HIV, other STIs and Teen Pregnancy: Group-Based Abstinence Education Interventions for Adolescents

Summary of CPSTF Finding

The Community Preventive Services Task Force (CPSTF) finds insufficient evidence to determine the effectiveness of group-based abstinence education interventions delivered to adolescents to prevent pregnancy, HIV and other sexually transmitted infections (STIs). Evidence is considered insufficient because of inconsistent results across studies.


Abstinence education (AE) interventions promote abstinence from sexual activity (either delayed initiation or abstinence until marriage) and mention condoms or other birth control methods only to highlight their failure rates if at all. These interventions usually include messages about the psychological and health benefits of abstinence and could also include other components, such as media campaigns and community service events.

This review evaluated AE interventions delivered in school or community settings to groups of adolescents (10 19 years old), and most adhered to eight federal guidelines that were required to obtain federal funding (the Federal A-H guidelines).

CPSTF Finding and Rationale Statement

Read the full CPSTF Finding and Rationale Statement for details including implementation issues, possible added benefits, potential harms, and evidence gaps.

Promotional Materials

Community Guide News

About The Systematic Review

The CPSTF finding is based on evidence from a systematic review of 21 studies with 23 study arms (search period 1988 – August 2007). The systematic review was conducted on behalf of the CPSTF by a team of specialists in systematic review methods, and in research, practice, and policy related to preventing HIV, other STIs, and teen pregnancy.

Summary of Results

Twenty-one studies with 23 study arms qualified for analysis in this review. The meta-analyses found the following results for primary outcomes:
  • Sexual activity: decrease of approximately 16% (odds ratio [OR]=0.81, 95% confidence interval [CI] 0.70, 0.94; 21 study arms)
    • This decrease is statistically significant, however, effect estimates differed by study design with larger effects for nonrandomized controlled trials compared to randomized controlled trials.
    • Studies with both designs had problems such as differences in follow-up time and multiple studies conducted by the same investigators.
  • Frequency of sex OR=0.77, 95% CI 0.57, 1.04 ( 5 study arms)
    • This result, although in the favorable direction (decrease in frequency), is statistically nonsignificant. Also, effect estimates differed by study design with larger effects for nonrandomized controlled trials compared to randomized controlled trials.
  • STIs: increase of approximately 8% that was statistically nonsignificant (OR=1.08, 95% CI 0.90, 1.29; 9 study arms)
  • Pregnancy: increase of approximately 12% (OR=1.15, 1.00, 1.32; 10 study arms)
    • Although this increase is statistically significant, sensitivity analyses suggest that the effect estimate is unreliable.
  • AE had no meaningful effect on any of the following secondary outcomes:
    • Number of sexual partners
    • Use of protection
    • Unprotected sexual activity
  • In two of the studies, at least one relevant outcome was reported that could not be included in the meta-analysis because of too little information to calculate an odds ratio. The results for these studies were consistent with the results of the meta-analysis.

Summary of Economic Evidence

An economic review of this intervention was not conducted because CPSTF did not have enough information to determine if the intervention works.


Applicability of this intervention across different settings and populations was not assessed because CPSTF did not have enough information to determine if the intervention works.

Evidence Gaps

CPSTF identified several areas that have limited information. Additional research and evaluation could help answer the following questions and fill remaining gaps in the evidence base. (What are evidence gaps?)

The following outlines evidence gaps for these group-based behavioral interventions to prevent or reduce the risk of adolescent pregnancy, HIV, or other STIs: comprehensive risk reduction, and abstinence education interventions.

Across both reviews, there was no consistent evidence of differential effects on outcomes for any of the 12 critical moderator variables (gender, virginity status, age, race/ethnicity, setting, dosage, focus, deliverer, multicomponent, targeting, study design, and comparison group type). Also, the majority of the studies examined interventions delivered to coed groups and results were not reported by gender. This limits the ability to determine differential effectiveness by gender for comprehensive risk-reduction and abstinence education interventions. This limitation extends to the evaluation of the effectiveness of parental participation as well, since it was an uncommon component in these reviews and often had low participation rates.

More consistent reporting of moderator variables by study authors is needed to clarify which of these (or other characteristics) may maximize the effectiveness of adolescent sexual behavior interventions. In addition, common measures of sexual behavior and standard intervals for follow-up assessments of these outcomes would allow for more comparability across studies and lead to a better determination of the overall public health impact of these interventions.

In terms of economic efficiency, future research is needed to examine how cost benefit or cost-effectiveness estimates vary depending on age, gender, and risk status of participants. For programs with objectives beyond pregnancy and STI prevention, future research needs to evaluate the full impact of such programs from a societal perspective, including non-health outcomes such as improved employment potential, and higher future earnings of program participants. Finally, for school-based programs, additional research needs to address the impact on school resources where the facilities, staff, or time from the school systems may be used for these programs.

