Databases in MS research: pitfalls and promises.

A database is an organized repository of data. Prospective collection of patient information in a database ('databasing') has been attempted by a few consortia of MS investigators over the past 10 years. This approach promises to facilitate epidemiologic research in MS and investigation of the natural history of the disease and how it might be altered by long-term treatments such as interferon beta. Databasing has some advantages over clinical trials in assessing new therapies, primarily because the focus is on long-term effectiveness in an entire population rather than short-term statistical significance in a highly selected population. The limitations of databasing and strategies to overcome these limitations are addressed.