In southern Africa, HIV/AIDS is considered to be a critical factor conditioning rural economic development , exacerbating already difficult problems with climatic variability and poverty. In their efforts to use household surveys to obtain information on rural adult mortality and morbidity and their effects on rural household livelihoods, Michigan State University researchers and their local collaborators in the Ministries of Agriculture of Rwanda (MINAGRI) and Mozambique (MADER), have learned various lessons. Using household surveys to estimate the impact on the households requires careful attention to detail as well as skilled use of econometric tools. The difficult modeling issues involved in such estimation is not be discussed in this paper. Rather, we focus on the basic data collection required and the formulation of effective survey questions. Based on the survey instruments used in these two countries, suggestions are made to improve the ability of researchers to estimate mortality rates, evaluate changes in demographic composition of the households, and elicit information from households regarding incidence of illness and death, their effects on household livelihoods, and the households' response strategies. This paper provides recommendations for future survey instruments to improve the knowledge base so critical for the design of interventions.