Improving data quality in community-based seafood consumption studies by use of two measurement tools.

A seafood consumption study w as conducted in Glynn County, Georgia, to address concern about bioaccumulation of mercury from a nearby hazardous waste site in people who ate potentially contaminated seafood from this area. Seafood consumption levels were ascertained with two data collection tools: a questionnaire and a dietary diary. The use of two instruments allowed for more detailed analysis to reveal discrepancies in responses between the two instruments, to improve reliability of study results, and to reduce recall bias. Implementation of the questionnaire was relatively easy and provided a broad characterization of consumption patterns in the area. The dietary diary was more time-consuming, resulting in a reduction in participation rates. It provided, however, more detailed information with which to address community concerns about adverse health effects from mercury exposure. Overall, individuals who participated in this study were able to make broad generalizations about the amount of seafood in their diet but were less accurate in estimating specific seafood consumption levels. In addition, the level of concordance between the questionnaire and the dietary diary was low with respect to seafood consumption levels. For investigators examining consumption patterns in a community, the decision to use a questionnaire, a dietary diary, or both will be influenced by the objectives of the study, the level of community concern, the number of study staff, and available resources.