Abstract
INTRODUCTION: A decreased frequency of unprotected sex during episodes of concurrent relationships may dramatically reduce the role of concurrency in accelerating the spread of HIV. Such a decrease could be the result of coital dilution - the reduction in per-partner coital frequency from additional partners - and/or increased condom use during concurrency. To study the effect of concurrency on the frequency of unprotected sex, we examined sexual behaviour data from three communities with high HIV prevalence around Cape Town, South Africa. METHODS: We conducted a cross-sectional survey from June 2011 to February 2012 using audio computer-assisted self-interviewing to reconstruct one-year sexual histories, with a focus on coital frequency and condom use. Participants were randomly sampled from a previous TB and HIV prevalence survey. Mixed effects logistic and Poisson regression models were fitted to data from 527 sexually active adults reporting on 1210 relationship episodes to evaluate the effect of concurrency status on consistent condom use and coital frequency. RESULTS: The median of the per-partner weekly average coital frequency was 2 (IQR: 1-3), and consistent condom use was reported for 36% of the relationship episodes. Neither per-partner coital frequency nor consistent condom use changed significantly during episodes of concurrency (aIRR=1.05; 95% confidence interval (CI): 0.99-1.24 and aOR=1.01; 95% CI: 0.38-2.68, respectively). Being male, coloured, having a tertiary education, and having a relationship between 2 weeks and 9 months were associated with higher coital frequencies. Being coloured, and having a relationship lasting for more than 9 months, was associated with inconsistent condom use. CONCLUSIONS: We found no evidence for coital dilution or for increased condom use during concurrent relationship episodes in three communities around Cape Town with high HIV prevalence. Given the low levels of self-reported consistent condom use, our findings suggest that if the frequency of unprotected sex with each of the sexual partners is sustained during concurrent relationships, HIV-positive individuals with concurrent partners may disproportionately contribute to onward HIV transmission.