The majority of children infected with Plasmodium falciparum in areas of stable endemicity do not develop severe, life-threatening disease. It is important to identify risk factors for the minority who do. Case-control studies in which children with severe disease are compared with children with non-severe disease and with community controls, avoid some of the ethical and logistical problems inherent in such an undertaking. This paper discusses methodological aspects of case-control studies of severe malaria including case and control definitions, selection of cases and controls, potential risk factors, sample size calculations and analysis. Although specifically concerned with malaria, many of these issues are equally relevant to case-control studies of other infectious and parasitic diseases in a tropical environment.