Refining case-mix adjustment. The research evidence.

We review case-mix adjustment, which is the process of adjusting for differences in the cases treated in different hospitals so that their costs or outcomes can be compared. We examine the Medicare payment system, which rests on case-mix adjustment, and identify areas, including outlier payments, in which payment accuracy might be improved without better measurement of the severity of illness. There is no available measure of severity of illness that would produce a large improvement in the accuracy of Medicare payments if used to supplement or replace the system of diagnosis-related groups. Evidence regarding whether better measurement of severity would substantially change the distribution of payments across hospitals is mixed. Considerable evidence suggests that the intensity of medically appropriate treatment for patients in the same diagnosis-related group varies substantially for reasons other than the severity of illness. Despite great demand for measures of the quality of care, important technical problems must be solved before we can be confident that differences in case-mix-adjusted outcomes reflect differences in the quality of care.