Study Characteristics

Interventions included in this review were:
  • Targeted to adolescents
    • Girls only
    • Boys only
    • Girls and boys together
  • Delivered in group settings in schools or communities
  • Led by adult or peer educators
  • Implemented as single or multicomponent programs
  • Tailored to groups or individuals

Content of these interventions addressed prevention of:

  • HIV and STIs
  • Pregnancy


Chin HB, Sipe TA, Elder RW, et al. The effectiveness of group-based comprehensive risk-reduction and abstinence education interventions to prevent or reduce the risk of adolescent pregnancy, HIV, and sexually transmitted infections: two systematic reviews for the Guide to Community Preventive Services. American Journal of Preventive Medicine 2012;42(3):272-94.

Community Preventive Services Task Force. Recommendations for group-based behavioral interventions to prevent adolescent pregnancy, Human Immunodeficiency Virus, and other sexually transmitted infections: comprehensive risk reduction and abstinence education. American Journal of Preventive Medicine 2012;42(3):304-7.

Sipe TA, Chin HB, Elder RW, Mercer SL, Chattopadhyay SK, Jacob V, Community Preventive Services Task Force. Methods for conducting Community Guide systematic reviews of evidence on effectiveness and economic efficiency of group-based behavioral interventions to prevent adolescent pregnancy, Human Immunodeficiency Virus, and other sexually transmitted infections. American Journal of Preventive Medicine 2012;42(3):295-303.

Barbot O. Getting our heads out of the sand: using evidence to make systemwide changes. American Journal of Preventive Medicine 2012;42(3):311-12.

Weed SE. Sex education programs for schools still in question: a commentary on meta-analysis. American Journal of Preventive Medicine 2012;42(3):313-15.

Wiley DC. Using science to improve the sexual health of America’s youth. American Journal of Preventive Medicine 2012;42(3):308-10.

Analytic Framework

Effectiveness Review

Analytic Framework see Figure 1 on page 276

When starting an effectiveness review, the systematic review team develops an analytic framework. The analytic framework illustrates how the intervention approach is thought to affect public health. It guides the search for evidence and may be used to summarize the evidence collected. The analytic framework often includes intermediate outcomes, potential effect modifiers, potential harms, and potential additional benefits.

Summary Evidence Table

Effectiveness Review

Summary Evidence Table

Included Studies

The number of studies and publications do not always correspond (e.g., a publication may include several studies or one study may be explained in several publications).

Effectiveness Review

Anderson NLR, Koniak-Griffin D, Keenan CK, Uman G, Duggal BR, Casey C. Evaluating the outcomes of parent-child family life education… including commentary by Hayman LL.Scholarly Inquiry for Nursing Practice 1999 Fall; 13(3): 211-38 (64 ref).

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.

Borawski EA, Trapl ES, Lovegreen LD, Colabianchi N, Block T. Effectiveness of Abstinence-only Intervention in Middle School Teens. American Journal of Health Behavior Vol 29(5) Sep-Oct 2005, 423 434 2005<

Christopher F, Roosa MW. An evaluation of an adolescent pregnancy prevention program: Is “Just Say No” enough? Family Relations: Interdisciplinary Journal of Applied Family Studies 1990;39(1):68-72.

Clark MA, Trenholm C, Devaney B, Wheeler J, Quay L. Impacts of the Heritage Keepers Life Skills Education Component. 2007. Mathematica Policy Research, Inc.

Clark LF, Miller KS, Nagy SS et al. Adult identity mentoring: reducing sexual risk for African-American seventh grade students. J Adolesc Health 2005 October;37(4):337.

Jorgensen SR, Potts V, Camp B. Project Taking Charge: Six Month Follow-Up of a Pregnancy Prevention Program. Family Relations: Interdisciplinary Journal of Applied Family Studies 1993;42:401-6.

Kirby D, Korpi M, Barth RP, Cagampang HH. The impact of the Postponing Sexual Involvement curriculum among youths in California. Fam Plann Perspect 1997;29(3):100-8.

St Pierre TL, Mark MM, Kaltreider DL, Aiken KJ. A 27-Month Evaluation of a Sexual Activity Prevention Program in Boys and Girls Clubs Across the Nation. Family Relations: Interdisciplinary Journal of Applied Family Studies 1995;44:69-77.

Trenholm C, Devaney B, Forston K, Quay L, Wheeler J, Clark MA. Impacts of Four Title V, Section 510 Abstinence Education Programs. Mathematica Policy Research, Inc.; 2008.

