Professor of Epidemiologic Methods
I am one of the few people in the world with a joint doctorate in Biostatistics and Epidemiology. As a result, I can freely speak the languages of both disciplines, and switch between the two cultures, playing the role of interlocutor for either. My own research is motivated by problems which arise in epidemiology and require biostatistical solutions. In particular but by no means exclusively, I have focused on methods for study design and data analysis which reduce bias in estimation and inference due to measurement error or misclassification in the exposure variable. I have recently received funding from the NIH to embark on an extensive project of methods development and re- and new analysis of several major studies of the effects of long-term exposure to constituents of air pollution on the risk of overall, cardiovascular and lung cancer mortality. This work is joint with colleagues in the Netherlands, Israel, and the University of Washington in Seattle. The goal is to substantially reduce, if not eliminate, exposure measurement error as a major source of bias in the available results to date, and involves solving challenging mathematical and computational problems in the realm of survival data analysis.
My website is one of the most visited at HSPH, because it contains much user-friendly well-documented freeware implementing non-standard methods useful in epidemiologic research. I am the statistician for the Nurses’ Health Study 2, the Health Professionals Follow-up Study, the Pooling Project of Prospective Studies of Diet and Cancer in Men and Women, the Harvard PEPFAR Dar es Salaam site, Trials of Vitamins in Dar es Salaam, and the multitude of spawn of these efforts.
My most recent interest has been to work actively with others in Epidemiology, Nutrition, Environmental Health, Global Health, and SHDH, to greatly increase global public health efforts at HSPH. In particular, I am interested in developing, testing and implementing public health oriented preventive interventions for achieving Millenium Development Goals 4 & 5, to complement health systems strengthening efforts of great interest currently, and developing, testing, and implementing preventive interventions to abate the global diabetes and cardiovascular disease epidemics. Expertise in monitoring and evaluation are a critical specific contribution that I bring to the table in these endeavors.
The role of adjusting for measurement error and misclassification in causal inference; implementation science; big data, ‘omics and the study of disease heterogeneity Cutter Symposium, Boston, MA, November 2013
Methods to Correct Measures of Effect for Bias due to Exposure Measurement Error Statistical Society of Canada Introductory Overview Lecture, Edmonton, Alberta, May 2013