Sexual health in adolescents

BMJ 2007; 334 doi: https://doi.org/10.1136/bmj.39087.374653.BE (Published 18 January 2007) Cite this as: BMJ 2007;334:103

Sexual health in adolescents: methodology and evidence

Stammers’ editorial[1] on sex education raises a number of
interesting public health issues. However, in the light of our recent
investigation into abstinence-based programmes for HIV prevention in high-
income countries, several of his assertions appear flawed on key
methodological principles.

-Conclusions drawn from Henderson et al.-
Stammers’ conclusions do not follow logically from the trial design and
results reported by Henderson et al.,[2] because his editorial has not
drawn appropriate attention to the actual comparison made. Stammers
concludes, “The trial found no significant difference between the
intervention and control groups ... The results should stimulate urgently
needed change from previous ineffective approaches to school sex
education.” This assertion overlooks that Henderson et al. compared the
SHARE programme to existing sex education programmes, which already
included 7 to 12 lessons of information provision and discussion.[2] In
fact, the evaluation did not show that SHARE (or sex education in general)
is ineffective; it simply showed that SHARE is no more or less effective
than existing sex education programmes. This is an important distinction.
The trial results do not necessarily motivate changes to existing methods;
instead, they suggest only that SHARE does not improve on the standard sex
education already delivered in east Scotland.

-The relevance of behavioural and biological outcome measures-
We agree with Stammers’ assessment of the clinical relevance of outcomes
frequently measured in sexual health programme trials. As we have
previously suggested,[3] the field of HIV prevention could benefit from
consensus on a set of operationalised and clinically relevant outcome
measures, both biological and behavioural, for the evaluation of risk.
This recommendation may apply to the fields of pregnancy prevention and
sexual health as well. The medical evaluation of HIV, other sexually
transmitted infections (STIs), and pregnancy will yield the most
clinically relevant assessments of sexual health promotion programmes,
especially as there are limits to the reliability of self-reported data[4-
7] and to the predictive validity of sexual behaviour outcomes[8 9] as
indicators of actual risk.

However, while medical evaluations of biological outcomes indicate
risk most directly, self-reported sexual behaviour data remain relevant.
Stammers suggests that evaluations showing protective effects for
“intermediate outcomes” such as condom use might lead to “false claims of
success, whereas more robust outcome measures such as rates of
terminations, unplanned conceptions, and sexually transmitted infections
show no benefit.” Here, Stammers does not account for the substantial
problem of floor effects. Often, rates of terminations, conceptions, and
STIs are so low that sample sizes of many hundreds or thousands of
adolescents would be necessary to detect a significant programme effect.
In the US, for example, the prevalence of HIV among young adults is
estimated at approximately 1.0 per 1,000 in the general population.[10]
To detect a difference of even one HIV infection per trial arm, a two-arm
trial among a general young adult population would require data for at
least 2,000 individuals at follow-up. It would not require as many
participants to detect a significant difference in pregnancy rates, but
the problem of floor effects remains relevant. Most trials lack such
expansive sample sizes, but they are often adequately powered to detect
significant behavioural differences.

As a result, while medically evaluated biological outcomes are
crucial, sexual behaviour data such as condom use are a necessary proxy
for most trials. The trials that find significant results for
behavioural, but not biological outcomes are not necessarily making “false
claims” of programme success; few are adequately powered to reliably
detect corresponding differences in biological outcomes. Notably, even
small trials should collect data on biological outcomes, because the
aggregation of many underpowered studies in a meta-analysis can produce
adequate power to detect significant effects.

-Causes of recent declines in adolescent pregnancy rates-
Recent research also contests Stammers’ suggestion that declines in
adolescent pregnancy rates (specifically, US adolescent pregnancy rates)
are primarily attributable to delayed first sex. A study published this
month by Santelli et al.[11] analyses data that correspond to decreases in
the US teen birth rate during the years 1995-2002. Analysing data for
youth aged 15-19, the study found that “the overall pregnancy risk index
declined 38%, with 86% of the decline attributable to improved
contraceptive use.” This contrasts with earlier research cited by

