Skip to main content

01.12.2019 | Research article | Ausgabe 1/2019 Open Access

BMC Public Health 1/2019

Revealing a safer sex option to reduce HIV risk: a cluster-randomized trial in Botswana

BMC Public Health > Ausgabe 1/2019
Noam Angrist, Mogomotsi Matshaba, Lesego Gabaitiri, Gabriel Anabwani



1.8 million new HIV infections occur every year, disproportionately affecting adolescent girls and young women. Abstinence-only risk avoidance approaches have had limited impact on reducing new infections. This cluster-randomized trial examines a risk reduction approach to curbing risky sex for school-going girls in Botswana.


The unit of randomization was the school (n = 229). Intervention participants received a 1-h intervention revealing a safer sex option: dating same-age partners which have 5-9x lower HIV prevalence than older partners. Primary outcomes were pregnancy as a proxy for unprotected sex and HIV. Secondary outcomes included self-reported sexual behavior. Generalized linear multilevel models with school-level robust variance for adjusted relative risk ratios were used in an intention-to-treat analysis.


At a 12-month follow up, the intervention reduced pregnancy with an adjusted Relative Risk Ratio (aRRR) of .657 [95% CI .433–.997] significant at the 5% level. Effects were largest at junior school (aRRR = .575 [95% CI .394–.841]) and in rural areas (aRRR = .518 [95% CI .323–.831]), significant at the 1% level. There were no significant effects for primary school students, suggesting age of sexual debut and related mechanisms are critical factors in the intervention’s effectiveness. Moreover, baseline beliefs of which partner is riskiest mediate the magnitude of effects.


Information on safe sex options can change sexual behavior. The success of the intervention working across contexts will depend on various factors, such as age of sexual debut and baseline beliefs.

Trial registration

Pan African Clinical Trials Registry PACTR20190183704​7199. Registered 31 December 2018. Retrospectively registered. This study adheres to CONSORT guidelines.
Über diesen Artikel

Weitere Artikel der Ausgabe 1/2019

BMC Public Health 1/2019 Zur Ausgabe