Impact of HIV and sexual health education on the sexual behaviour of young people:
Item
- Title
-
Impact of HIV
and sexual health education
on the sexual behaviour
of young people:
- extracted text
-
I
Impact of HIV
and sexual health education
on the sexual behaviour
of young people:
t
a review update
UNAIDS
j
UNICEF • UNDF • UNFFA
UNESCO « WHO • WOFtlD BANK
I
Impact of HIV
and sexual health education
on the sexual behaviour
of young people:
a review update
UNAIDS 1997
1
Table of Contents
5
Abstract
7
Introduction
11
Methodology
12
Findings
12
13
15
16
18
20
20
22
23
25
Controlled intervention studies
Other intervention studies
Cross-sectional surveys
International or national comparison studies
Exceptions
Discussion
Methodological limitations
Broader education issues
Features of successful programmes
The social context: gender
27
Implications for programme planners
29
Conclusion
30
References
42
47
59
62
Table 1: Controlled intervention studies
Table 2: Other intervention studies
Table 3: Cross-sectional surveys
Table 4: International or national comparison studies
UNAIDS
5
Abstract
assess the effects of HTV/AIDS and sexual health education on
young people’s sexual behaviour, a comprehensive literature review was
commissioned by the Department of Policy, Strategy, and Research of UNAIDS,
the Joint United Nations Programme on HTV/AIDS. Sixty-eight reports were
reviewed. Of 53 studies that evaluated specific interventions, 27 reported that
HTV/AIDS and sexual health education neither increased nor decreased sexual
activity and attendant rates of pregnancy and STDs. Twenty-two reported that HTV
and/or sexual health education either delayed the onset of sexual activity, reduced
the number of sexual partners, or reduced unplanned pregnancy and STD rates.
Only three studies found increases in sexual behaviour associated with sexual
health education. Hence, little evidence was found to support the contention that
sexual health and HTV education promote promiscuity. The interpretative value of
this research was somewhat compromised, however, because of inadequacies in
study design, analytic techniques, outcome indicators, and reporting of statistics.
Future education programmes need to incorporate the features that have been
associated with successful interventions in the past, as well as mechanisms by
which their impact can be evaluated. Programme evaluation should be grounded in
solid study design, and valid and appropriate statistical techniques. Gender and
developmental stage of students are issues for the educator and researcher at both
the design and the evaluation stages of sexual health/HTV education development.
UNAIDS
7
Introduction
iSexual health education for children and young adults is one of the
most hotly debated and emotional issues facing policy makers, national AIDS
programme planners, and educators today. Arguments have raged over how
explicit education material should be, how much there should be, how often it
should be given, and at what age to initiate education. Indeed, the question has been
asked: Why educate adolescents about sex, sexual health and sexually
transmissible diseases (STDs) at all?
Sexual debut for most young people occurs during their teenage years.
Sexual experience among young people has been estimated in a number of
countries: At age 15 years, 53% of young people in Greenland, 38% of young
people in Denmark (Werdelin, Misfeldt, Melbye & Olsen, 1992), and 69% of young
people in Sweden (Klanger, Tyden, & Ruusuvaara, 1993) have experienced
intercourse. By age 18/19 years, the percentage that are sexually active has been
reported as 54.1% in the United States, 31% in the Dominican Republic (Westhoff,
McDermott & Holcomb, 1996), 66.5% in New Zealand (Paul el al., 1995), and
51.6% in Australia (Rodden, Crawford, Kippax & French, 1996). Age of debut has
been estimated at a median of 17 years in England (Wellings et al., 1995) and a mean
of 15.95 years in the United States (Zelnik & Shah, 1983), and 16.8 years in Sweden
(Schwartz, 1993). Therefore, the majority of young people have begun to have
sexual intercourse before they leave their teens, and at least half by the age of 16.
Use of contraception and STD prevention has been reported to vary across
adolescence according to the age at which initiation occurs. Condoms (Kraft, Rise
& Tneen, 1990) and contraception (Faulkenberry, Vincent, James & Johnson, 1987;
Mosher & Bachrach, 1987; Zelnik & Shah, 1983) are more likely to be used the later
sex is initiated. Education on these topics has been found to modify that pattern (see
Tables 1 to 4), and appears to be more effective if given prior to first intercourse (see,
for example, Howard & McCabe, 1990), that is, in adolescence or pre-adolescence.
Partner turnover rate is greater during adolescence and the early
twenties than in later years (Billy, Tanfer, Grady & Klepinger, 1993; Paul et al.,
1995). This is true not only for numbers of casual partners, but also for those
relationships perceived as being regular and monogamous (Rosenthal, Moore &
Brumen, 1990). Although these serially monogamous pairings may be of short
duration, their regular status, in the minds of many of the young people in them,
confers safety with respect to STD transmission (Rosenthal et al., 1990).
Unprotected sex is viewed as not risky because the partner is a regular partner as
opposed to a casual one. Thus unprotected sex occurs with multiple partners, but the
cumulative risk is rendered invisible by the apparent monogamy and commitment
of each discrete relationship.
UNAIDS
9
The advent of the HIV/AIDS pandemic has further inflamed the debate. The
necessarily frank treatment in education programmes of historically taboo sexual
practices (e.g., anal sex, homosexual sexual practice) has rekindled fears as to how
young people will respond to the information presented to them. Whereas HIV
education and sexual health education for young people clearly have fundamental
points at which they diverge (Gillies, 1994), both have been subject to the same
criticism, namely, that the discussion of sexual health for purposes other than the
promotion of abstinence is an incitement and enticement to precocious sexual
involvement (Allgeier & Allgeier, 1988; Nazario, 1992; Scales, 1981; Thomson,
1994; Vincent et al., 1994; Whatley & Trudell, 1993). It is clear that such criticism
has had, and will continue to have, a significant effect on the extent and nature of
HTV and sexual health education (Scales, 1981; Udry, 1993; Vincent et al., 1994).
For this reason a thorough examination of the validity of that contention is essential.
The aim of the current review is not to assess the relative merits of HTV
and sexual health education programmes, nor is it to theorize about why some
approaches appear more successful than others in reducing the unintended
consequences of adolescent sexual activity. Both of those issues have been
addressed comprehensively elsewhere; the findings are summarized in the
discussion section of this review (Kirby, 1992; Kirby et al., 1994; Kirby, 1995;
Mellanby et al., 1992; Oakley, Fullerton & Holland, 1995; Visser & van Bilsen,
1994- Christopher, 1994). This report will complement and extend the previous
work’by bringing together data collected within and outside the United States, and
will eo beyond the scope of previous reviews that are restricted to school-based
interventions (Kirby, 1995; Kirby, 1992; Kirby et al., 1994; Stout & Rivara, 1989),
covering interventions conveyed in tertiary education institutions (e.g., Marcotte
& Logan, 1977), clinical settings (e.g., Mansfield, Conroy, Emans & Woods,
1993) by mail distribution (e.g., Kirby, Harvey, Claussenius & Novar, 1989) and
through public campaigns (e.g., Herlitz, 1993) and community groups (e.g.,
Postrado & Nicholson, 1992). The primary intention is to inform policy makers
programme planners, and educators about the impact of HIV and sexual heal
education on the sexual behaviour of young people as described in the published
literature. The review includes:
• a presentation of studies, summarized in Tables 1 to 4;
• a discussion of key findings under each study type (see Methodology for
the way the studies were classified);
. conclusions, drawn from the data, about the impact of HIV and sexual health
education on the sexual behaviour of young people;
. a general discussion of the methodological problems that have compromised
the assessment of this body of literature,
• a discussion of broader issues in education with respect to gender and social
context;
• a listing of features of successful programmes;
UNAIDS
11
Methodology
j/hirteen literature databases were searched and international experts
in the field were consulted to obtain relevant material1. Where possible, articles
were translated into English. The articles cited in this review are representative
rather than exhaustive.
