Environmental Risk and Meningitis Epidemics in Africa

of meningococcal meningitis occur in areas with particular environmental characteristics. We present evidence that the relationship between the environment and the location of these epidemics is quantifiable and propose a model based on environmental variables to identify regions at risk for meningitis epidemics. These findings, which have substantial implications for directing surveillance activities and health policy, provide a basis for monitoring the impact of climate variability and environmental change on epidemic occurrence in Africa. E pidemics of meningitis occur worldwide. However, the " meningitis belt " of Africa's Sahel region has the greatest incidence of the disease, with large epidemics attributed to predominantly group A meningococci; the endemic levels found in this region would be considered epidemic elsewhere. Although factors predisposing populations to meningitis epidemics are poorly understood, population susceptibility, introduction of new strains, poor living conditions, and concurrent infections have all been implicated. Epidemics occur throughout Africa in the dry season, coincide with periods of very low humidity and dusty conditions, and disappear with the onset of the rains, suggesting that these environmental factors may also play an important role in the occurrence of the disease (1–3). Lapeyssonnie (4) observed in 1963 that epidemics largely occurred in a semi-arid zone south from the Sahara, with 300–1,100 mm mean annual rainfall, and Chees-brough et al. (5) suggested in 1995 that areas that are humid throughout the year have low disease rates. In West Africa, Waddy (1958) described an area that suffered epidemics as having " … only one definable frontier, the junction of the forest…with the savanna…, when there is an abrupt change from a permanently humid climate to one with a severe dry season " (6). Epidemics have been rarely reported from the humid forested or coastal regions, even when neighboring areas are severely affected. The timing of future outbreaks is unpredictable; tools that identify the key environmental factors associated with areas prone to meningitis epidemics would help us to understand the basis for these outbreaks and eventually optimize prevention and control activities. We describe a model that predicts the probability, based on environmental information, of an area experiencing an epidemic of meningitis. Details of all known meningitis epidemics occurring before 2000 in countries comprising continental Africa were compiled from information documented in the published literature and unpublished institutional reports at the end of June 2001. All epidemics reported in the medical literature were identified in PubMed's online database …