Epidemiologic studies of organic micropollutants in drinking water.

Epidemiologic studies have been conducted in order to make a quantitative statement about associations between drinking water contaminants and disease. The basic measures of the association are a rate ratio or relative risk and rate difference or attributable risk. The appropriateness of this measure is dependent on components of study design, data collection, and the analysis of epidemiologic data, and these must be evaluated for each study to determine precision (lack of random error) and validity (lack of systematic error). Internal validity includes considerations for preventing selection bias, minimizing observation bias, and assessing, preventing, and controlling confounding bias within a particular study. No single epidemiologic study is likely to provide a definitive answer, and the results of epidemiologic studies must be interpreted in the context of other scientific information. Epidemiologic studies of organic micropollutants in drinking water have been reviewed and are summarized based on these considerations.