7th IAS Conference on HIV Pathogenesis, Treatment and Prevention (IAS 2013)


WEPDB0105 - Poster Discussion Abstract

Effects of HIV antiretroviral therapy on sexual and injecting risk-taking behaviour: a meta-analytical review

Presented by Joseph S. Doyle (Australia).

J.S. Doyle1,2,3, L. Degenhardt4, A. Pedrana1, E. McBryde1, R. Guy5, M. Stoove1, E. Weaver1, A. Grulich5, Y.-R. Lo6, M.E. Hellard1,2,3

1Burnet Institute, Centre for Population Health, Melbourne, Australia, 2Alfred Health, Department of Infectious Diseases, Melbourne, Australia, 3Monash University, Epidemiology and Preventative Medicine, Melbourne, Australia, 4University of New South Wales, National Drug and Alcohol Research Centre, Sydney, Australia, 5University of New South Wales, Kirby Institute, Sydney, Australia, 6World Health Organisation, Division of Combating Communicable Diseases, Manilla, Philippines

Aims: To assess whether antiretroviral agents (ARVs) used for HIV treatment or prevention are associated with changes in reported sexual or injecting risk-taking behaviour, or incidence of sexually transmitted infections (STIs).
Methods: A systematic review was conducted of any studies of antiretroviral agents among HIV-seropositive participants or antiretroviral prophylaxis among HIV-seronegative participants, compared to no antiretroviral use, published up to 1 December 2011. Outcomes of interest included measures of sexual risk-taking behaviour (unprotected intercourse and STIs) and injecting-risk (needle/syringe lending, borrowing or re-use). Journal databases and conference archives were searched for eligible studies. Two reviewers extracted data and appraised the quality of each included study. A random-effects model was used in all statistical analyses. The review protocol followed Cochrane and PRISMA guidelines for conduct and reporting of systematic reviews and was registered prospectively.
Results: Sixty English-language studies met the selection criteria. Fifty-eight studies containing 30,641 participants (median=265) reported sexual risk-takingbehaviour outcomes; eleven studies containing 16,138 participants (median=439) reported STI incidence; and four studies containing 1,600 participants (median=284) reported injecting risk-taking behaviour. All included studies were observational. Sexual risk-taking was lower in those receiving ARVs compared with those not receiving ARVs (odds ratio (OR) 0.71, 95%CI 0.62-0.81, p< 0.001; heterogeneity I2=80%), without any evidence of publication bias demonstrated by Funnel plot. STI incidence was also lower among individuals on ARVs (OR 0.38, 95%CI 0.33-1.01; p=0.05; I2=92%), however there was no difference in injecting risk-taking behaviour with antiretroviral use compared with those not receiving ARVs (OR 0.90, 95%CI 0.60-1.35; p=0.6; I2=0%).
Conclusions: Despite concerns that antiretroviral might increase sexual or injecting risk-taking , research findings suggests sexual risk-taking is reduced in people on ARVs: the reasons for this are not yet clear, though mutually reinforcing effects of HIV treatment and prevention messages among people receiving antiretroviral are possible.

Back to the Programme-at-a-Glance