Hospital Teaching Status and Medicare Expenditures for Complex Surgery

Abstract

Objective: To evaluate the relationship between hospital teaching intensity, Medicare payments, and perioperative outcomes.

Background: Several emerging payment policies penalize hospitals for low-value healthcare. Teaching hospitals may be at a disadvantage given the perception that they deliver care less efficiently.

Methods: Using Medicare Provider and Analysis Review files, we studied patients from age 65 to 100 years who underwent abdominal aortic aneurysm (AAA) repair (n = 71,422), pulmonary resection (n = 93,056), or colectomy (n = 277,619) from 2009 to 2012. Patients' hospitals were categorized into quintiles of teaching intensity (very major, major, minor, very minor, and nonteaching hospitals) based on the resident-to-bed ratio. Risk-adjusted 30-day Medicare payments were price-standardized to account for graduate medical education payments, disproportionate share costs, and regional wage-index adjustments. Risk-adjusted perioperative outcomes were also assessed.

Results: Comparing risk-adjusted Medicare payments per episode of surgery, very major teaching hospitals were $14,145 more expensive than nonteaching hospitals for AAA repair ($45,570 vs $31,426; P < 0.001), $9,812 more expensive for pulmonary resection ($39,550 vs $29,738; P < 0.001), and $19,147 more expensive for colectomy ($51,893 vs $32,746; P < 0.001). However, after accounting for social subsidies and regional variation in Medicare spending, risk-adjusted Medicare payments were not statistically different between very major teaching hospitals and nonteaching hospitals for AAA repair ($29,946 vs $27,993; P = 0.18) and pulmonary resection ($25,407 vs $26,813; P = 1.00); a statistically significant but attenuated difference persisted for colectomy ($34,949 vs $30,352; P < 0.001). Very major teaching hospitals generally had higher risk-adjusted rates of serious complications and readmissions, but lower risk-adjusted rates of failure to rescue and 30-day mortality than did nonteaching hospitals.

Conclusions: After price-standardization to account for intended differences in Medicare spending, risk-adjusted Medicare payments for an episode of surgical care were similar at teaching hospitals and nonteaching hospitals for three complex inpatient operations.