As part of my training as a specialist registrar in infectious diseases and tropical medicine, I have spent the past year in Malawi learning “real” tropical medicine. Of the many things I have learnt in this year, the most striking has been that tropical medicine is not quite what I thought it would be. I had expected to become familiar with the classic tropical diseases: malaria, schistosomiasis, and trypanosomiasis, to name a few.

But, in reality, I have seen little of these diseases. The reason is obvious: one disease has surpassed all others in terms of morbidity and mortality, pushing aside the other, more typical, tropical diseases. That disease is, of course, HIV. About 75% of the adult general medical inpatients are HIV positive at the hospital where I have been working. The huge economic and social impact of HIV on populations in Africa has been well described. Less well described has been the impact on the working practice of a tropical physician.

Tropical physicians are used to the idea of rationing resources, and it seems that in the pre-HIV days a way of practising had evolved to limit excessive spending of scarce resources by knowing the local epidemiology of diseases and relying strongly on clinical skills. For example, it is quite feasible to diagnose lobar pneumonia without chest radiography and the full array of “routine” blood tests that a pneumonia patient …