An exploration of awareness bias in two environmental epidemiology studies.

Awareness bias in environmental health research is the tendency to report more illness because of concerns arising from proximity to a hazard in the absence of a measurable biological effect. Such bias complicates the interpretation of self-reported symptoms in communities exposed to emissions from heavy industry. We used data from two epidemiologic studies in Northeast England where community concerns existed about health risks from industry. An association between proximity to industry and self-reported respiratory and nonrespiratory illnesses and symptoms had been found in one study but not in the other. An indicator of concern about industrial pollution was constructed from responses to a 17-item questionnaire about issues that had caused stress or anxiety. Univariate and multivariate analyses of health outcome variables in both studies showed that individuals with "industry-related worries" reported more illness, irrespective of proximity to industry. We conclude that self-reported illness was influenced by both worry and proximity to industry, but that worry about the hazard had the greatest effect on self-reported illness. We suggest that because absolute certainty about the role and extent of awareness bias in environmental epidemiology studies is unlikely to be achieved, self-reported data should be supplemented with other observations.