Recent yellow fever epidemics in West Africa have underlined the discrepancy between the official number of cases and deaths and those estimated by a retrospective epidemiological investigation. During the yellow fever epidemic that broke out in western Mali in September 1987, a total of 305 cases and 145 deaths were officially notified, but estimates revealed true figures abut five times higher. This paper attempts to discuss the factors that hindered early case detection and more complete reporting. They were, first, the insufficient training on the clinical diagnosis, the blood sampling method for laboratory confirmation, and the curative treatment of patients (resulting in low utilization of services); second, the lack of an action plan to prepare in advance a quick response to the epidemic, affecting reporting procedures at the peripheral level and active case‐finding during the outbreak; and third, the lack of laboratory facilities for a quick confirmation of the disease. The difficulties experienced during the yellow fever epidemic in Mali demonstrated the importance of a preparedness strategy for epidemic control, based on an integrated approach of epidemiological surveillance within basic health service activities. The need for regional collaboration and for institutionalized funds in the donor community that could be mobilized for epidemic preparedness activities is also emphasized. Copyright © 1990, Wiley Blackwell. All rights reserved
