Despite widespread implementation of a broad range of HIV (human immunodeficiency virus) prevention strategies and programmes in the last decade, the evaluation of interventions has progressed little beyond quantifying such indicators as the numbers of condoms distributed rather than the numbers of cases of HIV prevented. Economic evaluation in any form is thus poorly represented in the published literature. This absence of rigorous evaluation of costs and effectiveness of HIV prevention strategies occurs against a backdrop of increasing resource constraints and the related need to consider the opportunity costs of any expenditure. For any given objective or set of objectives, governments and other funding agencies are increasingly aware of the need to choose the strategy or combination of strategies bearing the least cost.
This paper presents the results of a cost analysis of HIV prevention strategies as a step towards developing the literature on economic evaluation. Cost analysis can indicate relatively low-cost strategies and provide useful information on affordability to those who are involved in the planning, management and evaluation of health strategies in developing counties [1, 2]. Similar disease-specific cost analyses have been performed before in the health sector; for example, Barlow & Grobar  reviewed the costs of controlling parasitic diseases, and Phillips et al.  reviewed the costs of selected intervention for the prevention of diarrhoea. In addition to presenting data on the costs and outputs of HIV prevention projects, this paper discusses some of the difficulties faced in economic evaluation in this area, and suggests directions for further research.
Case studies of currently operating programmes representing six broad categories of HIV prevention strategies were chosen for study. The six strategies were:
(1) Promotion of safer sexual behaviours through mass strategies, which usually includes teaching essential facts about HIV, promoting healthy behaviour, reducing anxiety about casual transmission, and preventing discrimination against those infected . This strategy usually involves minimal targeting of vulnerable groups.
(2) Promotion of safer sexual behaviours through person-to-person education, which involves more intensive intervention targeted at vulnerable groups for which the potential benefit from a given input is perceived to be high.
(3) Provision of condoms through social marketing, which encompasses market research, product importation, branded packaging, advertising, and distribution (with retail costs often subsidized by donors or government) .
(4) provision of sexually transmitted disease (STD) treatment and prevention services, which is important because STDs both rank among the top ten most important health problems in developing countries in terms of healthy life-years lost  and can facilitate transmission of HIV .
(5) Prevention of unsafe drug use behaviours for intravenous drug users (IVDUs) who are unable or unwilling to enter drug abuse treatment programmes, which includes education in the need to eliminate needle sharing, instruction in disinfecting contaminated injection equipment, bleach distribution, and needle exchange.
(6) Provision of a safe blood supply for transfusion, which involves screening donated blood for HIV and the disposal of infected blood.
A sample of case studies, representing each of these HIV prevention strategies, was sought on the basis of availability of cost and output data, and potential generalizability. As far as possible at least one case study for each strategy was selected from each of the three broad income categories of developing countries as defined by the World Bank (low, low-middle, and upper-middle income countries) . In addition, a fourth category, that of eastern European, former socialist economies, was added. …