A. James O’Malley, PhD, is an associate professor of health care policy, with a specialty in statistics, in the Department of Health Care Policy at Harvard Medical School. Problems in medicine and health care policy have motivated Dr. O’Malley’s methodological research, which focuses on analysis of complex survey data, latent variable modeling, social networks, causal inference, and statistical applications in cardiology. Dr. O’Malley is developing likelihood-based methods for estimation of multilevel covariance structures. This work will extend to research areas involving surveys with structured nonresponse, as in the Consumer Assessment of Healthcare Providers and Systems (CAHPS) study. He is also developing methods for causal inference for randomized clinical trials afflicted by noncompliance and subsequent nonresponse (such phenomena are quite common in psychiatry and other areas of mental health), and is in the process of developing related methods for observational data analysis.

Dr. O’Malley’s newest methodological interest is the analysis of longitudinal social network data. For the past several years, Dr. O’Malley has been interested in evaluating new technologies, particularly devices. He was instrumental in developing a new statistical approach to predict the outcomes that would be expected had the patients tested with a new device actually received the control device instead. This approach was applied to coronary artery stents and has recently been considered for trials of other types of devices.

Dr. O’Malley received his BSc (honors) and PhD degrees in statistics from the University of Canterbury in Christchurch, New Zealand. He then spent the last two years of his PhD at Purdue University, where he simultaneously earned an MS in applied statistics.