Background It is estimated that 2.27 million people are living with HIV or AIDS in India. The epidemic is concentrated and predominately driven by marginalised groups. The Avahan Programme in India is one of the largest HIV prevention programmes targeted at high risk populations within a single country in the world. It provides grants to state lead partners (SLPs), who in turn provide grants to non-government organsiations (NGOs) at the district level to deliver multi-component interventions (including peer outreach, STI services, and structural interventions). This study presents the costs of implementing these interventions to female sex workers, men who have sex with men and transgender from 23 districts in the four Southern states of India over 4 years.
Methods Financial and economic costs were prospectively collected. Costs by input and activity, unit costs of interventions between 2004 and 2008 were analysed. Economic costs are presented in US $ 2008 using 3% discount rate.
Results The total economic cost of the intervention over 4 years for 23 districts was US$ 27 341 121(range per district US$307 597–US$3 147 790) for an estimated target population of 93 345. In year 4, the average annual economic cost per estimated population at the NGO level across the 23 districts was US$70 (US$30–174). The average annual financial cost at current prices was US$59. Over the four years, at the NGO level, capital costs accounted for 10% of total costs. The main recurrent costs were personnel costs (46%) and the materials and supplies for sexually transmitted infections (STI) services (13%). Examining both SLP and NGO costs, programme management, information and grant management costs activities accounted for 27% of total cost; followed by capacity building (20%), STI services (19%), peer outreach (including behaviour change communication, condom provision) (16%) and structural interventions (11%). The proportion of cost that was spent on direct services such as peer outreach and STI services increased as the programme scaled up. Costs for structural activities also increased from 5% in year 1 to 11% by year 4.
Conclusions Assessing costs over the life of the project helps to identify how costs vary with the changing needs and strategies of the programme. We will conduct further analysis to examine which factors most influence costs (local price/wages, programme intensity, community involvement etc). This cost data can assist the realistic planning of large scale long term HIV prevention programmes in the future.