HIV This Week

Bryant J, Baxter L, Hird S. Non-occupational postexposure prophylaxis for HIV: a systematic review. Health Technol Assess. 2009;13(14):1-82.

Bryant and colleagues review the evidence on the clinical effectiveness and cost-effectiveness of non-occupational postexposure prophylaxis (PEP) for HIV. Eleven electronic databases were searched from inception to December 2007. Selected studies were assessed, subjected to data extraction using a standard template and quality assessment using published criteria. Studies were synthesised using a narrative approach with full tabulation of results from all included studies. One clinical effectiveness study meeting the inclusion criteria was identified, a cohort study of PEP in a high-risk HIV-negative homosexual male cohort in Brazil. The quality of the study was generally weak. Seroincidence in the cohort as a whole (2.9 per 100 person-years) was very similar to that expected in this population (3.1 per 100 person-years, p > 0.97), despite the seroconversion to HIV being 1/68 in the PEP group and 10/132 in the group not receiving PEP. High-risk sexual activities declined over time for both PEP and non-PEP users. Four economic evaluations met the inclusion criteria of the review. The methodological quality of the studies was mixed. The studies are constrained by a lack of published data on the clinical effectiveness of PEP after non-occupational exposure, with effectiveness data derived from one study of occupational PEP. Their generalisability to the United Kingdom is not clear. Results suggest that PEP following non-occupational exposure to HIV was cost saving for men who have unprotected receptive anal intercourse with men, whether the source partner is known to be HIV positive or not; heterosexuals after unprotected receptive anal intercourse; and people who inject drugs with contaminated equipment previously used by a known HIV-positive person. PEP following non-occupational exposure to HIV was cost-effective for all male-male intercourse (unprotected receptive and insertive anal intercourse, unprotected receptive oral sex, and ‘other’) and was possibly cost-effective for people who inject drugs and high-risk women. Four additional studies were identified giving further information about adverse events associated with PEP after non-occupational exposure to HIV. The majority of participants experienced adverse events with the most common being nausea and fatigue. Rates were generally higher in participants receiving triple therapy than in participants receiving dual therapy. Completion of PEP therapy was variable, ranging from 24% to 78% of participants depending on type of therapy. Toxicity was the main reason for discontinuation of treatment. It is not possible to draw conclusions on the clinical effectiveness of non-occupational PEP for HIV because of the limited evidence available. The review of cost-effectiveness suggests that non-occupational PEP may be cost-effective, especially in certain population subgroups; however, the assumptions made and data sources used in the cost-effectiveness studies mean that their results should be used with caution.

Editors' note: Clinical trials of post-exposure prophylaxis (PEP) for occupational exposure are neither ethical nor practical and so, based on animal studies, occupational PEP is standard practice everywhere. For non-occupational exposure, many countries have 5-day PEP starter kits in emergency rooms and other settings for people who may have been exposed to HIV during rape. This review underscores the need for better data collection on the use and outcomes of non-occupational PEP and for counselling about adverse events to improve adherence during the 4-week course of antiretroviral medication. It suggests that PEP for non-occupational exposure could be cost-effective particularly following exposure to blood/body fluids known to contain HIV.

The aim of the study was to determine the cost-effectiveness of HIV non-occupational post-exposure prophylaxis (PEP) in Australia. A retrospective cost analysis of a population-based observational cohort of 1601 participants eligible for non-occupational PEP in Australia between 1998 and 2004 was carried out. Guinot and colleagues modelled non-occupational PEP treatment costs and combined them with effectiveness outcomes to calculate the cost per seroconversion avoided. They estimated the cost-utility of the programme, and sensitivity and threshold analysis was performed on key variables. The average of non-occupational PEP cost per patient was A$1616, of which A$848 ( 52%) was for drugs, A$331 ( 21%) for consultations, A$225 ( 14%) for pathology and A$212 ( 13%) for other costs. The cost per seroconversion avoided in the cohort was A$1 647 476 in the base case analysis, and A$512 410 when transmission rates were set at their maximal values. The cost per quality-adjusted life-year (QALY) was between A$40 673 and A$176 772, depending on the risks of HIV transmission assumed. The authors conclude that in their base case, non-occupational PEP was not a cost-effective intervention compared with the widely accepted Australian threshold of A$50 000 per QALY. It was only cost-effective after receptive unprotected anal intercourse exposure to an HIV-positive source. Although non-occupational PEP was a relatively well-targeted intervention in Australia, its cost-effectiveness could be improved by further targeting high-risk exposures.

Editors' note: The cost per HIV infection averted in this Australian non-occupational PEP study was estimated at well over a million dollars, with the result that non-occupational PEP was found cost-effective in Australia only in the context of unprotected anal sex with a known HIV-positive man. Such cost-effectiveness studies are useful for determining the best use of available resources in any setting for non-occupational post-exposure prophylaxis.