Comparison of self-reported and observed water contact in an S. mansoni endemic village in Brazil.

Estimates of exposure are critical for immuno-epidemiologic and intervention studies in human schistosomiasis. Direct observation of human water contact patterns is both costly and time consuming. To address these issues, we determined whether individuals residing in a Schistosoma mansoni endemic village in Brazil could accurately self-report their water contact patterns. We compared the results of a water contact questionnaire to the present gold standard, direct observation of water contact in 86 volunteers, aged 8--29. We administered a survey to estimate volunteers' frequency and type of water contact and directly measured each volunteers' water contact patterns during 5 weeks of detailed water contact observations. We found a poor correlation between self reported frequency of contact and directly observed exposure (rho=0.119, P=NS). The questionnaire data was supplemented by information about average body surface area of exposure and duration of contact for specific activities derived from observations of this cohort. This 'supplemented questionnaire' data was significantly correlated with their exposure index (rho=0.227, P=0.05). It provides a starting point from which questionnaires may develop to provide a more cost-effective and less labor intensive method of assessing water contact exposure at the level of the individual.