Do all teaching hospitals deserve an add-on payment under the prospective payment system?

Although teaching hospitals are widely discussed in terms of 115 university teaching hospitals, Medicare's prospective payment system pays about 1,000 teaching hospitals an additional amount per case. This paper investigates two frequent justifications for this add-on payment: 1) teaching hospitals have sicker patients, and 2) they produce medical education. To test assertion 1, I reviewed case-mix analyses and performed regression analyses, which provided little evidence that nonuniversity teaching hospitals have sicker patients. With regard to assertion 2, I argue that the first 10 or 15 residents per 100 beds function as hospital-based and attending physicians and, accordingly, should not be included in the add-on payment. I conclude with two policy options that would focus the add-on payment on university hospitals.