Health Research and Policy


DATE: March 14, 2013
TIME: 1:15 - 3:00 pm
LOCATION: Medical School Office Building, Rm x303
TITLE: Measurement error correction for survival data analysis with covariates that are functions of time-varying exposure histories
SPEAKER: Donna Spiegelman
Professor of Epidemiologic Methods
Departments of Epidemiology and Biostatistics, Harvard School of Public Health

Nutritional and environmental epidemiologists are often interested in estimating the prospective effect of functions of a time-varying exposure history such as the cumulative exposure variable -- the sum of all exposures up to the present time -- or the cumulatively updated average exposures -- the running average of all exposures up to the present time -- in relation to chronic disease endpoints such as cancer and cardiovascular disease incidence and mortality. By re-calibrating the measurement error model within each risk set, a risk set regression calibration (RRC) method has been proposed for Cox models in this setting. An algorithm for a bias-corrected point estimate of the relative risk using an RRC approach is developed, followed by the derivation of an estimator of its variance. New developments take into account the correlation structure of the repeated exposure measurements within subjects, which impact both the point estimate and its variance. Emphasis is on methods applicable to the main study/external validation study design, which is standard in nutritional and environmental epidemiology. Limitations of the current validation study designs are discussed and partial solutions to these limitations are proposed. Simulation studies under several realistic assumptions about the error model and correlation structures were conducted, which demonstrated the validity and efficiency of this method in finite samples. The method is applied to a study of long-term exposure to fine particulate matter in air pollution in relation to all-cause mortality in the Nurses' Health Study. Time permitting, meta-regression methods to extrapolate the available data beyond the locations where it was collected will be discussed. User-friendly public software is available implementing these methods and, time permitting, use will be demonstrated (

This is joint work with Xiaomei Liao and Samuela Pollack.

Suggested readings:
Liao X-M, Zucker DM, Li Y, Spiegelman D. Survival Analysis with Error-Prone Time-Varying Covariates: A Risk Set Calibration Approach. Biometrics, 2011.

Spiegelman D. Approaches to uncertainty in exposure assessment in environmental epidemiology. Annual Review of Public Health, 2009; 31:149-63.

Footer Links: