Dose Response Modeling in Epidemiologic Cohort Studies


Principal Investigator:
Ellen Eisen, Adjunct Professor of Environmental Health

Department of Environmental Health and Department of Epidemiology


Dates of Research:
May 5, 2004 — April 30, 2009


This proposal builds on current work applying semiparametric statistical methods to model nonlinear exposure-response curves with smooth functions of exposure in occupational cohort studies (R01CA81345). In the course of adapting penalized splines for survival analysis, there have been two issues identified that need further development: 1) criteria for choosing the optimal amount of smoothness; and 2) diagnostic statistics to identify influential observations and measure their influence on the fitted values of the curve. These issues will be examined in the presence of the right skewed distributions typical of workplace exposures. Comparisons willl be made of the goodness of fit and resistance of penalized splines to that of two other common smoothers: restricted cubic spines, and a locally weighted regression smoother called “”loess.”” They will re-analyze two cancer mortality studies:larynx, lung, prostate and rectal cancer in autoworkers exposed to metalworking fluids (MWF) and lung cancer in California’s diatomaceous earth industry exposed to crystalline silica. The methodological investigations will focus on MWF and silia because, by clarifying the shapes of their exposure-response curves, one can make a substantial contribution to the current controversies over their control in the workplace.