Related Experiment Videos
The Cirrhosis Administrative Frailty Index (CAFI) Predicts Mortality Among Hospitalized Patients With Cirrhosis
Lucia Calthorpe1, Catherine Lee2, Cynthia Fenton3
1Department of Surgery, UCSF, San Francisco, CA, USA.
Insights
The Cirrhosis Administrative Frailty Index (CAFI) effectively predicts mortality in hospitalized cirrhosis patients. Higher CAFI scores indicate increased risk, aiding in risk stratification for better patient care.
Area of Science:
- Hepatology and Gastroenterology
- Clinical Epidemiology
- Health Services Research
Background:
- The Cirrhosis Administrative Frailty Index (CAFI) is the first ICD-based frailty index for cirrhosis patients.
- CAFI has been validated against in-person measures of physical frailty.
Purpose of the Study:
- To evaluate the CAFI's ability to predict mortality in a national cohort of hospitalized cirrhosis patients.
- To assess the association between CAFI scores and in-hospital mortality.
Main Methods:
- Utilized data from the National Inpatient Sample (NIS), 2019-2022, for adult cirrhosis hospitalizations.
- Computed CAFI using ICD-10 codes and employed multivariable logistic regression to analyze mortality risk.
- Determined CAFI cut points for risk stratification and performed stratified analyses by demographic and clinical factors.
Main Results:
- Among 642,487 admissions, 6.4% experienced in-hospital mortality.
- Each point increase in CAFI was associated with a 77% increased odds of mortality (OR=1.77).
- CAFI cut points stratified patients into low (3% mortality), intermediate (7% mortality), and high-risk (12% mortality) groups.
Conclusions:
- CAFI demonstrated strong external validity for predicting mortality in hospitalized cirrhosis patients.
- CAFI can be broadly applied to administrative datasets to account for frailty in cirrhosis research.
- This enhances the rigor of population-based studies involving cirrhosis patients.
Introduction:
The Cirrhosis Administrative Frailty Index (CAFI) is the first ICD-based frailty index specific to cirrhosis patients and validated against an in-person measure of physical frailty. We aimed to evaluate its ability to predict mortality in a nationally representative cohort of hospitalized patients with cirrhosis.
Methods:
Hospitalizations of adult patients with cirrhosis were identified in the National Inpatient Sample (NIS), 2019-2022. The CAFI was computed for each hospitalization using ICD-10 codes associated with the admission. Multivariable logistic regression was used to determine the association between CAFI and in-hospital mortality. Cut points for CAFI were derived by optimizing model fit (using Akaike Information Criterion) in multivariable logistic regression models predicting in-hospital mortality. Stratified analyses by age, sex, and Baveno stage were performed to determine whether these factors moderate the association between CAFI and mortality.
Results:
Among 642,487 admissions, 6.4% of patients died in-hospital. CAFI ranged from 4.3 to 14.7. Adjusting for age, sex, race/ethnicity, cirrhosis etiology, Baveno stage, and hepatorenal syndrome, each point increase in CAFI was associated with 77% increased odds of mortality (OR=1.77, 95%CI: 1.75, 1.79). CAFI cut points of 8.50 and 9.01 stratified patients into low (50%), intermediate (20%), and high-risk (30%) groups, with in-hospital mortality rates of 3%, 7%, and 12%, respectively.
Conclusions:
The CAFI, developed in an ambulatory cohort, demonstrated strong external validity for mortality in hospitalized cirrhosis patients. Together, these findings suggest that CAFI could be broadly applied to account for frailty in analyses of administrative datasets, thereby enhancing the rigor of population-based research in cirrhosis patients.
Related Concept Videos
01:23Cirrhosis I: Introduction
01:24Cirrhosis II: Pathophysiology
01:25Ultrasound II: Endoscopic Ultrasound and FibroScan
Endoscopic Ultrasound (EUS):