Objective: To estimate cost-effectiveness and capacity requirements for providing antiretroviral drugs to pregnant HIV-infected women in rural South Africa.
Setting: Hlabisa health district, where HIV prevalence among pregnant women was 26.0% in 1997.
Methods: Calculation of the number of paediatric HIV infections averted under three scenarios, and their cost. No intervention was compared with scenario A (zidovudine delivered within current infrastructure), scenario B (zidovudine delivered through enhanced infrastructure), and scenario C (short-course zidovudine plus lamivudine delivered through enhanced infrastructure). Cost-effectiveness was defined as cost per infection averted and cost per potential life-year gained. Capacity was determined in terms of staff and infrastructure required to effectively implement the scenarios.
Results: With no intervention, 657 paediatric HIV infections were projected for 1997. In scenario A this could be reduced by 15% at a cost of US$ 574 825, in scenario B by 42% at US$ 1520770, and in scenario C by 47% at US$ 764901. In scenario C, drugs accounted for 76% of costs, whereas additional staff accounted for 18%. Cost per infection averted was US$ 2492 and cost per potential life-year gained (discounted at 3%) was US$ 88. Cost of scenario C was equivalent to 14% of the 1997 district health budget. At least 12 extra counsellors and nurses and one laboratory technician, together with substantial logistical and managerial support, would be needed to deliver an effective intervention.
Conclusion: Although antiretrovirals may be relatively cost-effective in this setting, the budget required is currently unaffordable. Developing the capacity required to deliver the intervention would pose both a major challenge, and an opportunity, to improve health services.