The focus of the review is on research that studied the behavioural
impact of HIV/AIDS and sexual health education on young people. Research that
dealt solely with knowledge and attitudes about sex has been excluded, because
of the poor association between attitudes and knowledge on the one hand, and
behaviouFontfie other (Kirby,-1985b): This also mean's that only the behavioural
outcomes of multifaceted studies are reported. Similarly, studies describing only
policy and services, with no behavioural impact analysis, have been excluded.
Behavioural outcome is most commonly assessed by comparing people who
did or did not receive HIV/AIDS or sexual health education in terms of adolescent
pregnancy, abortion and birth rates, STD infection rates, and self-reported
sexual activity.
Included in this report are data dating from the mid-1970s, even though
some of the research was conducted before the advent of HIV/AIDS. Although the
content of sexual health education has changed in the last 20 years, the basic
findings from early studies are still relevant with respect to the relationship of
sexual health education to sexual behaviour.
The studies considered in this review were classified into four types:
controlled intervention studies; other intervention studies; cross-sectional surveys;
and national and international comparison studies. The general findings arising
from research under these headings will be described, with key studies discussed
in detail. Three additional studies whose findings conflict with those of the general
corpus of research will be discussed outside that framework.
1 The databases were: PSYCLIT; SOCIOFILE; APAIS; AUSTROM; MEDLINE; FAMILY
RESOURCES; EMBASE; MENTAL HEALTH ABSTRACTS; PASCAL; SOCIAL SCISEARCH; PAIS
INTERNATIONAL; DISSERTATION ABSTRACTS ONLINE; CURRENT CONTENTS.
UNAIDS
13
Vincent et al. (1987), for example, demonstrated the potential for a
dramatic decrease in rates of adolescent pregnancy through the provision of sexual
health education and family planning services. The programme was instituted
within a portion of a county in South Carolina (USA), with the remainder of the
county and three other counties serving as control areas. The intervention involved
education for adult leaders, such as community agency professionals, religious
leaders, and parents. There were also school-based sexual health education for
students from grades K through 12, broadcasting of programme initiatives and
messages through the media, and integration of sexual issues into mainstream
health promotion. After two to three years of programme implementation, the area
in which the intervention was conducted experienced a 35.5/1000 reduction in the
estimated pregnancy rate for females 14 through 17 years of age, as compared with
14 4/1000 in the non-intervention area of the target county, and increases of 5.5
(P<0.002), 16.4 (P <0.001) and 13.9/1000 (P <0.0001) in the control counties. The
study demonstrated that the effects of sexual health education initiatives may be
observed on a scale larger than that of a single school or college class, or institution.
The randomized and controlled design of those intervention studies
permits an accurate assessment of the impact of particular sexual health education
programmes. Tight control over programme content and the study sample allows
valid comparison with the absence of intervention conditions, although one study
that reported comparative reductions in birth rates in the treatment group failed to
state the statistical significance of the findings (Williams et al., 1985). From those
studies it would appear that sexual health education does not lead to greater sexual
activity but may lead to safer and more responsible choices for young people.
[other intervention studies
The bulk of the studies of the relation of HTV and sexual health
education to sexual activity were non-experimental designs. A total of 38 studies
(see Table 2) are grouped in this category. Because of this large number, the text
describing study outcomes has been divided into sections on sexual activity and the
markers of pregnancy, abortion and births; contraceptive use; and condom use,
while the table summarizing the .findings has been divided by programme type.
Sexual activity, pregnancy, abortion, and births
Fourteen studies reported reductions in sexual activity, pregnancies,
births, or abortions (Dycus & Costner, 1990; Daures et al., 1989; Edwards,
Steinman, Arnold & Hakanson, 1980; Howard & McCabe, 1990; Mansfield,
Conroy, Emans & Woods, 1993; Mellanby, Phelps, Crichton & Tripp, 1995;
Nafsted, 1992; National Committee on Health Education, 1978; St. Pierre, Mark,
Kaltreider & Aiken, 1995; Schinke, Blyth, Gilchrist & Burt, 1981; Sellers,
McGraw & McKinlay, 1994; Slap, Plotkin, Khalid, Michelman & Forke, 1991;
UNAIDS
15
Contraceptive use
A number of studies have demonstrated increased use of contraception
among the sexually active following sexual health education (Berger et al., 1987;
Blanchard et al., 1993; Eisen & Zellman, 1987; Berlitz, 1993; Howard & McCabe,
1990; Wielandt & Jeune, 1992; Sakondhavat et al., 1988). Other studies have
indicated that although sexual health education does not generally produce an
increase in coital activity, such education may lead to increases in alternative and
safer practices (in terms of pregnancy or HTV transmission) such as masturbation
or oral sex (Dignan et al., 1985; Yarber & Anno, 1981; Zuckerman et al., 1976).
Condom use
Eight recently published studies evaluated education campaigns that
focused on HIV/AIDS issues and the promotion of condom use (Blanchard et al.,
1993; Goertzel & Bluebond-Langner, 1991; Herlitz, 1993; Kipke et al., 1993;
Rotherum-Borus et al., 1991; Siegal et al., 1995; Turner et al., 1993; Wielandt &
Jeune, 1992). Four of those reported no change in condom use post-test (Goertzel
& Bluebond-Langner, 1991; Kipke et al., 1993; Siegal et al., 1995; Turner et al.,
1993), but six (the remaining four plus Jemmott et al., 1992; and Mansfield et al.,
1993) reported post-intervention increases in condom use with no accompanying
increase in sexual activity or lowering of age of first intercourse. For example, in
Switzerland, Blanchard et al. (1993) serially surveyed first- through fourth-year
apprentices in the Swiss canton of Vaud in 1987, 1990, and 1992 regarding their
sexual behaviour, knowledge, and attitudes. Over that five-year period, the young
people had been exposed to the Swiss Stop-AIDS campaign, which promoted safer,
rather than reduced, sexual activity. From 1987 through 1992 there were dramatic
increases in regular condom use, and no lowering of age of first intercourse. Slap
et al. (1991) reported a decrease from 30% to 24% in condom use among the
sexually active from baseline to post-intervention. This was due to the fact that
seven baseline condom users were not sexually active at follow-up.
These findings from non-experimental studies demonstrated that
education can lead to increases in the extent to which safer sex is practiced but does
not necessarily result in more sexual activity. Interpretation of file findings of six
studies remains speculative, as details of statistical significance were not provided
for observed differences (Dycus & Costner, 1990; Edwards et al., 1980; Marcotte
& Logan, 1977; National Committee on Health Education, 1978; Sakondhavat et
al., 1988; Schinke et al., 1981).
