I found the case very interesting. Perhaps because it is closely related to the kind of work I do as an Operations Analyst for a healthcare system. As was mentioned in another opening post, Oakley and his team failed to take seasonality into consideration. Two weeks in January are surely not representative of an entire year, especially in terms of patient volume; thus annualizing two weeks data is an obvious error. I would expect the patient volume to be lower than average in January as it is not a peak of cold/flu season and patients prefer to have their annual checkups during warmer seasons (mind you the hospital is in Massachusetts). This hunch is confirmed by analysis of the data in Exhibit 8: if we take values for Physician Visits/Year, combined NP – Patient & Employee Visits/Year and Intern/Resident Visits/Year and convert them to Visits/Day (taking into account that there are 8 Physicians, 1 Nurse Practitioner and 13 Interns/Residents at the clinic) we will get values of 5.3, 5.7 and 1.0 Visits/Provider/Day, which are very low (48 work weeks/year and 5 days/week work schedule assumed). Therefore, either PCU is overstaffed or more representative time span should be taken.

Another factor that Oakley’s team overlooked is the fact that the teaching hospitals receive direct and indirect Graduate Medical Education payments from Medicare that partially compensate for residency education costs and for higher patient care costs due to presence of teaching programs. According to Donald A. Young et al, the authors of “Medicare and the American Health Care System: Report to the Congress” (June 1997, ISBN: 9780788146763), in fiscal year 1995, the average “per resident” Medicare payment amount was about $67,000. When considering the cost of care provided by interns and residents this payment needs be taken into account and cost adjusted accordingly.

I am a bit