THE RELIABILITY OF RETROSPECTIVE STUDIES USING A ONE-YEAR RECALL PERIOD TO MEASURE DRACUNCULIASIS PREVALENCE IN GHANA

Background. Cross-sectional studies are usually adopted to assess the one-year period prevalence of dracunculiasis. They depend upon a recal l period of up to one year. This paper aims to examine the degree of a ccuracy with which villagers in an endemic region recall the occurrenc e of dracunculiasis during the 12 months prior to a cross-sectional su rvey. Methods. Two types of study were conducted in Tolon, a village i n Northern Ghana-retrospective and prospective. Data from people inter viewed in a bimonthly survey throughout 1990 were compared with the re sults of a cross-sectional retrospective study conducted in early 1991 . Results. The results showed that retrospective studies using a recal l period of up to one year to measure dracunculiasis prevalence gave o nly 59.9% of the actual cases, but there were roughly equal numbers of 'false positive' and 'false negative' cases, so that overall prevalen ce obtained was very close to the correct figure. Conclusions. Cross-s ectional surveys may be of use to obtain estimates of the one-year per iod prevalence of dracunculiasis, but are not likely to be reliable en ough for more detailed study.