AIDS: Modelling and Predicting

The three authors are to be warmly congratulated on their substantial, yet complementary, papers on an issue of major public health moment – how to get to grips quantitatively with the HIV-AIDS epidemic. Advances in the treatment of HIV disease are proving disappointing, more disappointing now (mid 1993) in fact than at any time in the past several years. Vaccine development too has been largely a succession of false dawns. Primary prevention, however, is demonstrably achievable, as seen by the shape of the HIV epidemic curve in male homosexual populations in much of the developed world, and injecting drug users in at least some communities. Individuals have an underrated capacity to respond to credible information, clearly witnessed in other chronic disease areas by the sharp declines in lung cancer and cardiovascular disease seen in, for example, the US, Britain and Australasia. The most pressing public health need in AIDS is for accurate estimates of the rate at which HIV transmission is occurring in different subsections of the community, defined by age, sex, geography and particularly transmission category. Only then can public health messages be appropriately shaped and directed, and more importantly, be made credible to the subpopulations concerned. There is little point in urging young adults to refrain from sexual pleasure if the message is not clearly substantiated by the facts. HIV disease is, however, a complex temporal process; there is a wide gap between the observations we can currently make on the disease in the population and the conclusions that one would like to draw about the pattern of HIV infection.