ABSTRACT
In order to enable the assessment of schistosomiasis control, it has to be monitored. The methods and techniques employed for monitoring have to be validated in turn. In the Malian National Schistosomiasis Control Programme, only samples of the village population are examined. For practical and economic reasons, the sampling procedure is not free of bias from self-selection. When rates of samples are compared with results of examining the whole population no significant difference between the rates was found. Infection rates were precise to about +/- 5%. The sensitivity and specificity of laboratory technicians put to the task of identifying and counting the eggs of Schistosoma haematobium or S. mansoni in urine or stool samples differed considerably. The lowest scores were obtained by a laboratory technician from a district health centre who accompanied the central control team for a month. The predictive values of samples read positive dropped significantly below a general prevalence of 20%. The technician from the district centre would read samples falsely positive in more than 70%. However, the same person would read only 10% of slides falsely negative, once prevalence has dropped below the threshold of 20%, when maintenance depends on the performance of basic health services. The amount of stool used for a Kato-Katz preparation is only a 25th of one gram. The distinction between light and intensive infections with S. mansoni is therefore based on counting 1 to 3 or 4 or more eggs in a preparation. In a poisson distribution with mean 1, the difference between 1 and 4 is not significant.(ABSTRACT TRUNCATED AT 250 WORDS)
