Using population attributable risk to choose HIV prevention strategies in men who have sex with men

BACKGROUND In Australia, HIV is concentrated in men who have sex with men (MSM) and rates have increased steadily over the past ten years. Health promotion strategies should ideally be informed by an understanding of both the prevalence of the factors being modified, as well as the size of the risk that they confer. We undertook an analysis of the potential population impact and cost saving that would likely result from modifying key HIV risk factors among men who have sex with men (MSM) in Sydney, Australia. METHODS Proportional hazard analyses were used to examine the association between sexual behaviours in the last six months and sexually transmissible infections on HIV incidence in a cohort of 1426 HIV-negative MSM who were recruited primarily from community-based sources between 2001 and 2004 and followed to mid-2007. We then estimated the proportion of HIV infections that would be prevented if specific factors were no longer present in the population, using a population attributable risk (PAR) method which controls for confounding among factors. We also calculated the average lifetime healthcare costs incurred by the HIV infections associated with specific factors by estimating costs associated with clinical care and treatment following infection and discounting at 3% (1% and 5% sensitivity) to present value. RESULTS Unprotected anal intercourse (UAI) with a known HIV-positive partner was reported by 5% of men, the hazard ratio (HR) was 16.1 (95%CI:6.4-40.5), the PAR was 34% (95%CI:24-44%) and the average lifetime HIV-related healthcare costs attributable to UAI with HIV-positive partners were $AUD102 million (uncertainty range: $93-114 m). UAI with unknown HIV status partners was reported by 25% of men, the HR was 4.4 (95%CI:1.8-11.2), the PAR was 33% (95%CI:26-42%) and the lifetime incurred costs were $AUD99 million. Anal warts prevalence was 4%, the HR was 5.2 (95%CI:2.4-11.2), the PAR was 13% (95%CI:9-19%) and the lifetime incurred costs were $AUD39 million. CONCLUSIONS Our analysis has found that although UAI with an HIV-positive sexual partner is a relatively low-prevalence behaviour (reported by 5% of men), if this behaviour was not present in the population, the number of infections would be reduced by one third. No other single behaviour or sexually transmissible infections contributes to a greater proportion of infections and HIV-related healthcare costs.