Although the Hospital Readmissions Reduction Program (HRRP) has decreased readmissions in targeted conditions, outcomes in high-risk subgroups are unknown. This study analyzed the impact of cirrhosis as a comorbidity on readmissions in conditions subjected to the HRRP.
Using a longitudinal analysis of the New York, Florida, and Washington State inpatient databases from 2009 to 2013, adult Medicare beneficiaries with a diagnosis-related group of targeted conditions by the HRRP—pneumonia, congestive heart failure (CHF), and myocardial infarction (MI)—were included. Exclusion criteria included inability to assess for readmission, previous liver transplant, or having a readmission not subject to penalty under the HRRP. A sensitivity analysis used the International Classification of Diseases, 9th Revision, Clinical Modification codes to identify pneumonia, CHF, and MI hospitalizations. The primary outcome was 30-day readmission, with secondary outcomes including 90-day readmission, trends, and cirrhosis-specific risk factors for readmission.
Of the 797,432 patients included, 8,964 (1.1%) had cirrhosis. Patients with cirrhosis had significantly higher 30-day readmissions overall (29.3% vs 23.8%, P < 0.001) and specifically for pneumonia and CHF, but not for MI. Thirty-day readmission rates significantly decreased in patients without cirrhosis (annual percent change −1.8%, P < 0.001), but not in patients with cirrhosis (P = 0.39). Similar findings were present for 90-day readmissions. A sensitivity analysis confirmed these findings. On multivariable analysis, cirrhosis was associated with significantly higher 30-day readmissions (odds ratio 1.13, P < 0.001).
When cirrhosis is comorbid in patients with conditions subjected to the HRRP, readmissions are higher and have not improved. Focused efforts are needed to improve outcomes in cirrhosis and other high-risk comorbidities within the HRRP cohort.
1Department of Gastroenterology and Hepatology, New York Presbyterian Hospital, Weill Cornell Medicine, New York,New York, USA;
2Division of Gastroenterology and Hepatology, University of Michigan Medicine, Ann Arbor, Michigan, USA;
3VA Center for Clinical Management Research, VA Ann Arbor Healthcare System, Ann Arbor, MI, USA;
4Department of Healthcare Policy and Research, Weill Cornell Medical College, New York, NY, USA.
Correspondence: Russell Rosenblatt, MD, MS. E-mail: RUR9017@NYP.ORG.
SUPPLEMENTARY MATERIAL accompanies this paper at http://links.lww.com/AJG/A175
Received January 10, 2019
Accepted March 18, 2019