[cross-sectional surveys
In the nine cross-sectional surveys reviewed, study participants were
not assigned randomly to treatment and control conditions, nor were interventions
manipulated by the investigators. Rather, respondents were surveyed as to whether
they had or had not received sexual health and/or contraceptive education and then
UNAIDS
17
All five of the comparison studies indicated that when and where there
was open and liberal policy as well as the provision of sexual health education and
related services (e.g., family planning) there were lower pregnancy, birth, abortion
2d STO mtes For example, Jones et al. (1985) used a 37-countiy comparison of
patterns of adolescent pregnancy to examine the impact of, mter aha, government
education policy, financial support for abortion and single parents, religiosity
openness about sexual health, ethnicity, and marriage laws, on adolescent
pregnancy and sexual activity. Findings from that study indicated that those
countries that rated higher on openness about sex were also those that
the lowest birthrates; teaching of birth control in schools was associated with low
adolescent fertility; and low birth rates were associated with low abortion rates In
a detailed analysis comparing the United States with Canada, England and Wales,
Sweden, the Netherlands, and France, the United States was found to have by far
the highest rates of adolescent pregnancy, birth, and abortion. Differences in
amount of financial support for unmarried mothers, mmonty issues and adolescent
unemployment did not account for the discrepant birth rates. If discouraging the
discussion of sex and access to family planning services in an effort to deter or
shield adolescents from sex were effective policies, the United States would have
been expected to have one of the lowest adolescent pregnancy rates. Instead, for
1980 15- through 19-year-olds in the United States had a pregnancy rate of
96/1000 females, over double that of the countries ranked second (England and
Wales- 45/1000) and nearly seven times that of the sexually liberal Netherlands
(14/1000'1. Countries that address young people’s sexual health in a frank, open
2d supportive manner experienced fewer of the negative consequences of sexual
activity yet did not see greater sexual involvement. Jones et al. conclude that
“incre^ing the legitimacy and availability of contraception and sexual health
education (in its broadest sense) is likely to result in declining adolescent
pregnancy rates” (1985, p. 61).
A complementary review by Singh (1986) examined, on a state-by
state basis within the United States, factors linked to variations in adolescent
pregnancy. With regard to education about sex, state policy and its implementation
varied widely within and between states. Sexual health education was quantified by
documenting the proportions of adolescents receiving sexual heal* education in
junior and senior high schools, the amount of class time devoted to that‘
whether parental consent was required, and the openness of each state s policy
towards permitting sexual health education. The only statisticafiy si^an
finding was an inverse relationship between the proportion of ™r high^schoo
students receiving sexual health education and pregnancy rates. Unfortunately, this
smdy did not gather direct measures of levels of sexual activity, only pregnancy
rates. Therefore, it is not clear whether lower pregnancy rates were due to less
sexual activity or more effective contraceptive use.
There is an interesting contrast between the Singh (1986) analysis and
the aforementioned analysis by Jones et al. (1985). Within the United States a
higher abortion rate was correlated inversely with birth rates. The international
study, however, found a positive correlation between birth and abortion rates: that
UNAIDS
19
those approaches that advocate abstinence will achieve decreases in sexual
involvement or guard against promiscuity. Whatley & Trudell (1993) questioned
the validity of two abstinence programmes as comprehensive sexual health
education. Criticisms included: insufficient and inaccurate information; reliance on
scare tactics; ignoring the realities of adolescents’ lives; reinforcing gender
stereotypes; lack of respect for economic and cultural diversity; presenting only
one side of controversial issues; and inadequate evaluation of programme
outcomes. An abstinence-only approach ignores the developmental diversity in
young people’s sexual health, and marginalizes, and possibly alienates, those who,
for whatever reason, do not adopt the “no sex” option. Further, a programme that
precludes the discussion of prophylactic measures so as to not undermine the
abstinence message misses the opportunity to educate students who will become
sexually active in the future.
The third study that reported an association between sexual health
education^^no^^^exuaFactivity~waS _by~Marsiglio_& Mott (i986).Tn their
sample of 14- through 22-year-olds followed over a five-year period, prior
exposure to a sexual health education course was associated positively and
significantly with the initiation of sexual intercourse at 15 and 16 years of age, but
not at 17 or 18 years of age. As with any other statistical data, correlation does not
imply causality, but this result should not be overlooked. The effect of sexual health
education, however, was less important (according to the statistical model proposed
by the authors) than infrequent church attendance, parental education of less than
12 years, and ethnicity. The authors concluded that “it is unlikely that sexual health
education courses will substantially alter teenage [sexual] behaviour” (Marsiglio &
Mott, 1986, p. 161).
UNAIDS
21
>
Goertzel & Bluebond-Langner, 1991; Kipke et al., 1993; Yarber & Anno, 1981).
One of the five made only between-group comparisons at pretest and post-test
(Bellingham & Gillies, 1993). It is possible for a significant group-by-time
interaction to occur without between-group differences at either pretest or post-test.
The other four studies evaluated within-group pretest to post-test differences
separately for experimental and control groups. Again, important information
regarding the interactive effect of time and intervention is lost with such a design.
The remaining 16 studies used repeated measures analysis including a group-bytime interaction term, between-group differences in change scores from pretest to
post-test, and/or analysis of covariance adjusted for pretest measurements.
Third, drawing conclusions from some studies was difficult because
post-test measurements were made so close to programme completion. This is
particularly so for courses of short duration (Bellingham & Gillies, 1993;
Christopher & Roosa, 1990; Marcotte & Logan, 1977). Post-test measures of
behaviour may overlap the time period in which baseline measures were taken,
particularly if measures are monthly or yearly averages. At best this masks change
and at worst it confounds results. Longer follow-up time allows for sufficient
numbers to accumulate that statistical analysis of change may be validly
undertaken (Stout & Kirby, 1993). Delayed post-test measurements will also yield
information on the durability of change, and assist in the identification of correlates
of sustained change.
Fourth, in evaluations that take an experimental approach, the non
randomization of subjects to control and experimental conditions means that results
will always be subject to self-selection bias, unless the sampling procedure takes
account of this potential confounder (see sampling procedure of Bernard &
Schwartz, 1977). Yet the largest group of studies in this review employed nonexperimental designs. Oakley et al. (1995), in a critical review of the HTV/AIDS
prevention literature, questioned the ability of non-randomized controlled studies
to address adequately biases introduced in the sample selection process4.
Finally, when comparing experimental and control groups, researchers
should be mindful of the heterogeneity in sexual development of the students that
comprise these groups. Evaluation should include some assessment of interactive
effects of sexual developmental stage and the intervention. Differences between
entire groups only reveal aggregate change, which may veil important differential
change in a develop-mentally diverse group. As far as numbers will allow,
comparisons should be made between developmentally comparable subjects from
the control and experimental groups.
4 This review should not be read as endorsing randomized controlled designs or any other
experimental design over other, non-experimental approaches to evaluation of health promotion
interventions It merely notes that in an experimental approach certain hazards are likely to be
encountered depending on the techniques employed. In an area as complex as health promotion^
where behaviour and social practices are informed by a variety of information sources and
understood within particular discourses, it would be unwise to limit research to any one particular,
paradigm or to claim that there is only one way to evaluate results.
pi s ' 3^
IP(
UNAIDS
23
aims of the programme to an appreciable degree (de Gaston, Jensen, Weed &
Tanas, 1994). There is also a diversity of sources of information about sex to which
young people either deliberately or inadvertently are exposed. It cannot be assumed
that what is taught will be translated directly into behaviour, hence the weak
association between sexual knowledge, attitudes, and behaviours.
