David Harrington is involved in two different lines of research. The first is research in statistical methods for clinical trials and prospective cohort studies in which the time to an event is a primary outcome. He has worked in efficient nonparametric tests and regression methods for right-censored data, sequential designs for clinical trials, and nonparametric methods for estimating nonlinear covariate effects on survival. Recently, he and co-workers in the Department of Biostatistics have been studying methods for analyzing survival data when some covariates have missing observations. Missing data are common in both prospective and retrospective cohort studies, and simply ignoring cases with missing observations can lead to substantial biases in inference.
Dr. Harrington is also involved in collaborative research in cancer. He was the principal investigator of the Statistical Coordinating Center for the Cancer Care Outcomes Research and Surveillance (CanCORS) Consortium. This NCI-funded study is a network of sites around the country that are conducting a population based study of access to and outcomes from cancer care, with special focus on ethnic subgroups and subgroups defined by age. Dr. Harrington's primary interest is in hematologic malignancies, and he has worked in that area with Dana-Farber investigators as well. He and Margaret Shipp of the Dana-Farber co-chaired the International non-Hodgkin's Lymphoma Prognostic Factors Project, a collaboration of United States, Canadian and European treatment centers. This project produced an internationally agreed upon definition of risk factors for early relapse or death in patients with aggressive non-Hodgkin's lymphoma, and those definitions are now being used at many centers to decide which patients should be candidates for intensive chemotherapy, and which should be candidates for more standard approaches. He also works in inference in statistical genetics and longitudinal data.
Dr. Harrington is currently the Acting Chair of the Statistics Department in the Faculty of Arts and Sciences, Harvard University