Weed S, Ericksen IH, Lewis A, Grant GE, Wibberly KH. An Abstinence Program’s Impact on Cognitive Mediators and Sexual Initiation. Institute for Research and Evaluation. Unpublished Work

Weed S, Olsen JA, DeGaston J, Prigmore J. Predicting and Changing Teen Sexual Activity Rates: A Comparison of Three Title XX Programs. 1992. Unpublished Work

Weed S. Evaluation Report- Heritage Community Services Program Year 2004-2005 (2nd Year of Evaluation). 2005. Unpublished Work

Weed S, Ericksen IH, Birch PJ. An Evaluation of the Heritage Keepers Abstinence Education Program. 2005. Unpublished Work

Weed S, Anderson N. What Kind of Abstinence Education Works? Comparing Outcomes of Two Approaches. 2007. Unpublished Work

Young M, Core-Gebhart P, Marx D. Abstinence Oriented Sexuality Education Initial Field Test Results of the Living Smart Curriculum. FL Educator; 1992.

Search Strategies

The following outlines the search strategy for these group-based behavioral interventions to prevent or reduce the risk of adolescent pregnancy, HIV, or other STIs: comprehensive risk reduction, and abstinence education interventions.

With the assistance of a CDC librarian, the team searched for published studies in the following databases: CINAHL, MEDLINE, PsycINFO, PubMed, Sociological Abstracts, Web of Science, ERIC, POPLINE, NTIS, EPO, CRISP, and the online Cochrane Controlled Trials Register. In addition, we also reviewed references listed in all retrieved articles, published and unpublished reports provided by team members and elsewhere, and references from a search of an electronic database continuously updated and maintained by Prevention Research Synthesis (PRS) in the Division of HIV and AIDS Prevention at CDC.

The teams considered studies for inclusion if they were:

  • Published between 1988 and 2007
  • Published in English
  • Studies conducted in the United States


Portions of the search terms below in parentheses indicate allowances for variation of a keyword, such as the singular and plural versions.

  1. Adolescent(s)
  2. Teenager(s)
  3. Teen(s)
  4. Youth
  5. Young
  6. Student(s)
  7. School
  8. College
  9. Undergraduate
  10. (1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9)
  11. Human Immunodeficiency Virus
  12. HIV
  13. Acquired Immunodeficiency Syndrome or AIDS
  14. AIDS
  15. Sexually Transmitted Diseases
  16. STD
  17. STI
  18. (11 or 12 or 13 or 14 or 15 or 16 or 17 and prevent(ion/ing)
  19. Sex(ual) Behavior(s)
  20. (Sex(ual)) Risk(y) Behavior(s)
  21. Sex(ual) Risk Reduction
  22. Sex(ual) Risk Taking
  23. Sex(ual) Risk Avoidance
  24. Teen/adolescent/teenage pregnancy
  25. Unwanted pregnancy
  26. Unintended pregnancy
  27. Abstinence
  28. Postpon(ing) sex/intercourse
  29. Delay(ing) sex/intercourse
  30. Sexual Activity
  31. Sexual Acts
  32. Protected Sex
  33. Sexual Involvement
  34. Repeated Childbearing
  35. Repeat Pregnancies
  36. Parenting
  37. Cohabitation
  38. Mothers
  39. Fertility Control
  40. (18 through 39OR’d together)
  41. Program evaluation
  42. Outcome stud(ies)
  43. Primary Prevention
  44. Impact stu(dies)
  45. Follow-up stud(ies)
  46. Intervention
  47. Education
  48. Preventive/ion
  49. Evaluation/ing
  50. Program
  51. Promotion
  52. Outcome(s)
  53. Initiative
  54. Design
  55. (41through 54OR’d together)
  56. 10 and 40 and 55

Excluded Terms

Finally, to focus the search results to US-based studies, the following search terms/geographic locations were EXCLUDED:

  • Africa
  • Atlantic Islands
  • Australia
  • Britain/ish
  • Canada
  • Canad(ian)
  • Caribbean Region
  • Central Region
  • Central America
  • England
  • Europe
  • Greenland
  • Indian Ocean Islands
  • Latin America
  • Mexico
  • Oceania
  • Pacific Islands (not Hawaii)
  • South America

Note: This strategy was used rather than the INCLUSION of US-based studies because not all US-based studies are explicitly indexed as such; some allow for the assumption of a US-based study unless otherwise indicated.

Considerations for Implementation

CPSTF did not have enough evidence to determine whether the intervention is or is not effective. This does not mean that the intervention does not work, but rather that additional research is needed to determine whether or not the intervention is effective.