“The question is not whether children will get sexual health education,
but how and what kind they will receive. It is impossible to hide children
from sexual influences. Adult role models, television, advertisements and
parents bombard young children with them ... silence and evasiveness are
just as powerful teachers as a discussion of the facts” (McNab, 1981, p. 22).
Kirby (1985b) asserted that although the accusation that sexual health
education incites sexuafactivity is unfounded, itJs unrealistic^d^overly optimistic
to construe sexual health education as the panacea for unacceptably high rates of
adolescent STDs and unintended pregnancy. Sexual health education represents a
valuable resource that informs young people’s sexual contact but often it is not the
most influential, thus the potential of education in the development of behavioural
patterns must be assessed in the context of other influences on the sexual health of
young people (Goldman & Goldman, 1981; Spanier, 1976; Stout & Rivara, 1989).
[Features of successful programmes
Although it may be premature to state that education programmes and
provision of clinical services unequivocally reduce STD and unintended pregnancy
rates, the evidence here for reductions in their antecedents suggests that such
interventions have the potential to achieve those outcomes. This raises the question
as to which features of past interventions are associated with reductions in sexual
intercourse and in unprotected sexual intercourse. Several researchers have turned
their attention to this question, most notably Douglas Kirby and his associates
(Kirby, 1992; Kirby et al., 1994; Kirby, 1995). The findings of these three reviews
have given rise to a number of identifiable features associated with successful
outcomes, which will be briefly described here. The nine features identified in his
1995 review of 50 studies of interventions with young people below the age of 19
years are summarized below, as these support and reflect his earlier findings.
Kirby (1995) found that the following features were common
characteristics of programmes that successfully achieved delays in first intercourse,
and/or increased the use of contraception or condoms:
1. Social Influence Theory, Social Learning Theory or Cognitive-Behavioural
theories of behaviour underpinned the interventions;
2. the programmes were focused on the specific aims of delayed intercourse
and protected intercourse;
UNAIDS
25
[The social context: gender
The social context of human sexual health has recently been receiving
greater attention, particularly in the HIV/AIDS literature (Kippax, Crawford,
Waldby & Benton, 1990; Kirby, 1985b; Moore & Rosenthal, 1990; Thomson,
1994) Choosing to have or not to have sex or to use condoms has social meanings,
consequences, and implications for public and private identity (Hollway, 1984).
In the British Women, Risk and AIDS Project (Thomson & Scott, 1991), which
studied 500 young women 16 through 21 years of age, the authors examined the
perceived appropriateness of the sexual health education the women had received.
“By far the most common criticism of sexual health education at school was
that it had little or no relationship to the real choices and pressures around
sexual health that affected the young women in question... the concentration
upon the biology of human reproduction was consistently criticised for taking
no account of the context in which sexual behaviour takes place nor the personal
and social consequences of such behaviour” (Thomson & Scott, 1991, p. 6).
This also highlights the relevance of gender in the delivery of
education regarding HIV and sexual health. Male-to-female transmission of HIV,
for example, was estimated in one study of sero-discordant couples to be 23%
(Padian et al., 1987). In the Masaka district in Uganda, prevalence of HIV in girls
aged between 13 and 19 years old is 20 times that of boys in the same age group
(The status and trends of the global HIV/AIDS pandemic, 1996). Increased risk
arises out of not only a physical vulnerability, but also a social one. Often
responsibility for contraception and STD protection is located with females. This is
so even in the case of condom use, despite their being a male controlled
prophylactic. Messages to that effect make use of the stereotype that women are
responsible for their own sexual conduct and that of their actual or potential male
partners Women are implicitly asked to step out of their other gender stereotype of
passivity and guide the sexual encounter to safety with respect to disease
transmission. There is an inherent contradiction in asking women to ensure the use
of condoms or discouraging penetrative practices, when their culturally legitimized
role in most cultures is one of passivity (Waldby, Kippax & Crawford, 1993). That
is the meanings and assumptions that currently define and inform young women’s
and young men’s sexual lives are often at odds with the strategies proffered by
education campaigns (Kippax el al, 1990; Lever, 1995; Thomson & Scott, 1991).
This is most notable in steady or regular relationships where, in comparison to
casual encounters, condoms are much less likely to be used consistently (Plitcha,
Weisman, Nathanson, Ensminger & Robinson, 1992; Rodden et al., 1996).
The importance of gender considerations in formulating and delivering
HIV and sexual health education can be found in other studies focusing on practice.
The average age difference between females and their first male partners has been
estimated at 1.8 years in the United States and 2.3 years in Sweden (Schwartz,
1993); in Norway 83.7% of girls but only 28.4% of boys reported an older partner
UNAIDS
27
Implications for programme planners
^^esigning high quality programmes is a major challenge for
educationalists and policy makers (DMA Foundation for AIDS, 1997), often
overwhelmed by the array of data and by pressures from public opinion. This
review provides a foundation for policy makers to argue for the continued
development of programmes on life skills, HIV and STD, sexual health, and
reproductive health. The major points raised are these:
• education on sexual health and/or HIV does not encourage increased
sexual activity;
• good quality programmes help delay first intercourse, and protect
sexually-active youth from STD, including HTV, and from pregnancy;
• responsible and safe behaviour can be learned;
• sexual health education is best started before the onset of sexual activity;
• education has to be gender sensitive for both boys and girls;
• young people’s sexual health is informed by a wide range of sources;
• young people are a developmentally heterogeneous group and not all
can be reached by the same techniques.
In addition, studies show that effective education programmes:
• are grounded in Social Learning Theory;
• have focused curricula, giving clear statements about behavioural aims,
and feature clear delineation of the risks of unprotected sex and methods
to avoid it;
• focus on activities that address social influences;
• teach and allow for practice in communication and negotiation skills;
• encourage openness in communicating about sex;
• equip young people with skills for decoding media messages
and their underlying assumptions and ideologies.
The challenge for those who plan for the provision of fflV/sexual
health education is to take the discoveries made by researchers and apply them in
practice. Programme developers need to resist the temptation to design on the basis
of convention and current epidemiological data, and rely rather on evaluated best
practice and trend analysis. Grounding HIV/sexual health education in lessons
UNAIDS
29
*»
Conclusion
influences on young people’s sexual lives are not restricted to explicit
messages about sex. In pursuit of an appropriate and effective way-to promote
healthy, positive sexual behaviour, engagement with those influences is vital. It is
important that policy makers, programme managers, and teachers be aware that the
evidence indicates that safer sexual practice among young people may be achieved
through education. Future education programmes need to incorporate the features that
have been associated with successfill interventions in the past, as well as including
their own evaluation procedures. Programme evaluation should be grounded in solid
study design and _valid. and appropriate statistical techniques. The gender and
developmental stage of the student are issues for the educator and researcher at both
the design and evaluation stages of sexual health/HTV education development. Failing
to provide appropriate and timely information and services to young people for fear
of condoning and encouraging sexual activity is not a viable option.
