Can We Prevent HIV and STIs Through School Programs?

Fortune teller, sex risk, STIs, STDs
Fortune teller, sex risk, STIs, STDs
Although a positive impact on behavior change has been reported following the implementation of curriculum-based programs, there is a lack of clear evidence that school-based educational interventions directly prevent HIV/STIs.

Acquisition of HIV infection in adolescence persists with more than 8% of HIV infections in the United States diagnosed in those younger than 19 years of age.1 Unsafe sex remains the primary route of infection for this population. In addition, surveillance data from the US Centers for Disease Control and Prevention (CDC) show that the majority of cases of chlamydia and gonorrhea are reported in young adults aged 15 to 24 years. To address this, the National HIV/AIDS Strategy2 calls for information about HIV risks, transmission, and prevention; to be universally integrated into all educational environments.

School-based interventions aimed at decreasing high-risk sexual behaviors vary greatly. Some are explicitly sex education-oriented and include teaching the biology of HIV transmission, role-play to promote safe sex, and training resistance skills to overcome peer pressure to have sex. Other approaches, derived from social development models, aim to promote prosocial behavior more broadly as a means to prevent HIV and/or sexually transmitted infections (STIs).

In a recent meta-analysis and systematic review, researchers considered studies evaluating the effectiveness of urban school-based programs in the United States running between 1981 and 2008.3 Articles focused on programs targeting youth aged 10 to 19 years and measuring changes in HIV/STI incidence or HIV/STI testing were included. This is the first review assessing such programs by looking primarily at clinical outcomes in the United States.

Nine studies met the inclusion criteria; 3 were randomized controlled trials (RCTs) and 6 were nonrandomized studies. Three of the studies examined the same intervention at different follow-up periods. Comparator groups received standard sex education in 4 interventions, and there was no comparator group for 2 of the interventions (4 studies). One study compared change in STI incidence in a city following implementation of a condom availability program with a similar city that lacked such a program.

All the included studies were evaluated to be of low methodological quality and at high risk for bias, including a concern of positive publication bias for the non-RCTs. No studies were found that recorded HIV incidence as a primary outcome. Follow-up varied widely from 3 months to 18 years.

Seven studies reported STI incidence as an outcome with only 2 finding a significant reduction in STI incidence. Both of these positive studies assessed the Seattle Social Development Project. This comprehensive intervention focused uniquely on preteens in grades 1 through 6. Activities included in-service teacher training, parenting classes, and social competence training for the students. The project aimed to strengthen bonds in both schools and families to protect against socially unacceptable behaviors. Of note, there was no explicit sex education or discussion around sex. For the full 5-year intervention, a significant reduction in STIs was reported (risk ratio [RR] 0.36; 95% CI, 0.23-0.56)4 after 18 years of follow-up compared to a nonrandomized control group. This effect was not found for a “late-intervention” group with involvement only during grades 5 and 6.

Of two studies reporting changes in HIV testing uptake, one showed a significant outcome (RR 3.19; 95% CI, 1.24-8.24) in less than 12 months of follow-up. This intervention combined the Natural Opinion Leader model with the positive influence of popular music. The result was a group of 6 peer “musical opinion leaders” who wrote, recorded, and disseminated HIV prevention-themed music through in-class presentations and distribution of CDs with the composed song “Life is Too Short.”

Safer Choices, the other study with HIV testing as one of its outcomes, did not show an impact. This intervention focused on school-wide change using 5 components: (1) 20 classroom sessions, (2) parental education, (3) school-community linkages, (4) peer resources, and (5) establishment of a “School Health Promotion Council”. Safer Choices demonstrated no change in HIV testing within 12 months (RR 0.78; 95% CI, 0.41-1.47) or with follow-up at 19 and 31 months.

Secondary outcomes of the meta-analysis included frequency of intercourse, number of partners, and proportion initiating sexual intercourse in the 12 months following the intervention. None of these outcomes were reduced in the studies examined. Safer Choices did demonstrate a reduction in the rate of sex without a condom, and this was replicated in one of the Seattle Social Development Project follow-up studies. However, 2 other studies found no change in reported rate of sex without a condom.

Related Articles

The systematic review concluded that there was no persuasive evidence for the effectiveness of school-based interventions in HIV/STI prevention given the low methodological quality and paucity of significant positive findings. The authors note that the low prevalence of HIV and STIs in the populations studied leads to difficulty finding statistically meaningful results without impractically large sample sizes.

Aside from this meta-analysis, numerous other studies have evaluated the effect of curriculum-based programs on behavior change and have reported significant positive impacts. Kirby reviewed 56 studies, including 9 assessing the effect of abstinence promotion.5 Two-thirds of the studies looking at comprehensive sex and HIV/STI education demonstrated good evidence of a delay in initiation of sex and increased condom and contraceptive use. In contrast, only 2 of the 9 abstinence programs showed any positive effect on sexual behavior.

A recent Cochrane systematic review assessed the international picture.6 Focusing on clinical outcomes (HIV/STI prevalence, pregnancy), 8 RCTs (5 in Africa, 1 in South America, and 2 in Europe) with a total of 55,157 participants were evaluated. These trials found no evidence that educational programs delivered in a school setting reduced the prevalence of HIV, herpes simplex virus type 2, or syphilis. However, 2 of the studies assessed incentive-based programs to promote school attendance through cash transfers or free school uniforms. The results demonstrated a reduction in adolescent pregnancy and, in the case of cash transfers, a reduction in the prevalence of herpes simplex virus type 2.

In summary, there is a lack of clear evidence that school-based educational interventions directly prevent HIV/STIs. However, the demonstrated impact of programs to moderate high-risk sexual behavior may be a more appropriate measure of success.


  1. Centers for Disease Control and Prevention.  STDs and HIV— CDC Fact Sheet. Updated July 10, 2017. Accessed December 14, 2017.
  2. National HIV/AIDS strategy: updated to 2020. Updated January 31, 2017. Accessed November 13, 2017.
  3. Mirzazadeh A, Biggs MA, Viitanen A, et al. Do school-based programs prevent HIV and other sexually transmitted infections in adolescents? A systematic review and meta-analysis [published online August 8, 2017]. Prev Sci. doi: 10.1007/s11121-017-0830-0
  4. Hill K G, Bailey JA, Hawkins J D, et al. The onset of STI diagnosis through age 30: results from the Seattle Social Development Project intervention. Prev Sci. 2014;15:S19-S32.
  5. Kirby D. The impact of abstinence and comprehensive sex and STD/HIV education programs on adolescent sexual behavior. Sex Res Social Policy. 2008;5(3):18-27.
  6. Mason-Jones AJ, Sinclair D, Mathews C, et al. School-based interventions for preventing HIV, sexually transmitted infections, and pregnancy in adolescents. Cochrane Database Syst Rev. 2016;11:CD006417.