas
de acumular y reinvertir estos ahorros para apoyar un
programa efectivo de control de la TB.
e
I:
i
t.
>• -
!
h
E
fe
I
r ’
I
I
I
- Media
RF_DIS_5_B_SUDHA.pdf
Position: 1751 (5 views)