*
UNAIDS
31
adolescent by giving information in the third year (of secondary school)].
Contraception-fertilite-sexualite 1989; 17:1021 -6.
Davidson JK, Darling CA. The impact of college-level sex education on sexual
knowledge, attitudes and practices: the knowledge/sexual experimentation
myth revisited. Deviant Behavior 1986; 7:13-30.
Dawson DA. The effects of sex education on adolescent behavior. Family Planning
Perspectives 1986; 18:162-70.
de Fine Olivarius F, Worm A-M, Petersen CS, Kroon S, Lynge E. Sexual behaviour of
women attending an inner-city STD clinic before and after a general cam
paign for safer sex in Denmark. Genitourinary Medicine 1992; 68:296-9.
de Gaston JF, Jensen L, Weed SE, Tanas R. Teacher philosophy and program
implementation and the impact of sex education outcomes. Journal of
Research and Development in Education 1994; 27:265-70.
Dignan M, Denson D, Anspaugh D, C’mich D. Effects of sex education on sexual
behaviors of college students. Adolescence 1985; 20:171-8.
Division of Adolescent and School Health. Handbooks for evaluating HIV education.
Atlanta (GA): Centers for Disease Control and Prevention; 1995.
du Guemy J, Sjdberg E. Inter-relationship between gender relations and the HTV/AIDS
epidemic: some possible considerations for policies and programmes. AIDS
1993; 7:1027-34.
Dunne M, Donald M, Lucke J, Milsson R, Ballard R, Raphael B. Age-related increase
in sexual behaviours and decrease in regular condom use among adolescents
in Australia. International Journal ofSTDs and AIDS 1994; 5:41-7.
Dycus S, Costner GM. Healthy early adolescent development (11-13 year olds):
implementing a human sexuality curriculum for seventh graders. Elementary
School Guidance and Counselling 1990; 25:46-53.
Edelman MW, Pittman KJ. Adolescent pregnancy: black and white. Journal of
Community Health 1986; 11:63-9.
Edwards LE, Steinman ME, Arnold KA, Hakanson EY. Adolescent pregnancy
prevention services in High school clinics. Family Planning Perspectives
1980; 12:6-14.
Eisen M, Zellman GL. Changes in incidence of sexual intercourse of unmarried
teenagers following a community-based sex education program. Journal of
Sex Research 1987; 23:527-44.
Faulkenberry R, Vincent M, James A, Johnson W. Coital behaviors, attitudes, and
knowledge of students who experience early coitus. Adolescence 1987;
86:321-32.
Ford N, Fort-D’Auriol A, Ankomah A, Davies E, Mathie E. Review of Literature on
Health and Behavioural Outcomes of Population and Family Planning
Education Programmes in School Settings in Developing Countries.
Geneva: WHO/GPA; 1992.
UNAIDS
33
Kinke M Bover C, Hein K. An evaluation of an AIDS risk reduction education and skills
training (Arrest) program. Journal ofAdolescent Health 1993; 14:533-9.
Kippax S, Crawford J, Waldby C, Benton P. Women negotiating heterosex: implications for
AIDS prevention. Women’s Studies International Forum 1990; 13:533-42.
Kirby D. A review of educational programs designed to reduce sexual risk-taking
behaviors among school-aged youth in the United States. Santa Cruz (CA).
ETR Associates; 1995.
Kirby D School-based programs to reduce sexual risk-taking behaviors. Journal of
School Health 1992; 62:280-7.
Kirbv D The effects of selected sexuality education programs: toward a more realistic
view. Journal ofSex Education and Therapy 1985; 11:28-37. (1985a)
Kirby D Sexuality education: a more realistic view of its effects. Journal of School
\JlealthA9^^2^^i5^__________
Kirby D, Barth RP, Leland N, Fetro JV. Reducing the risk: impact of a new curriculum
on sexual risk-taking. Family Planning Perspectives 1991; 23:253-63.
Kirby D Harvey P, Claussenius D, Novar M. A direct mailing to teenage males about
condom use: its impact on knowledge, attitudes and sexual behavior. Family
Planning Perspectives 1989; 21:12-18.
Kirby D Resnick M, Downes B, Gunderson P, Potthoff S, Zelterman D, et al. The effects
of school-based health clinics in St. Paul on school-wide birthrates. Family
Planning Perspectives 1993; 25:12-16.
Kirbv D Short L, Collins J, Rugg D, Kolbe L, Howard M, et al. School-based programs
to reduce sexual risk behaviors: a review of effectiveness. Public Health
Reports 1994; 109:339-60.
Kirbv D Waszak C, Ziegler J. Six school-based clinics: their reproductive health
services and impact on sexual behavior. Family Planning Perspectives 1991,
23:6-16.
Klanger B Tyden T, Ruusuvaara L. Sexual behaviour among adolescents in Uppsala,
’ Sweden. Journal ofAdolescent Health 1993; 14:468-74.
Kraft P, Rise J, Traeen B. The HTV epidemic and changes in the use of contraception
among Norwegian adolescents. AIDS 1990; 4:673-778.
Kroger F, Wiesner PJ. STD education: challenge for the 80’s. Journal ofSchool Health
1981;51:278-81.
Ku LC, Sonenstein FL, Pieck JH. The association of AIDS education and sex education
with sexual behavior and condom use among teenage men. Family Planning
Perspectives 1992; 24:100-106.
Kvalem I, Sundet J, Rivo K, Eilertsen D, Bakketeig L. The effect of sex education on
adolescents’ use of condoms: applying the Solomon Four-group Design.
Health Education Quarterly 1996; 23:34-47.
UNAIDS
35
Nafsted P. Evaluation of a programme to prevent unwanted pregnancies among young
women in the Stovener district, Oslo, 1988-1990. Tidsskrift for Den norske
leegeforening 1992; 112:3112-14.
National Association of State Boards of Education, United States ot America. Someone
at school has AIDS: a complete guide to educational policies concerning
HIV infection. 1996.
National Committee on Health Education. Living together: a familyplanning project on
Gotland, Sweden. Stockholm: National Swedish Board of Health and
Welfare, Sweden; 1978.
Nazario SL. School teaches the virtues of virginity. Wall Street Journal 1992 Feb 20;
Sect. B:1,3.
N1GZ Netherlands Institute for Health Promotion and Disease Prevention. A school
policy on AIDS/STD education and sexual health: an exemplary brochure.
Amsterdam: European Information Centre “AIDS and youth”; 1995.
Oakley A, Fullerton D, Holland J. Behavioural interventions for HIV/AIDS prevention.
yl/DS 1995; 9:479-86.
Padian N, Marquis L, Francis D, Anderson R, Rutherford G, O’Malley P, et al.
Male-to-female transmission of Human Immunodeficiency Virus. Journal of
the American Medical Association 1987; 258:788-90.
Paul C, Dickson N, Davis P, Lay Yee R, Chetwynd J, McMillan N. Heterosexual
behaviour and HIV risk in New Zealand: data from a national survey.
Australian Journal ofPublic Health 1995; 19:13-18.
Philliber SG, Tatum MC. Sex education and the double standard in school. Adolescence
1982; 17:273-83.
Pick-de-Weiss S, Diaz-Loving R, Andrade-Palos P, David HP. Effect of sex education on
the sexual and contraceptive practices of female teenagers in Mexico City.
