How Useful Is Linear Regression Analysis in Detecting the Existence of Dose–Response Relationships in Large-Scale Epidemiologic Studies When Only a Fraction of the Population Is Sensitive? The Case of Methylmercury

The existence of a dose response in epidemiologic studies is generally determined from the linear regression slope after controlling for covariates. This approach assumes the entire population is equally sensitive to the toxicant and that response is a function only of dose and a random error function. However, sensitive subpopulations have been identified for a variety of toxicants possibly including methylmercury (MeHg). The study of MeHg exposure in the Seychelles Islands has failed to find significant effects (dose–response slope not significantly different from zero) while other studies have found such effects. Using data on the error function in developmental test scores and MeHg exposure distributions from that study, and assuming plausible dose–response relationships for sensitive subpopulations, we conducted Monte-Carlo simulations of the power of linear regression analysis to detect a dose–response relationship from the total sample (n=700), and to compare dose–response slopes in the total and sensitive populations. Linear regression did not reliably detect a dose–response relationship for most scenarios when sensitives were 5% of the total and for some scenarios when sensitives were 10% of the total. We also found that the dose–response slope for the total population underestimated the sensitive dose–response slope in all cases by about an order of magnitude. These findings may have important implications for detection and quantification of dose–response relationships from epidemiologic studies.