Ahmed A. Elkafrawy, MD1, Osama Kaddourah, MD1, Kevin F. Kennedy, MS2, Alisa Likhitsup, MD3
1University of Missouri Kansas City School of Medicine, Kansas City, MO; 2Mid America Heart Institute, Kansas City, MO; 3Saint Luke's Liver and Transplant Specialists, University of Missouri Kansas City, Kansas City, MO
Introduction: Hepatic encephalopathy (HE) is a common diagnosis of hospital admissions. Our aim in this study is to identify trends, causes and predictors of 30-day readmissions in patients admitted with HE.
Methods: The Nationwide Readmissions Database (NRD) was queried to identify index admissions with HE between 2010-2015 using the ICD-9 code 572.2. All the patients non-electively admitted within 30 days of Index admission were included. We excluded patients who died during their index admission or left against medical advice. Chi-Square, Fisher’s Exact and Student-t tests were used for statistical analysis to compare baseline characteristics among patients with and without 30-day readmission. Multivariate logistic regression models were used to predict factors associated with readmissions. All analyses were performed with SAS 9.4 (Cary, NC).
Results: We identified 196,730 eligible index admissions. The 30-day readmission rate was 29.4% and stable over time (Figure 1). Median (IQR) age was 60 (52, 68) years, 42% were female and the median length of stay was 5 (3, 10) days. Among patients with 30-day readmission, there is a higher proportion of patients with Medicaid (23.5 vs 20%, p< 0.001), residing in the metro area (32.3 vs 29.5%, p< 0.001), being discharged from metro-teaching hospital (p< 0.001). The hospital bed size and ownership setting were similar between the groups. The proportion of patient readmitted to the hospital was lower among patients who were discharged to a skilled facility (24 vs 27%, p< 0.001) but higher among patients who were discharged to home with self-care (53 vs 50%, p< 0.001). The main reason for 30-day readmission was liver disease-related complications (37.6%), followed by sepsis, mental health or substance use disorder (Figure 2). Multivariate analysis showed Metro-teaching hospital, presence of diabetes, renal failure, presence of ascites, required paracentesis, presence of portal hypertension were predictive factors associated with 30-day readmissions (Table 1).
Discussion: Approximately a third of patients admitted with HE were readmitted to the hospital, and the trend has been stable over a 6-year period. From our study, health-care provider education in the metro-teaching hospitals regarding cirrhosis, renal failure, and ascites management will likely improve 30-day readmission rate among patients with HE.
Citation: Ahmed A. Elkafrawy, MD; Osama Kaddourah, MD; Kevin F. Kennedy, MS; Alisa Likhitsup, MD. P2428 - THIRTY-DAY READMISSIONS IN PATIENTS ADMITTED WITH HEPATIC ENCEPHALOPATHY BETWEEN 2010-2015: A STUDY FROM THE NATIONWIDE READMISSION DATABASE. Program No. P2428. ACG 2019 Annual Scientific Meeting Abstracts. San Antonio, Texas: American College of Gastroenterology.