P0912 - Investigating Primary Factors Contributing to Cirrhosis Readmissions at a Safety-Net Academic Medical Center in New Mexico

Sunday, October 22, 2023

3:30 PM – 7:00 PM PT

Location: Exhibit Hall

Presenting Author(s)

Justin M. Joseph, MD (he/him/his)

University of New Mexico
Albuquerque, NM

Justin M. Joseph, MD¹, Aleksandr Birg, MD²
¹University of New Mexico, Albuquerque, NM; ²University of New Mexico Hospital, Albuquerque, NM

Introduction: Cirrhosis is an increasingly common condition that can emerge as the terminal stage of many liver diseases. New Mexico has the highest liver disease mortality at 41.1 deaths per 100 000 residents (CDC, 2022). Decompensated cirrhosis has a high symptom burden from its multiple complications leading to high healthcare utilization. This results in repeat readmissions, which can strain healthcare systems at capacity. Readmission rates at 90 days are an important outcome measure in quality improvement efforts. Studying patient-specific and hospitalization-specific characteristics associated with 90-day readmission can help decrease overall hospital burden and improve care.

Methods: A retrospective, case-control study was performed to identify patients admitted with cirrhosis-related complications between 7/1/2019 and 1/1/2020 at a safety-net academic medical center. Data was collected regarding demographics, in-hospital events, and discharge/readmission. Patients with readmission within 90 days were identified. Patients without readmission were compiled into a control group. Multivariate logistic regression on Prism 9 (GraphPad) was used to identify statistically significant factors for readmission (P< 0.05).

Results: 442 patients were identified for this study. 95 (21.5%) were readmitted within 90 days. 347 (78.5%) were used as a control group. Factors significantly associated with readmission include nosocomial infection (aOR 5.97, 95% CI 1.35-28.47), proton pump inhibitor (PPI) prescribed on discharge (aOR 2.06, 95% CI 1.14-3.75), discharge against medical advice (aOR 3.12, 95% CI 1.06-9.00), and discharge to long-term acute care (aOR 2.77, 95% CI 1.04-7.19). Nosocomial infections in readmitted patients had an intra-pulmonary (50%) or intra-abdominal source (50%).

Discussion: This study shows that patients admitted with a cirrhosis-related diagnosis were more likely to be readmitted if they had a PPI prescription on discharge, left against medical advice, had a nosocomial infection, or discharged to long-term acute care. Long-term PPI use without an evidence-based indication is linked to SBP, pneumonia, hepatic encephalopathy, and mortality. PPIs may alter the intestinal microbiome, resulting in bacterial translocation. PPI stewardship is a potential quality improvement intervention that may reduce hospital readmission.

References:
CDC. Chronic Liver Disease/Cirrhosis Mortality by State.
https://www.cdc.gov/nchs/pressroom/sosmap/liver_disease_mortality/liver_disease.htm. 2 Mar 2022.

