Project 2 People

Miguel Hernan -- Project Leader Miguel Hernan is a Professor of Epidemiology in the Department of Epidemiology and the Department of Biostatistics at the Harvard T.H. Chan School of Public Health. He is an Affiliated Faculty Member of the Harvard-MIT Division of Health Sciences and Technology and the Associate Director of the HSPH Program on Causal Inference. Professor Hernan's research and teaching are focused on methodology for causal inference, including comparative effectiveness research to guide policy and clinical decisions. In an ideal world, all decisions would be based on randomized experiments. For example, public health recommendations to avoid saturated fat or medical prescription of a particular painkiller would be supported by long-term studies that compared the effects of interventions randomly assigned to large groups of people from the target population who complied with their assignment. Unfortunately, randomized experiments are often unethical, impractical, or simply too lengthy for timely decisions. The next best thing to a randomized experiment is an observational study that closely mimics a randomized experiment. Though causal inferences from observational data are risky, the best available evidence for decision-making will often come from well designed and properly analyzed observational studies. Because there is no alternative to observational studies, there is a need to keep improving them.

Francesca Dominici -- Professor and Senior Associate Dean of Research (co-PI) Francesca Dominici is a Professof of Biostatistics in the Department of Biostatistics and Senior Associate Dean for Research at the Harvard T.H. Chan School of Public Health. Her main research focuses on developing statistical methods for integrating and analyzing large and heterogeneous datasets to evaluate the health impacts of new discoveries and interventions. She is committed to: 1) advancing statistics by disseminating new methodology that better accounts for confounding and model misspecification in drawing inferences from large observational studies; and 2) applying the newly developed methodology to large and complex databases to address critical questions in public health and to impact policy.

Xabier Garcia-De-Albeniz -- Research Associate Xabier Garcia-De-Albeniz is a Research Associate in the Department of Epidemiology in the Department of Epidemiology at the Harvard T.H Chan School of Public Health. A Medical Oncologist trained in Barcelona, Garcia-De-Albeniz earned his Sc.M. in Epidemiology from Harvard in 2012. He is a member of the board of the cooperative group GEMCAD (Multidisciplinary Spanish Group for GI malignancies research) where I collaborate designing and analyzing clinical research. His research interest moves towards any methodology that may help improving patient care as fast and accurately as possible.

Sebastien Haneuse -- Assistant Professor Sebastien Haneuse joined the Department of Biostatistics as an Assistant Professor of Biostatistics in October of 2010. The focus of his research is the development of novel study designs to address bias encountered in observational studies. To a greater extent, methods development in the statistical sciences focuses on post-hoc analysis techniques that attempt to overcome deficiencies in the data collection process. In many instances, however, there may still be insufficient information in the observed data to inform the analysis technique. Examples include inferring individual-level associations from ecological data and the adjustment for selection bias; without additional information one must resort to making empirically unverifiable assumptions. His research looks to augment the data collection process with information that can then be used to directly address the various biases.

Deborah Schrag -- Professor Dr. Schrag is a Professor of Medicine at Harvard Medical School and the Chief of the Division of Population Sciences in the Department of Medical Oncology at the Dana-Farber Cancer Institute. Dr. Schrag is a health services researcher focusing on the study of cancer care delivery. She describes the patterns and outcomes of cancer treatment in order to determine how well treatments with efficacy established in the clinical trial setting are translated into practice in non-research settings. This involves strategic use of observational and found data sources and application of statistical techniques to evaluate the impact of treatment interventions in non-experimental settings. Recent work has focussed on technology diffusion and efforts to determine what determines how rapidly new treatments are adopted and the factors that drive utilization. Her current project seeks to evaluate the quality of care delivered to indigent patients with cancer who are insured by the State Medicaid programs in New York and California. By using Medicaid enrollment and claims histories linked to other data sources including hospital discharge abstracts and tumor registry data, the goal is to prioritize areas for improving care delivery to the poor. Ultimately, the goal is to inform design of sustainable systems architecture for ongoing surveillance of the quality of cancer care.

Tyler VanderWeele -- Professor Tyler VanderWeele is a Professor of Epidemiology in the Department of Epidemiology and the Department of Biostatistics at the Harvard T.H. Chan School of Public Health. His methodologic research concerns how we distinguish between association and causation in the biomedical and social sciences and the study of the mechanisms by which causal effects arise. The current focus of his work includes the analysis of pathways, assessments of interaction, and the evaluation of network and spillover effects in which one person’s exposure will affect the outcomes of another. His research employs counterfactual theory and ideas from causal inference to clarify and formalize concepts used by epidemiologists, biomedical researchers and social scientists. His empirical work has been in the areas of perinatal, psychiatric and genetic epidemiology; various fields within the social sciences; and the study of religion and health. In perinatal epidemiology, he has worked on evaluating prenatal care indices, on the analysis of trends in birth outcomes, and on assessing the role of preterm birth in mediating the effects of prenatal exposures on mortality outcomes. In genetic epidemiology, I have been studying gene-environment interaction and the pathways by which genetic variants operate. In psychiatric epidemiology, I have been studying the feedback and inter-relationships between depression, loneliness and subjective well-being. His work in the social sciences has included the study of educational interventions, micro-finance programs, social network effects, and judicial decisions. His work in religion and health is oriented towards assessing the mechanisms by which religion and spirituality affect health outcomes.

Cory Zigler -- Assistant Professor Cory is an Assistant Professor of Bioatstistics in the Department of Biostatistics. His primary research focus is the development of Bayesian causal inference methods for problems in public health and biomedical research. Recently, he has been developing methods for assessing the public health impact of air quality regulations and for advancing Bayesian methods that use propensity scores for comparing the effectiveness of clinical treatments using large administrative claims data.

Copyright by Xihong Lin, 2011