DOCTORAL TRAINING OF AFRICAN SCIENTISTS

There are two principal rationales for doctoral training of African sc ientists in health: 1) these scientists are essential for the nations of sub-Saharan Africa to define and implement their own health priorit ies, and 2) the research they perform is essential for development. Ho wever, this training is difficult because of its expense (> $20,000 pe r year), because many developed country mentors are unaware of the rea lities of research in sub-Saharan Africa, and because major difference s in salary provide a financial disincentive to return. We describe a training strategy that reduces attrition because it is linked to the i nvestigators' responsibilities before and after training, and to home country priorities. This strategy requires a close relationship betwee n the developing country (on-site) and developed country (off-site) me ntors, with joint participation in the selection and funding process, followed by course work and short-term, independent projects off-site that lead to a thesis project in the developing country, and subsequen tly to a defined professional position in the developing country after completion of the doctoral degree. For this strategy to succeed, the developed country mentor must have both field experience and investiga tive expertise; the developing country mentor must have an understandi ng of modern biology, as well as clinical and epidemiologic experience . In addition, we would Like to emphasize that the long-term retention of these talented, highly-trained individuals requires a similar long -term commitment by their developed country mentors, well beyond the s hort term of most research funding.