Exposure Measurement Error Correction in Longitudinal Studies with Discrete Outcomes

• Activity Number: 144 - Methods for Missing and/or Misclassified Data
• Type: Contributed
• Date/Time: Monday, August 8, 2022 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): Ce Yang* and Molin Wang and Donna Spiegelman and Ning Zhang and Unnati Mehta and Jamie Hart
• Companies: Harvard University and Harvard T.H. Chan School of Public Health and Yale School of Public Health and University of North Carolina at Chapel Hill and Harvard School of Public Health and Harvard School of Public Health
• Keywords: Air pollution; generalized estimating equation; longitudinal data; measurement error; anxiety

Abstract

Environmental epidemiologists are often concerned with estimating the effect of functions of time-varying exposure histories on health outcomes, for example, anxiety disorders. However, the individual exposure measurements that constitute the history upon which an exposure history function is constructed are usually subject to measurement errors. To obtain unbiased estimates of the effects of such mismeasured functions in longitudinal studies with discrete outcomes, a method based on the main/validation study design is developed. In particular, various estimation procedures are explored when an internal validation study presents. Simulation studies were conducted to assess its performance compared to standard analysis, and we found that the proposed method had good performance in terms of finite sample bias reduction and nominal coverage probability. As an illustrative example, we applied our proposed methods to a study of long-term exposure to PM2.5, in relation to anxiety disorders in the Nurses’ Health Study.