W. T. Mehtsun1,2, I. Papanicolas2,4, K. D. Lillemoe1,3, A. K. Jha2,3 1Massachusetts General Hospital,Department Of Surgery,Boston, MA, USA 2Harvard School Of Public Health,Department Of Health Policy And Management,Boston, MA, USA 3Harvard School Of Medicine,Brookline, MA, USA 4The London School Of Economics And Political Science,LSE Health,London, , United Kingdom
Introduction: Hospital readmissions have recently become a focal point of national health policy efforts. The past decade has seen the implementation of public reporting and financial penalties targeted at improving readmission rates in U.S. hospitals. Consequently, the reduction of excess readmission following inpatient surgery has now become a national priority for clinicians, hospitals, and health policy leaders. Whether this increased attention and effort has led to an observable decrease in surgical readmission rates is unclear. Therefore, we sought to determine whether national 30-day readmission rates after common and complex surgeries are declining.
Methods: Using 100% inpatient Medicare-fee-for-service beneficiary data from 2005 to 2014, we identified patients undergoing the following complex and common surgical procedures: coronary artery by-pass graft (CABG), abdominal aortic aneurysm repair (AAA repair), esophagectomy, pancreatectomy, cystectomy, pulmonary lobectomy, colectomy, appendectomy, cholecystectomy, hip replacement, and knee replacement. For each procedure, we used linear regression models with hospital fixed effects to calculate yearly 30-day surgical readmission rates – risk adjusted by age, gender, urgency, and Elixhauser comorbidity indicators.
Results: Among Medicare fee-for-service patients, who underwent the selected procedures during the study period (N=7,411,230), 30-day surgical readmission rates declined for 8 out of 11 procedures. The greatest declines in 30-day surgical readmission rates were in patients who underwent CABG (-0.52% per year, p<0.001), hip replacement (-0.41% per year, p<0.001), and pulmonary lobectomy (-0.36% per year, P<0.01). There were non-significant declines in 30-day surgical readmission rates among patients who underwent esophagectomy (-0.15% per year, p=0.80), pancreatectomy (-0.02% per year, p=0.70), and cystectomy (-0.04% per year, p=0.70). We found that improvements in 30-day surgical readmission rates occurred mainly within hospitals.
Conclusions: Declines in 30-day surgical readmission rates were observed for most procedures, with the greatest readmission reductions seen among procedures that were targeted by pay-for-performance programs.