INTERNAL VALIDATION OF MEDICARE CLAIMS DATA

The medicare database is commonly used for epidemiology and health ser vices research, but validation of its data by chart review or question naire may be difficult and expensive. Since hospital and physician bil ls are independent in Medicare, however, these two data sources can be used to supplement and corroborate each other. This ''internal valida tion'' is illustrated here for hip fracture and prostatectomy. Agreeme nt of the hospital and physician data streams regrading site of hip fr acture (neck vs other), treatment of hip fracture (internal fixation v s arthroplasty), and type of prostatectomy (transurethral resection of prostate, open, or radical) was excellent with percentage of agreemen t generally between 89% and 99%, and kappa statistics typically betwee n 0.74 and 0.95. When validation with outside data sources is not read ily available, such internal validation of Medicare data may be valuab le.