Community exposure to hazardous waste disposal sites: assessing reporting bias.

A household health survey of residents living near two hazardous waste disposal sites in Calcasieu Parish, Louisiana was conducted in 1981-1982 as part of a comprehensive study of the effects of those sites on the environment and on the health of nearby residents. An unexposed community was included in the health survey for comparison. Due to media coverage and public concern about the sites, two potential indices of reporting bias, hypochondriasis and respondent's opinion about the environmental effects of waste sites, were included in the survey. Because air and water quality data showed little evidence that hazardous concentrations of chemicals were being released from the sites, questions were raised about the interpretation of the health survey data. The data were analyzed, therefore, for the association between symptom reports and the potential indices of reporting bias. Hypochondriasis scores were associated with symptom reports regardless of location of residence while an individual's opinion showed a different pattern by area of residence. Respondents living near one of the waste disposal sites who answered "yes" to the opinion question were 2-3 times more likely to report some types of symptoms than residents of the comparison community. In contrast, there was little difference in symptom reports between the exposed and comparison communities for those answering "no" to the opinion question. The usefulness of self-reported symptom data in studies of communities near hazardous waste disposal sites is discussed, and attention is called to the need to develop measures sensitive to reporting bias in epidemiologic studies.