Article Text
Abstract
Introduction Condom less commercial sex work is a common mode of HIV transmission in China. This study systematically reviews the impacts of behavioural interventions on condom use and HIV testing uptake among female sex workers (FSW) in China.
Methods Chinese and English language peer-reviewed articles published between January 2000 and December 2013 were searched in five electronic databases. Odds ratios (OR) were calculated by comparing the levels of improvements in condom use and HIV testing uptake by various intervention strategies. Study quality was assessed for included studies. This review followed the PRISMA guidelines and was registered in PROSPERO (CRD42014013466).
Results One hundred and twenty-eight studies met inclusion criteria. Meta-analyses indicated that FSW in the post-intervention period were 2.3–5.0 times more likely to use condoms with male clients in their last sexual act and 2.3–3.4 times more likely to consistently use condoms in the last month than in the pre-intervention period. In particular, multiple intervention sessions were more effective in improving condom use among FSW with male clients (OR = 5.6, [4.0–7.8]) than a single intervention session (OR = 3.3, [2.8–3.8]). Behavioural interventions also improved past-12-month HIV testing uptake 4.6 fold (95% CI, 2.9–7.4). Comprehensive intervention programs, which include health education plus additional sexual health care services, testing for HIV infection and counselling services, were more effective (OR = 8.1, [4.0–16.7]) in improving HIV testing uptake compared with health education only programs (OR = 2.7, [1.6–4.5]). Longer intervention duration (>12 months) did not increase effectiveness in improving condom use or HIV testing rate among Chinese FSWs.
Conclusion Behavioural interventions are effective in improving condom use and HIV testing uptake among Chinese FSW. This review highlights both the potentials and limitations of condom promotion interventions targeting female sex workers.
Disclosure of interest statement None.