Author + information
Discharge disposition following transcatheter aortic valve replacement (TAVR) is highly varied. Little is known about the impact of different post-acute care settings on hospital readmissions. The aim of this study was to determine if readmission rates vary by discharge destination after TAVR.
Using the 100% Medicare Inpatient Standard Analytic Files, we retrospectively analyzed 11,698 Medicare patients who underwent TAVR between 2011-2013 and were discharged from the hospital alive. Multivariate regression models were used to control for differences in patient demographics, comorbidities and complications. Sub-analyses were performed to assess the impact of discharges to home or home health (HH). Experienced hospitals (at least 30 TAVR discharges to home or HH), were stratified according to high or low HH utilization. High HH users were defined as those that discharge 75% or more of the combined home or HH population to HH. Low HH users were defined as those that discharge less than 25% of the combined population to HH.
Among 11,698 patients discharged alive after TAVR, discharge destination was home in 33%, home with HH in 34%, skilled nursing facility (SNF) in 25% and rehabilitation or long term care hospital (LTCH) in 7%. Unadjusted 30-day readmission rates were highest in patients sent to rehabilitation or LTCH (25%), followed by SNF (24%), HH (18%) and home (15%). After regression adjustment, readmission rates remained highest in patients sent to rehabilitation or LTCH (19%), followed by SNF (19%), HH (15%) and home (13%). Among 77 experienced hospitals (7,852 patients), there was wide latitude in the choice between home and HH, with the percentage discharged to HH ranging from 8% to 92%. At 30-days, the average readmission rate for the combined home and HH population was 16.8% and 16.4% (p=0.79) in high HH and low HH users, respectively.
In this study, the readmission rate following TAVR was associated with choice of discharge destination. Patients sent to inpatient post-acute care (rehabilitation or LTCH, SNF) had the highest rates of readmission. Patients discharged home (with or without HH) had the lowest readmission rates. Hospitals show substantial variation in the percentage of TAVR cases warranting HH, with virtually identical rates of 30-day readmission for high and low HH users. These results suggest the choice of discharge to home or HH may be discretionary. In this analysis, HH does not appear to reduce readmissions compared to discharge home.