Journal ofPsychology and Human Sexuality’ 1990; 3:71 -93.
Plitcha S, Weisman C, Nathanson C, Ensminger M, Robinson J. Partner-specific
condom use among adolescent women clients of a family planning clinic.
Journal ofAdolescent Health 1992; 13:506-11.
Postrado L, Nicholson H. Effectiveness in delaying initiation ot sexual intercourse of
girls aged 12-14: two components of the Girls Incorporated Preventing
Adolescent Pregnancy Program. Youth and Society’ 1992; 23:356-79.
Rees B, Zimmennan S. The effects of formal sex education on the sexual behaviors and
attitudes of college students. Journal of the American College Health
Association 1974; 22:3 70-1.
Rodden P, Crawford J, Kippax S, French J. Sexual practice and understandings of “safe”
sex: assessing change among 18 to 19 year old Australian tertiary students,
1988-1994. Australian Journal ofPublic Health 1996; 20:643-9.
Roosa MW, Christopher FS. Evaluation of an abstinence-only adolescent pregnancy
prevention program: a replication. Family Relations 1990; 39:363-7.
UNAIDS
37
Siegal D, DiClemente R, Dubin M, Krasovsky F, Saliba P. Change in junior high school
students’ AIDS-related knowledge, misconceptions, attitudes, and HIVpreventative behaviors: effects of a school-based intervention. AIDS
Education and Prevention 1995; 7:534-43.
Singh S. Adolescent pregnancy in the United States: an interstate analysis. Family
Planning Perspectives 1986; 18:210-20.
Slap G, Plotkin S, KhalidN,Michelman D, Forke C. A Human Immunodeficiency Virus
peer education program for adolescent females. Journal of Adolescent
Health 1991; 12:434-42.
Smith M. Teen incentives program: evaluation of a health promotion model for
adolescent pregnancy prevention. Journal of Adolescent Health Education
1994; 25:24-9.
Spanier GB. Sex education and premarital sexual behavior among American college
—----- students. Adolescence T978; 13:659-74.
Spanier GB. Formal and informal sex education as determinants of premarital sexual
behavior. Archives ofSexual Behavior 1976; 5:39-67.
Stout JW, Kirby D. The effects of sexuality education on adolescent sexual activity.
Paediatric Annals 1993; 22:120-6.
Stout JW, Rivara FP. Schools and sex education: does it work? Paediatrics 1989; 83:375-9.
The status and trends ofthe global HIV/AIDS pandemic. Satellite symposium organized
by AIDSCAP, Francois-Xavier Bagnoud Centre for Health and Human
Rights, and Joint United Nations Programme on HIV/AIDS (UNAIDS);
1996 July; Vancouver, Canada.
Thomson R Prevention, promotion and adolescent sexuality: the politics of school sex
education in England and Wales. Sexual and Marital Therapy 1994; 9:115-26.
Thomson R, Scott S. Women, Risk and AIDS Project: learning about sex: young women
and the social construction of sexual identity. (WRAP Paper 4). London:
Tufnell Press; 1991.
Traeen B, Lewin B, Sundet J. The real and the ideal; gender differences in heterosexual
behaviour among Norwegian adolescents. Journal of Community and
Applied Social Psychology 1992; 2:227-37.
Turner JC, Korpita E, Mohn LA, Hill WB. Reduction in sexual risk behaviors among
’ college students following a comprehensive health education intervention.
Journal ofAmerican College Health 1993; 41:187-93.
Udry JR. The politics of sex research. Journal ofSex Research 1993; 30:103-10.
UNAIDS Integrating STD/HIV prevention in the school setting: a position paper.
Geneva: UNAIDS 1997.
Vincent ML, Clearie AF, Schluchter MD. Reducing adolescent pregnancy through
school and community based education. Journal of the American Medical
Association 1987; 257:3382-6.
UNAIDS
39
Young M, Core-Gebhart P, Marx D. Abstinence-oriented sexuality education: initial field
test results of the Living Smart curriculum. Family Life Educator 1992; 10:4-8.
Zabin LS, Hirsch MB, Smith EA, Streett R, Hardy JB. Evaluation of a pregnancy
prevention program for urban teenagers. Family Planning Perspectives
1986; 18:119-26.
Zelnik M, Shah F. First intercourse among young Americans. Family Planning
Perspectives 1983; 15:64-72.
Zuckennan M, Tushup R, Finner S. Sexual attitudes and experience: attitude and
personality correlates and changes produced by a course in human sexuality.
Journal of Consulting and Clinical Psychology 1976; 44:7-19.
Tables
Table 1: Controlled intervention studies
Table 2: Other intervention studies
Table 3: Cross-sectional surveys
Table 4: International or national comparison
studies
Study
Intervention
Sample
Key findings for impact on sexual behaviour
Main et al.
1994, USA
15 sessions over 1 semester HIV
prevention using social cognitive
theory on risk behaviour nonns,
factual knowledge & skills
development
N = 419 cont. &
560 exp. M & F in 9th
through 12th grade
•At 6-month follow-up no difference in initiation
of coitus in experimentals (16%) compared
to controls (17%: P = 0.98), or frequency of
intercourse in those active at pretest (P = 0.533)
• Of those sexually active at pretest, experimental
group reported fewer sexual partners in past
2 months compared to controls at post-test
(P = 0.046)
Smith,
1994, USA
8 sessions incl. self-esteem, STDs
communication, decision-making,
sexuality in workshops plus
sessions using role-playing for
skills & negotiation rehearsal
N = 60 cont. &
60 exp. M & F,
mean age = I5J years
•Greater reduction from baseline to immediate
follow-up in frequency of intercourse in the last
2 months for experimentals (3.5/month
to 1.19/month) than controls (3.95/month
to 2.74/month) P <0.05
• Significant increase from baseline
to immediate follow-up in contraceptive use
due to intervention (P <0.005)
N= 577 cont.
& 739 exp. M & F
in 9th & 11 th grade
•At 3-month follow-up, significant reduction
in experimental group compared to controls
in sexual intercourse with partners who used
drugs intravenously (P <0.05)
•Greater monogamy (P <0.05) and consistent
condom use (P <0.05) in experimental group
compared to controls
•No significant difference in changes to rates
of abstinence from pretest to post-test (P = 0.6)
Walter & Vaughan, 6 x I period over 2 days AIDS
1993, USA
prevention curriculum using
health belief, social cognitive
& social influence models incl.
condom use negotiation
& refusal of sex
c
-u >
GO
Study
Intervention
ABSTINENCE PROGRAMS
Christopher
6 sessions incl. sex refusal skills,
& Roosa
self-esteem, consequences of sex,
1990, USA
life goals & family values
Sample
Key Findings for impact on sexual behaviour
•At immediate post-test there was a significant
TV = 129 cont. &
191 exp. M & F
increase in mean lifetime sexual interaction
aged 12 through 13 years for those taking the programme (P <0.02),
but not for controls
•No significant differences in coital behaviour
(no P-value given)
Jorgensen et al,
1993, USA
6 week abstinence-based
pregnancy prevention programme
incl. self development, family
values, pregnancy & STDs
N = 52 cont. &
39 exp. M & F 7th grade
•At 6-month follow-up, for pre-programme
virgins there was a marginal difference in
initiation of sexual activity in recipients of
education (23%) compared to controls (50%)
P = 0.051
Miller et aL,
1993, USA
Home-based 6 x 20 min. sex
education videos discussing
puberty, abstinence, gender
equality, sexual anatomy,
decision-making and refusal-ofsex skills
N = 290 cont. &
258 exp. with 7th &
8th grade M & F
children in culturally
Mormon area
• Follow-up measures taken at 3 and 12 months
• Low rates of sexual intercourse overall (3-5%)
• Significant increase in sexual interaction over
time (P <0.001) for both groups, but no group
by time interaction (P = 0.662)
c
-u f
CJl
Table 2: Other Intervention Studies: Summary of Study Design, Type of Intervention, and Key Findings *
Study
Intervention
HUMAN SEXUALITY COURSES
Baldwin et al.
