The Impact of Peri-discharge Interventions on 30-Day Readmission and 90-Day Mortality Rates for Patients Admitted With Hepatic Encephalopathy

Abstract

Introduction: Hepatic encephalopathy (HE) is a significant cause of both hospital readmissions and mortality in patients with end stage liver disease. In this study, we aimed to determine predictors for hospital readmissions and mortality among patients admitted with HE, and to evaluate the impact of 2 peri-discharge interventions designed to reduce hospital readmissions for HE. Methods: We retrospectively reviewed a total of 312 HE patients who were admitted to our health system from January 2010 to September 2013. Multivariate analysis was performed to identify clinical variables that were predictors of 30-day readmissions and 90-day mortality rates. Two peri-discharge interventions (medication reconciliation [MedAction Plan] and post-discharge phone call) implemented in 2013 were also evaluated for their efficacy in reducing hospital readmissions and 90-day mortality. Results: Of the 312 patients in the study, there were 458 documented HE admissions with 28% of the cohort having more than 1 index admission. The majority of the patients were male (69%) and Caucasian (65%) with a mean age of 57. Chronic hepatitis C was the most common cause for cirrhosis. Mean MELD was 20 and 35% were already listed for transplant at the time of admission. Most patients were admitted with Grade 1-2 HE, and dual therapy with lactulose and rifaximin was used by 30% of patients prior to admission. 30-day readmission rate was 34% and 90-day mortality rate was 12%. After adjusting for MELD, predictors of 30-day readmission were hyponatremia on discharge and serum ALT, while predictors of 90-day mortality were 30-day readmission and diabetes. Being listed for transplant was associated with reduced 90-day mortality (OR 0.181; [0.055, 0.596]). A total of 18 patients received peri-discharge interventions and were compared to patients who received no intervention during the same time period. Compared with no intervention, the post-discharge phone call was associated with a non-statistically significant reduction in hospital readmissions. However, the MedAction plan was associated with a statistically significant increased risk for readmissions (RR 2.5333; CI [1.4136, 4.5400]; p=0.0018). Neither intervention was associated with reduced 90-day mortality. Conclusion: The findings of a significant readmission and mortality rate in this patient population correlates well with prior studies in the literature. The trend toward reduction of 30-day readmission among patients who received the post-discharge phone call is an important finding with potential medical and cost implications. Future studies should include an evaluation of factors that impact the effectiveness of peri-discharge interventions in this patient population.