20th International AIDS Conference - Melbourne, Australia


WEAC0104 - Oral Abstract Session

Is use of antiretroviral treatment (ART) associated with decreased condom use? A meta-analysis of studies from low- and middle-income countries (LMICs)

Presented by Caitlin Kennedy (United States).

C. Kennedy1, K. Armstrong2, V. Fonner1, M. Sweat2, K. O'Reilly2

1Johns Hopkins Bloomberg School of Public Health, International Health, Baltimore, United States, 2Medical University of South Carolina, Psychiatry and Behavioral Sciences, Charleston, United States

Background: With rapidly expanding ART access in LMICs, concerns remain about increases in risky sexual behavior due to treatment optimism, physical health improvements, and assumptions of non-infectiousness with low/undetectable viral load. Conversely, ART may be associated with less risky behavior due to reduced depression, increased hope, and regular medical contact. Research from high-income countries has shown no association between ART and unsafe sex, but despite a rapidly expanding evidence base, no published meta-analyses have examined this issue in LMICs. We conducted a meta-analysis of the association between ART and condom use among PLHIV in LMICs.
Methods: Following PRISMA guidelines, we systematically reviewed and meta-analyzed data from peer-reviewed journal articles from 1/1/1990-5/9/2012. Articles were identified through a comprehensive search of electronic databases, hand searching of key journals, and secondary reference screening. Studies were included if they presented pre-post or multi-arm measures of ART and sexual behavior in LMICs. Two trained coders independently abstracted data using standardized forms. Meta-analysis was conducted using random effects models, stratifying by gender and partner type.
Results: Of 6646 initial citations, 37 studies met the inclusion criteria. Studies came primarily from sub-Saharan Africa (N=30), with additional studies from Thailand (N=4), Brazil (N=2), and India (N=1). Most used cross-sectional or prospective cohort designs, although there were two retrospective cohorts and one non-randomized controlled study. In meta-analysis, PLHIV on ART were more likely to report consistent condom use (Odds ratio [OR]: 1.8, 95% confidence interval [CI]:1.4-2.5, p< 0.000, I2=65.4%) and condom use at last sex (OR:2.3, 95% CI: 1.5-3.6, P< 0.000, I2=88.6%) than those not on ART. This finding was strikingly consistent across sub-analyses by gender and partner type (OR range:1.4-2.6), although not always statistically significant. The strongest effects were seen in consistent condom use with HIV-/unknown status partners (OR:2.6, 95% CI:1.8-3.7, p< 0.000) and spouses (OR:2.6, 95% CI:1.2-5.7, p=0.013). Heterogeneity across analyses was high.
Conclusions: Meta-analytic results show that in LMIC, PLHIV on ART are more likely to report condom use than those not on ART. This is encouraging news for continued expansion of ART programs in LMICs and suggests that “treatment as prevention” may be true in more ways than one.

Back to the Programme-at-a-Glance