External Validation of The Hospital Frailty Risk Score (HFRS) to Identify Liver Cirrhosis Patients at Risk for Frailty Mortality: A Multi Centre Retrospective Cohort Study in Asians

Background/Objectives: Frailty is a well-established predictor of adverse outcomes in older adults and patients with chronic disease. The Hospital Frailty Risk Score (HFRS), derived from administrative ICD-10 data, was previously developed and validated in Western populations to identify frailty risk. This study aimed to validate the HFRS in an Asian cirrhotic population and evaluate its ability to predict mortality and hospital readmissions. Methods: Patients with liver cirrhosis and at least one hospital admission in 2018 from two tertiary public hospitals in Singapore were identified using ICD-10 codes. HFRS was calculated using a two-year lookback period according to Gilbert’s methodology. Patients were categorized as no risk (score = 0), low risk (score < 5), intermediate risk (5–15), or high risk (>15). Outcomes assessed included 30-day mortality, 30-day hospital readmission, and 2-year mortality, with follow-up until 31 December 2020. Multivariate logistic regression analysis was performed. Results: A total of 814 patients were included, of whom 629 (77.3%) had frailty-related ICD-10 codes. Increasing HFRS category was independently associated with worse outcomes. Compared with the reference group, 30-day mortality increased stepwise with frailty: low risk odds ratio (OR) 25.74 (95% CI 1.54–431.33, p = 0.02), intermediate risk OR 40.42 (95% CI 2.42–676.14, p = 0.01), and high risk OR 50.92 (95% CI 2.92–887.08, p < 0.01). Higher HFRS was also associated with increased 30-day readmission and 2-year mortality across all risk categories. Conclusions: Frailty-related diagnoses are significant independent predictors of mortality and hospital readmission in patients with liver cirrhosis. HFRS is a cost-effective and scalable tool using administrative data to identify high-risk patients and support clinical risk stratification.