10 week human sexuality course
1990, USA
on STDs/AI DS, contraception
through lectures & readings
Sample
Key findings for impact on sexual behaviour
TV = 107 cont. &
141 exp. M&F
freshman to senior
sexually active students
•At immediate post-test, no significant
differences in number of vaginal or oral
partners for experimental dr control groups
since the course began (P>0.05)
•No significant changes in dondom use during
vaginal or oral sex for experimental or controls
pre- to post-test (P >0.05) ?
•Significant difference at pretest in levels of
premarital intercourse for exp. (67%) compared
to control subjects (43%: ^<0.05), but no
significant changes by immediate post-test
• Increase in oral sex in experimental group
(F<0.05)
i
Dignan et al.
1985, USA
3 hours x 15 week human
sexuality course (no course
content details given)
TV = 103 cont. &
Dycus & Costner
1990, USA
9-week human sexuality
curriculum involving parents &
school counsellors on HIV, STDs,
& decision-making
• A drop in the pregnancy rate from 30
N= 364 M&F
aged 12 through 15 years (at pre-programme) to 8 tor girls age between
12 & 15 years of age in th£ first year of the pilot
programme (no P-value given)
101 exp. M&F
sophomore year students
c
Rees &
Zimmerman
1974, USA
Human sexuality on homo
sexuality, family planning,
STD, contraception & abortion
♦ See endfor explcuuifion ofuhhivvicitions
N = 230 M&F
college students
• No increases in percentage sexually active
(M at pretest = 79%, at imimediate post-test
73%, no P-value given; F at pretest 61%, at
immediate post-test 62%, no P-value given)
Study
Intervention
Sample
Key findings for impact on sexual behaviour
PUBLIC CAMPAIGNS
Blanchard et al.
Public “Stop AIDS” information
1993, Switzerland campaign carried out in schools,
youth centres at social & sporting
events over 5 years
N= 2911 M&
F 16 through 19 years
of age
•No significant trends in proportion sexually
active over 5 years
•Of those sexually active, no increase in number
of partners over 5 years (no P-value given)
• Large increase in regular condom use from
1987 to 1992 (M approximately 22% to 42%;
F approximately 10% to 32%,: Statistically
significant but no P-value specified)
de Fine Olivarius
et al.
1992, Denmark
Effect of 1985 public campaign
promoting barrier methods
of contraception and limiting
numbers of partners from 1984
to 1988
N = 2365 F aged
attending STD clinic,
mean age = 25.2 years
•No change in total numbers of sexual partners,
frequency of intercourse, births, pregnancies,
abortions, chlamydia, herpes, and cervical
dysplasia (P >0.05)
• Over 4-year period, decrease in gonorrhoea
from 22% to 6% (P <0.01), but increase
in genital warts from 4% to 10% (P <0.05)
Berlitz
1993, Sweden
Effect of 1987 nationwide
campaign on AIDS from 1986
to 1989
N= 11025 M&F
aged between
16&44 years
•No statistically significant changes in percentage
reporting coital activity or number of sexual
partners (no P-value given)
• Increase in use of condoms among singles
with no regular partner (1986: 24% compared
to 1989: 35%, no P-value given)
c
-u
<o
s
Intervention
Study
Wielandt & Jeune
1992, Denmark
Pretest prior to 1985 public
information of HIV/AIDS
from 1984 to 1989
Sample
N= 1381 M&F
aged 16-20 years
Key findings for impact on sexual behaviour
•No difference in age profiles for first coitus
in 1984 & 1989 (median age F: 1984 = 16.7
years, 1989 = 16.8 years; M = 16.9 years
in both 1984 & 1989)
• From 1984 to 1989 more condoms used for
first coitus: M = 36.5% to 61 % (P <0.001);
F = 42.1% to 62% (P <0.001)
ABSTINENCE PROGRAMS
St. Pierre et al,
1995, USA
12 sessions over 3 months,
N= 53 cont. &
booster programme at 1 & 2 years 99 exp. M&F,
mean age =13.6 years
(for only half of experimental
group), on resistance, skills, peer
pressure, abstinence promotion
using role-playing &
advertisement analysis
Young et al,
1992 USA
24 sessions on self-esteem,
puberty, sexual decision-making,
parent-child communication,
abstinence promotion
N = 66 cont. &
60 exp. M&F junior
high school students
(N.A.S)
•No significant differences in exp. & cont. pretest
virgins in initiation of intercourse, or subsequent
frequency & time since last intercourse at
immediate, 1 year & 2 years post-test
• Reduction in pre-programme non-virgins
in frequency & time since last sexual intercourse
at 1 year (P <0.1), & 2 years (P <0.05) post-test
compared to cont. & exp. subjects who also
conipleted booster programme
• At 1 week post-test, 6 exp. & 3 cont. who
had had intercourse in last month at baseline,
had not had intercourse in last month, and
0 exp. & 3 cont. who had not had intercourse
at baseline in the last month, had had
intercourse in the last month (P <0.001.)
c
tn
Study
Intervention
Sample
Marcotte & Logan
1977, USA
3-day medical sex education
course incl. sex role socialization,
physiology, cross-cultural
comparison of sexuality
& sexual health
N = 41 M&F
medical students
mean age = 24.9 years
Mellanby et al.
1995, UK
Postrado &
Nicholson
1992, USA
Key findings for impact on sexual behaviour
O1
• 70.9% at pretest and 75.6% at immediate
post-test had sexual intercourse regularly
(no P-value given)
•Frequency of intercourse: 9.4/month at pretest
and 9.7/month at post-test (no P-value given)
Iq
$
s
£
§
§
§
i
25-30 x 1 hour education
by doctors, teachers & peers
on puberty, reproduction,
contraception, relationships,
assertiveness training through
role play & group work
N = 5398 cont. &
1175 exp. M&F,
15 through 16 years
(I) 6 x 2 hours on pregnancy
prevention, peer pressure,
resistance skills, assertiveness by
discussion & films, & (II) 5x2
hours with parent on parent-child
communication about sex
N = 117 cont. &
295 exp. F,
12 through 14 years
•After approximately one year post-intervention,
controls were 1.45 (95% CI: 1.13 - 1.87) times
more likely than programme students to be
sexually active at 15.5 to 16.5 years of age
5
3c:
2
S
§
s
tn
5
2
•At 1-year follow-up, pretest virgins doing
component I just as likely to initiate sex as
non-participants (OR: 1.0, P = 0.974)
• For component II, at 1-year follow-up pretest
virgin non-participants marginally more likely
to initiate intercourse than participants
(OR: 2.6, P = 0.054)
1§
§
o
3
I
p?