-The evaluation of “Managing the Pressures Before Marriage” by Blake
et al.[13]-
Referring to a randomised controlled trial by Blake et al., Stammers
reports that adding parent-child homework “greatly enhanced” the
abstinence-only programme’s effectiveness as measured by “usual
intermediate outcomes only, such as self efficacy and intention to have
sex.” Although Stammers then calls for more evaluations that assess
sexual debut, he does not mention that Blake et al. did assess sexual
debut at short-term follow-up: the trial found no significant difference
between trial arms in whether students had ever had sexual intercourse.
The evaluation also found no significant effects on whether students
reported having sexual intercourse in the previous 3 months. Attitudes
can be important mediators of sexual activity, but the results of Blake et
al. actually indicate that the addition of parent-child homework did not
enhance short-term programme effects on sexual behaviour. (This finding
was also limited by floor effects, as only 6% of participants in the trial
reported ever having had sex.[13]) It is interesting to note the
contradiction between Stammers’ suggestion that sex education studies
assessing condom use make “false claims of success,” and his acceptance of
attitudinal outcomes of an abstinence-only programme trial as indications
of “greatly enhanced” effectiveness.

-The effectiveness of abstinence-based programmes-
Safe sex and “saving sex” need not be mutually exclusive focuses of sexual
health interventions, as Stammers has also acknowledged in the past.[14]
No one disputes that consistent abstinence from oral, anal, and vaginal
sex can be an effective way to avoid pregnancy, STIs, and the sexual
acquisition of HIV. However, given limited resources, a major issue is
whether it is more effective for programmes to promote only abstinence, or
to promote abstinence along with condom use and other safe-sex strategies.

The most methodologically rigorous systematic reviews to date have
documented no behavioural or biological evidence that abstinence-only
programmes can reduce sexual risk with respect to HIV infection[15] or
pregnancy,[16] as compared to a range of control groups. One review has
even indicated that abstinence-only programmes may cause harm.[16]
However, rigorous reviews[17-20] suggest that hierarchical interventions
that promote both abstinence and condom use (i.e., “abstinence-plus”
programmes) have had long-term protective effects on self-reported sexual
behaviour and/or pregnancy. (To date it appears that only three trials of
abstinence-plus HIV prevention programmes[21-23] have assessed self-
reported STI incidence or treatment, without finding significant effects
among approximately 1,734 participants. The control groups for these
studies were usual care[22 23] and a time-matched HIV prevention program
without skills training.[21] All may have been underpowered to detect
significant differences, and more evaluations are necessary to assess this

Direct comparisons between abstinence-only and abstinence-plus
programmes are lacking. In our research, we have found only one
randomised controlled trial[24] that explicitly compared an abstinence-
only programme against an equal-format programme that prioritised
abstinence but also recommended condom use. This evaluation found no
significant differences in condom use, incidence or frequency of sex, or
number of partners (with analyses representing approximately 194 students
aged 18-21 in the two-arm comparison), but it was limited to six-week
follow-up and faced serious methodological limitations. Many cite Jemmott
et al.[25] as another example, but the abstinence-focused arm in this
trial acknowledged the protective effects of condoms and was therefore not
an abstinence-only intervention. The lack of abstinence-only vs.
abstinence-plus programme trials is problematic, and more evaluations
directly comparing the two programme types will provide key evidence in
this ongoing investigation.


1. Stammers T. Sexual health in adolescents: "saved sex" and parental
involvement are key to improving outcomes. BMJ 2007;334(103-4).

2. Henderson M, Wright D, Raab M, Abraham C, Parkes A, Scott S, et
al. Impact of a theoretically based sex education programme (SHARE)
delivered by teachers on NHS registered conceptions and terminations:
final results of a cluster randomised trial. BMJ 2007;334:132-6.

3. Underhill K, Montgomery P, Operario D. Reporting deficiencies in
trials of abstinence-only programs for HIV prevention. AIDS 2007;21(2):266

4. Rosenbaum J. Reborn a virgin: adolescents' retracting of virginity
pledges and sexual histories. Am J Public Health 2006;96(6):1098-1103.

5. Brener N, Grunbaum J, Kann L, McManus T, Ross J. Assessing health
risk behaviors among adolescents: the effect of question wording and
appeals for honesty. J Adolesc Health 2004;35:91-100.

