We propose to conduct detailed studies of several unresolved issues in the descriptive epidemiology of osteoporotic fractures. Using Medicare claims data, we will focus on fractures of the hip, distal forearm, proximal humerus, pelvis, and proximal tibia occurring in people over 65 years old. Incident fractures will be identified using hospital and physician Medicare claims for 1985 and 1986 from Maine, New Hampshire, Vermont, Massachusetts and a 5 percent sample of claims for the elderly U.S. population. Corresponding population data obtained from the Medicare enrollment files will be used to compute fractures rates. Building on the experience of the investigators in using these and similar data sets, we plan to assess the effects of factors such as water fluoridation, sunlight, season of the year, race, and sex on the risk of the specified osteoporotic fractures. The research will be unique in using data referring to the entire elderly U.S. population, and in considering simultaneously through Poisson regression the multiple personal and environmental factors noted above, as they relate to fractures at several anatomic sites. We also propose to validate the utility of Medicare data for fracture epidemiology by (1) comparing the Medicare data regarding fractures with that from selected orthopedic practices in New England, by (2) carefully assessing the Medicare data itself for internal inconsistencies, and by (3) assessing the impact of treatment in Veterans' Administration hospitals on rate estimates.
Showing the most recent 10 out of 11 publications
