VA Residency Training Program Rankings May Predict Cost of Care but not Hospital Readmission or Mortality Rates
FINDINGS:
- Veterans treated by a team of VA physicians affiliated with a higher-ranked medical/surgical residency training program had 10% lower healthcare costs compared to Veterans at the same hospital who were treated by a team of VA physicians affiliated with a lower-ranked training program – and up to 25% lower costs for more complicated conditions (e.g., heart failure, COPD).
- Differences in cost largely were the result of diagnostic-testing rates: the physician team affiliated with the lower-ranked program took longer to order tests, and ordered more of them.
- Hospital readmission rates and mortality were unrelated to the physicians’ training program.
BACKGROUND:
A major question in healthcare is the underlying source of geographic variation in spending. These differences are often ascribed to different preferences and training among physicians. Graduate medical education is part of VA’s statutory mission, and VAMCs are located near academic medical centers to enhance training: 107 of the 126 U.S. medical schools are affiliated with a VAMC. Investigators in this natural experiment analyzed data from nearly 30,000 Veterans from one large, urban VA hospital who had been randomly assigned by standard hospital procedure to teams comprised of physicians affiliated with one of two medical/surgical residency training programs. One program was affiliated with one of the higher-ranked medical schools in the U.S., while the other program ranked lower according to measures such as the pass rate for Board exams. Patient characteristics were similar across teams, and both teams had access to the same facilities, nursing staff, and specialists. In comparing physician teams associated with the two programs over 13 years, investigators assessed length of hospital stay, readmissions, number and timing of procedures, healthcare costs, and mortality.
LIMITATIONS:
- This study focused on only two residency training programs and one VA hospital.
- Variations in care might be explained both by the differences in the quality of physicians accepted into the programs and the quality of their clinical training; it is difficult to separate the two in this study.
- These results are driven by residents in training as opposed to attending physicians; therefore, differences found in this study could change or fade over time as residents gained more experience.
AUTHOR/FUNDING INFORMATION:
Dr. Wagner is part of HSR&D’s Health Economics Resource Center (HERC) in Menlo Park, CA.
Doyle J, Ewer S and Wagner T. Returns to Physician Human Capital: Evidence from Patients Randomized to Physician Teams. Journal of Health Economics December 2010;29(6):866-82.
