at the National Institutes of Health, and a pediatric surgery-transplant surgery fellowship in Cincinnati. Her research in mixed chimerism to induce tolerance to organ allografts and treat nonmalignant diseases such as sickle cell anemia and autoimmune disorders is being applied clinically in six Food and Drug Administration-approved phase I trials. Dr. Ildstad holds several patents related to her research in expanding bone marrow transplantation to treat nonmalignant disease by optimizing the risk-benefit ratio through graft engineering and partial conditioning. She is the founding scientist of Chimeric Therapies, Inc., a biotechnology company focused on bone marrow graft engineering, and she serves on the board of directors of the company. Dr. Ildstad is a member of the Institute of Medicine, is serving as correspondent for the Committee on Human Rights, and is a member of the Institute of Medicine Committees on Organ Procurement and Transplantation and Multiple Sclerosis: Current Status and Strategies for the Future.
Bruce Levin, Ph.D., is professor and head of the Division of Biostatistics at the Columbia School of Public Health. He received a Ph.D. from Harvard University and has been on the faculty at Columbia University since 1974. His research interests are in biostatistics, data analysis, logistic regression, and sequential analysis. He is principal investigator of the Statistics, Epidemiology, and Data Management Core of the HIV Center for Clinical and Behavioral Studies. He is the senior statistical consultant on several multicenter randomized clinical trials in the field of stroke neurology and cardiology, including the Warfarin and Aspirin Recurrent Stroke Study and its collaborative trials. Dr. Levin has also served for the past 8 years as a consulting statistical editor for the American Journal of Public Health. He has a long-standing interest in the intersection of statistical inference for categorical data and computational methods in statistics. This interest has resulted in the development of analytical methods for a variety of problems from a unified viewpoint (that of finite and infinite dimensional linear exponential families). A product of this work is the unique computing program MCLA for maximum conditional likelihood analysis of polytomous outcome data in large-scale epidemiological data sets with arbitrary levels of stratification. Dr. Levin also studies general empirical Bayes methods (e.g., to test homogeneity hypotheses in large sparse data sets). Using these methods, Dr. Levin explores alternative designs for clinical trials, including nonrandomized designs that ensure allocation of new treatment to those subjects in most dire need of therapy. A related area of interest is sequential trial design, both classical and innovative (e.g., designs that minimize ethical costs).