Malaria remains an important cause of death, especially in sub-Saharan Africa. Self-treatment with antimalarial drugs is a common practice that raises important issues for policy-makers. A number of important questions concerning factors related to self-treatment, adequacy of self-treatment and the role of self-treatment in malaria mortality remain unanswered. Although there are some common patterns, there is considerable diversity in treatment practices, even within a single country. Social science research on malaria treatment needs to move beyond description to evaluation of interventions. This will require a greater degree of methodological rigour and more attention to the generation of data that can be compared across time periods and studies. Definitions of malaria cases and the role of local disease categories in identifying cases need to be made more explicit. Illnesses should be classified by severity, using measures of perceived severity as well as biomedical signs of severity. Each treatment step should be considered in terms of four levels of analysis: who provided the treatment or advice, what the treatment was, where it was obtained and when it was taken in relationship to onset of illness.