Characteristic	Percentage or Median of Sample (IQR)	Unadjusted Odds Ratio (95% CI)	Adjusted Odds Ratio (95% CI)
Age	56 (19)	0.95	1.00 (0.98-1.025)
Male	61.54	1.13 (0.79-1.62)	1.31 (0.75-2.32)
Private Insurance	8.60	0.68 (0.28-1.69)	0.62 (0.20-1.68)
No Insurance	82.35	0.70 (0.26-1.88)	0.80 (0.25-2.17)
VA Insurance	2.04	1.04 (0.21-5.11)	1.40 (0.18-7.35)
Current Alcohol Use	42.53	1.08 (0.71-1.63)	1.03 (0.55-1.94)
Former Alcohol Use	14.25	1.14 (0.61-2.12)	1.05 (0.45-2.35)
Diabetes	31.67	0.95 (0.59-1.52)	0.73 (0.39-1.31)
Major Depressive Disorder	22.85	1.23 (0.77-2.09)	1.51 (0.80-2.82)
Direct Admission from Clinic/Physician's Office	6.79	0.91 (0.36-2.30)	0.85 (0.25-2.45)
Transfer from Another Hospital	14.71	0.59 (0.28-1.23)	0.70 (0.28-1.57)
Transfer from Skilled Nursing or Intermediate Care Facility	2.94	0.30 (0.04-2.37)	0.35 (0.02-2.06)
Admitted from Emergency Department	48.64	1.36 (0.77-1.67)	1.28 (0.76-2.15)
Hepatic Encephalopathy	11.09	1.06 (0.52-2.15)	1.78 (0.74-4.62)
Ascites	31.22	1.19 (0.76-1.87)	1.47 (0.77-2.78)
Spontaneous Bacterial Peritonitis	4.75	1.83 (0.72-4.65)	1.75 (0.74-6.16)
Bleeding Esophageal Varices	12.44	1.08 (0.72-1.63)	0.51 (0.20-1.24)
Portal Hypertension	42.53	1.08 (0.72-1.63)	0.90 (0.03-7.74)
Hepatocellular Carcinoma	11.09	4.87 (1.07-22.14)	3.43 (0.57-22.68)
HBV	1.13	0.91 (0.10-8.27)	0.83 (0.03-7.74)
HCV	36.24	1.39 (0.87-2.22)	1.51 (0.83-2.71)
EGD Performed	28.73	1.08 (0.67-1.73)	0.98 (0.21-4.34)
Diagnostic Paracentesis	12.67	1.34 (0.71-2.52)	0.90 (038-2.01)
Shock During Initial Hospitalization	16.06	0.82 (0.43-1.56)	0.90 (0.38-2.01)
Dialysis During Initial Admission	4.07	1.04 (0.34-3.24)	1.18 (0.28-4.19)
Seen By GI Consult Service	28.05	1.11 (0.69-1.79)	1.17 (0.27-5.17)
Nosocomial Infection	2.26	3.65 (1.04-12.88)	5.97 (1.35-28.5)
LOS of index hospitalization	5 (7)	1	1.00 (0.95-1.04)
ICU Days of Index Hospitalization	0 (2)	0.85	0.97 (0.89-1.05)
Lactulose and/or Rifaximin at Initial Discharge	40.27	1.31 (0.88-1.96)	0.57 (0.33-1.01)
PPI on Initial Discharge	40.50	1.26 (0.87-1.89)	2.06 (1.14-3.75)
SBP Prophylaxis on Initial Discharge	10.41	1.29 (0.64-2.59)	1.30 (0.54-2.97)
Discharge to Skilled Nursing Facility	27.15	0.87 (0.52-1.44)	0.80 (0.38-1.67)
Discharge to Long-Term Acute Care Facility	7.69	1.75 (0.82-3.71)	2.77 (1.04-7.19)
Discharge to Home with Home Health Care	4.75	1.14 (0.41-3.20)	1.71 (0.48-5.33)
AMA Discharge	4.75	2.74 (1.12-6.69)	3.12 (1.04-7.19)
Weekday Discharge	16.29	0.98 (0.70-1.37)	0.75 (0.39-1.50)
Not Hispanic or Latino	56.33	1.06 (0.73-1.54)	1.39 (0.82-3.28)
MELD-NA on Discharge	11 (8)	1.1	0.99 (0.95-1.03)

Table: Baseline Characteristics and Select Factors Associated with Readmission for Cirrhosis-Related Diagnoses

Disclosures:

Justin Joseph indicated no relevant financial relationships.

Aleksandr Birg indicated no relevant financial relationships.

Justin M. Joseph, MD¹, Aleksandr Birg, MD². P0912 - Investigating Primary Factors Contributing to Cirrhosis Readmissions at a Safety-Net Academic Medical Center in New Mexico, ACG 2023 Annual Scientific Meeting Abstracts. Vancouver, BC, Canada: American College of Gastroenterology.