6. Lauritsen JL, Swicegood CG. The consistency of self-reported
initiation of sexual activity. Fam Plann Perspect 1997;29(5):215-21.

7. Newcomer S, Udry J. Adolescents' honesty in a survey of sexual
behavior. J Adolesc Res 1988;3(3-4):419-23.

8. Peterman TA, Lin LS, Newman DR, Kamb ML, Bolan G, Zenilman J, et
al. Does measured behavior reflect STD risk? An analysis of data from a
randomized controlled behavioral intervention study. Project RESPECT Study
Group. Sex Transm Dis 2000;27(8):446-51.

9. O'Leary A, DiClemente RJ, Aral SO. Reflections on the design and
reporting of STD/HIV behavioral intervention research. AIDS Educ Prev
1997;9 Suppl 1:1-14.

10. Morris M, Handcock M, Miller W, Ford C, Schmitz J, Hobbs M, et
al. Prevalence of HIV infection among young adults in the United States:
results from the Add Health study. Am J Public Health 2006;96(6):1091-7.

11. Santelli J, Lindberg LD, Finer LB, Singh S. Explaining recent
declines in adolescent pregnancy in the United States: the contribution of
abstinence and improved contraceptive use. Am J Public Health

12. Santelli J, Abma J, Ventura S, Lindberg L, Morrow B, Anders J, et
al. Can changes in sexual behaviors among high school students explain the
decline in teen pregnancy rates in the 1990s? Journal of Adolescent Health

13. Blake SM, Simkin L, Ledsky R, Perkins C, Calabrese JM. Effects of
a parent-child communications intervention on young adolescents' risk for
early onset of sexual intercourse. Fam Plann Perspect 2001;33(2):52-61.

14. Stammers T. Abstinence under fire. Postgrad Med J 2003;79:365-6.

15. Kirby D, Laris B, Rolleri L. The impact of sex and HIV education
programs in schools and communities on sexual behaviors among young
adults. Research Triangle Park, NC: Family Health International, YouthNet
Program, 2006.

16. DiCenso A, Guyatt G, Willan A, Griffith L. Interventions to
reduce unintended pregnancies among adolescents: Systematic review of
randomised controlled trials. BMJ 2002;324:1426-1435.

17. Kirby D. Emerging Answers: Research findings on programs to
reduce teen pregnancy. Washington DC: National Campaign to Prevent Teen
Pregnancy, 2001.

18. Jemmott JB, III, Jemmott LS. HIV risk reduction behavioral
interventions with heterosexual adolescents. AIDS 2000;14 Suppl 2:S40-52.

19. Pedlow C, Carey M. HIV sexual risk-reduction interventions for
youth: A review and methodological critique of randomized controlled
trials. Behav Modif 2003;27:135-190.

20. Manlove J, Papillo AR, Ikramullah E. Not yet: programs to delay
first sex among teens. Washington, DC: National Campaign to Prevent Teen
Pregnancy, 2004.

21. St. Lawrence JS, Jefferson KW, Alleyne E, Brasfield TL.
Comparison of education versus behavioral skills training interventions in
lowering sexual HIV-risk behavior of substance-dependent adolescents. J
Consult Clin Psychol 1995;63(1):154-7.

22. Kirby D, Korpi M, Adivi C, Weissman J. An impact evaluation of
Project SNAPP: an AIDS and pregnancy prevention middle school program.
AIDS Educ Prev 1997;9(SA):44-61.

23. Boekeloo BO, Schamus LA, Simmens SJ, Cheng TL, O'Connor K,
D'Angelo LJ. A STD/HIV prevention trial among adolescents in managed care.
Pediatrics 1999;103(1):107-115.

24. Hernandez JT, Smith FJ. Abstinence protection and decision-
making: experimental trials on prototypic AIDS programs. Health Educ Res

25. Jemmott JB, III, Jemmott LS, Fong GT. Abstinence and safer sex
HIV risk-reduction interventions for African American adolescents: a
randomized controlled trial. JAMA 1998;279(19):1529-1536.

Competing interests:
None declared

Competing interests: No competing interests

30 January 2007
Kristen A Underhill
Research Officer
Don Operario, and Paul Montgomery
Centre for Evidence-Based Intervention, University of Oxford, 32 Wellington Square, Oxford, OX1 2ER