§
Sakondhavat et al.
1988, Thailand
Sex education including abortion,
contraceptive information, STDs
N= 520 M&F
attending vocational
school, mean age =
20.6 years
•After 1 year, no increase in sexual activity
(no P-value given)
• Increase in contraceptive use (no P-value given)
i
Study
Intervention
Sample
Key findings for impact on sexual behaviour
Schinke et al.
1981, USA
Cognitive behavioural
prevention with social worker
on reproduction, contraception,
problem solving, decision
making & interpersonal skills
(no information on course length)
N = 49 cont. &
44 exp. M & F
adolescents
(no age specified)
• Reduction in intercourse without contraception
in treatment compared to controls at 6-month
follow-up (Group 1: 5% versus 23%, Group 2:
7% versus 31%), at 9-month follow-up (Group 1:
8% versus 26%, Group 2: 11% versus 42%) &
at 12 months follow-up (Group 1: 6% versus
30%, Group 2: 11% versus 41%) (no P-values
reported)
Siegal et al.
1995, USA
12 sessions in 3 weeks on HIV,
skills to refuse sex, decision
making, sex education
N= 123 cont. &
434 exp. M & F
7th, 8th & 9th grade
•No significant difference iri changes in sexual
risk-taking (i.e., number of: partners, frequency
of sex, & condom use) froiti pretest to 3-month
follow-up post-test between experimental and
control groups (no P-value given)
Turner et al.
1993, USA
3 to 5 week college seminars on
STDs, safer sex, values, decision
making & assertiveness skills
N = 227 cont. &
341 exp. M & F
mean age = 18.3 years
•At 3-month follow-up, greater abstinence
in experimental males at post-test (42%)
compared to pretest (25%: P <0.05)
and compared to control group at post-test
(29%: P <0.05), but not for females
•No significant differences in number
of partners between pretest & post-test,
or between experimental & control groups
c
O1
cn
>
Study
Intervention
Sample
Key findings for impact on sexual behaviour
Mansfield et aL
1993, USA
Physician intervention: Cont.:
N = 43 standard &
standard HIV counselling on risk 47 extra intervention
assessment, condom use & supply M & F, mean age =
of free condoms; extra interven
17.6 years
tion incl. 20 minutes discussion
on HIV infection susceptibility,
prevention, HIV testing
Rotherum-Borus
etaL
1991, USA
3-30 x 90-120 minutes on HIV
knowledge, coping in risk
situations, skills identification
through videos & discussion
N= 61 cont. &
• Rates of abstinence in past 3 months
78 exp. M & F runaways was the same for experimental & control
aged 11 through 18 years groups at 3-month & 6-month follow-ups
•As number of sessions increased,
so did consistent condom use at 3-month
(p = 0.3, P <0.06) & 6-month follow-up
(p = 0.25, P <0.06)
Slap et aL
1991, USA
1 peer-counselling session
(5-30 minutes) on HTV
transmission, condom use,
risk behaviours, abstinence,
HTV testing & contraception
N = 241 F at adolescent
clinic aged 12 through
19 years
•At 2 month follow-up, no significant
differences between standard & extra
intervention groups in sexual behaviour
or condom use
• From baseline to 2-month follow-up,
0.4 reduction in number of partners
in last month (P <0.0001), and increase
in condom use always from 13% to 23%
(P <0.001) for both intervention groups
combined
• Significant decrease in reporting of sexual
intercourse in last 2 weeks from pretest
(21.3%) to 2-6 week follow-up (13.7%:
P<0.05)
• Decline in condom use always from pretest
(30%) to post-test (24%: P <0.05)
c
>
%
Table 3: Cross-Sectional Surveys: Summary of Study Design, Type of Survey, and Key Findings *
Study
Intervention
Sample
Key findings for impact on sexual behaviour
Anderson et al.
1990, USA
Survey about HIV/AIDS
education in schools
?V=8098 M&F
between 9th and 12th
grade
•Once HIV/AIDS knowledge, gender, race
& age were taken into account, HIV/AIDS
education had no effect on: having > 2 sexual
partners over lifetime and/or in last year,
or on always using condoms (P <0.05)
Dawson
1986, USA
Retrospective survey on receiving
education on menstruation, STD,
birth control, reproduction,
& sexual behaviour
N = 1888 F aged 15
through 19 years
•No effect on probability of initiation of sexual
activity (no P-value given)
•Education recipients more likely to use
contraception: Ever: (P <0.05) & at first
intercourse (if education is given prior
to initiation) (P <0.05)
Furstenberg et al.
1985, USA
Retrospective survey of sex
education & sexual behaviour
7V = 469M &F 15
through 16 years of age
•Prevalence of intercourse significantly higher in
those who did not have sex education (25.5%)
compared to those who did (16.5%: P <0.05)
Ku et al.
1992, USA
Evaluated formal education
in AIDS, birth control, STDs
& resisting sexual activity
N= 1,880 M 15 through
19 years of age
• HIV/AIDS education was associated with
a marginal increase in the number of those who
had no partners in the past year (4%: P <0.1),
& 9% increase in proportion using a condom
100% of the time (P <0.01)
* See enclfor explanation of abbreviations
c
cn
co
cn
Study
Intervention
Sample
Key findings for impact on sexual behaviour
•No relationship between attending a sex
education course in junior or senior high
and subsequent premarital sex behaviour
(X2 = 6.3,df=4:NS)
•No relationship of birth control instruction
to subsequent premarital sexual behaviour
(F: X2 = 2.2, df = 4:NS; M: X2 = 4.4, df = 4:NS)
Spanier
1978, USA
Retrospective survey on
attendance at a sex education
course at either junior or senior
high school, & sexual behaviour
N= 1177M &F
college students
Wellings et al.
1995, UK
Retrospective national survey
of sexual attitudes & lifestyles
N= I8876 M&F
• M whose main source of sex education was
aged 16 through 59 years school-based were significantly more likely
to be virgins at 16 years compared to those
whose main source was friends (P <0.05).
This relationship was non-significant for F
(P>0.05)
•M & F significantly more likely to use
contraception (P <0.05) & F more likely to use
condom at first intercourse (P <0.05; M: NS,
P >0.05) if school was main source of sexuality
information compared to friends as main source
M = males, F = females, cont. = control subjects, exp. = experimental subjects
c
CD
Study
Type of article
Siedlecky
1987, Australia
Commentary on issues
for young people’s sexuality
Singh
1986, USA
Interstate comparison within
the United States of sex
education, abortion, pregnancy
& birth rates
Sample
Key findings for impact on sexual behaviour
•Even though there has been an increase
in the number of school programmes
on sexuality education there has not been
a concomitant increase in adolescent
pregnancies & births ,
I
F 15 through 19 years
of age
I
•A higher proportion of white senior high
school students receiving sex education
was associated highly’with lower pregnancy
rates in white but not black females
(standardised P = -O.$f8: P <0.01)
•No significant relationship between
sex education & abortion rates (P >0.05)
c
o >
Go
- Media
6810.pdf
Position: 1804 (5 views)