I've noticed a recent spate of empirical papers assessing med mal, especially studies endeavoring to examine potential impacts flowing from various state tort reforms, drawing on the National Practitioner Data Bank Public Use Data File ("NPDB"). The Department of Health and Human Services ("HHS") maintains this data file in accordance with the Health Care Quality Improvement Act of 1986. Since September 1, 1990, laws require that (with some exceptions) all medical malpractice payments must be reported to HHS within 30 days of payment. Like any other database, however, NPDB is far from perfect. For example, NPDB records do not satisfactorily include the individual characteristics of each case (e.g., injury severity, plaintiff age, physician specialty, etc. are either not or poorly reported). Data file limitations aside, those interested in this literature can get more info on the NPDB here.