IntroductionKey populations such as men who have sex with men (MSM), drug users and sex workers are at high risk of HIV infection, but they are marginalised and hidden. Social network strategy (SNS) is purposeful to use social networks to generate social influence, accelerate behaviour change and achieve desirable outcomes among individuals or communities and have been increasingly used for HIV interventions. This study aims to investigate the effects of SNS on HIV prevention among key populations.MethodsWe searched six databases, including PubMed, Web of Science, Embase, Cochrane Library, ScienceDirect and Wiley for randomised controlled trials published between January 1999 and May 2019. Eligibility criteria included SNS conducted among key populations for HIV interventions, with a comparator group. Outcomes included changes in HIV high-risk behaviour, HIV seroconversion and other HIV outcomes. We used the risk ratio (RR) or mean difference with associated 95% confidence interval (CI) to assess the comparative efficacy between SNS and control methods on the selected outcomes. The GRADE system was used to assess the quality of evidence for the studies.ResultsOf 2818 citations identified, 28 trails from 24 papers met the inclusion criteria. The results showed that SNS was associated with less unprotected intercourse (RR 0.79, 95% CI 0.72 to 0.86) and sex with multiple partners (0.46, 95% CI 0.33 to 0.65). Additionally, relative to the control methods, SNS significantly reduced HIV seroconversion (0.65, 95% CI 0.53 to 0.81), improved HIV testing uptake (1.11, 95% CI 1.07 to 1.15) and promoted participant retention (1.03, 95% CI 1.00 to 1.06) among key populations. The Grading of Recommendations Assessment, Development and Evaluation system showed that trails were of moderate quality.ConclusionsThis review provides evidence that SNS can reach key populations who are currently not being reached by existing programmes and deliver HIV interventions through social networks, which decreases HIV sexual risk behaviour and HIV incidence and increases HIV testing uptake and participant retention.Trial registration numberCRD42019140533.