levels of social organization, mitigation interventions divert scarce resources from other uses, including efforts to prevent transmission. Thus, the value to society of any mitigation intervention should be as least as great as the cost of the resources devoted to the effort. Research on this issue might improve the efficiency of current expenditures, as well as justify a case for or against additional spending.

A great deal more is known about designing and implementing HIV-prevention programs than is known about providing care to the millions of people in sub-Saharan Africa already infected with the virus. Simple, cost-effective solutions to daily living problems faced by persons with AIDS, such as palliative care, part-time home care, and group counseling, may make larger, more expensive interventions unwarranted.

Research on impact should incorporate both qualitative and quantitative approaches to data collection and should evaluate both short-and long-term effects. Of particular interest is research that would permit an understanding of the impact of HIV/AIDS on poverty and on individual decision making. Research is needed to ascertain whether decreased life expectancy reduces willingness to save or invest in financial and real assets, in human capital, and in the relationships necessary to maintain social interactions. In the long term, the impact of HIV/AIDS on sub-Saharan Africa will depend on the strength and malleability of social and economic networks in accommodating the changes that are occurring.

The epidemic has already affected millions of households in sub-Saharan Africa and will continue to do so for at least the next 20 years. Efforts to mitigate the effects of the disease have been uncoordinated and poorly targeted, and their