Back to Annual Symposium Program

Objective: Regionalization has been proposed as a potential strategy to improve the outcomes of patients undergoing abdominal aortic aneurysm (AAA) repair. We sought to examine the implications of regionalization of AAA repair to highest volume hospitals using Medicare data between 1997 and 2007.
Methods: All Medicare patients undergoing non-ruptured AAA repair between 1997 and 2007 were identified by ICD9 (n=294,812). Hospitals were stratified by volume of aortic aneurysm repair and total Medicare payments into quintiles. Differences in mortality and Medicare payments (including DRG payments, outlier and readmission payments) were compared using bivariate and multivariate statistics.
Results: Over the last decade, there has been relatively little redistribution of cases to highest volume hospitals. Previously documented trends confirmed that patients undergoing surgery in highest volume hospitals are 29% more likely to undergo an endovascular repair (RR 1.29, 95% CI 1.27-1.31). Accounting for differences in patient factors and endovascular repair, risk-adjusted mortality rates varied from 4.1% in the lowest volume hospitals to 3.1% in the highest volume hospitals (RR 1.27, 95% CI 1.08-1.49). When Medicare payments were examined across quintiles of hospital volume, the findings were surprising. Using 2005-2006 as a benchmark, median Medicare payments were slightly higher in the highest volume hospitals ($24, 078 in lowest volume vs. $25,177 in highest volume, p<0.001). These differences in payments across hospital volume quintiles were not due to differences in outlier payments (complications) or readmission rates, but were related to higher DRG payments to the highest volume hospitals. Thus, moving aortic surgery for all patients in 2006 to only highest volume hospitals would result in an excess of 29 million Medicare dollars spent and save 130 lives per year. Interestingly, lowest Medicare-payment facilities have a lower mortality than highest payment facilities (2.8% vs. 4.8%, RR 0.59, p<0.001). These differences in mortality were not explained by differences in endovascular repair rates, patient co-morbidities, or hospital volume, but occurred because low Medicare-payment facilities have lower complications (lower readmission and outlier payments).
Conclusions: It appears that regionalization of AAA repair to highest volume hospitals would reduce mortality rates; however, this may increase healthcare expenditures by as much as 29 million dollars per year. Continued focus on preventing complications, rather than regionalization, appears to be a better strategy for reducing overall Medicare expenditures.