Functional vulnerability and all-cause mortality among community-dwelling United States adults with a history of hepatitis or chronic liver disease.

Background and aims: Frailty predicts death in advanced cirrhosis and among transplant candidates, but whether a frailty-like functional index identifies mortality risk in the far larger population of community-dwelling adults with liver disease, who are rarely assessed for it, is unsettled. We examined whether a survey-based functional vulnerability index (FVI) was associated with all-cause mortality among US adults reporting a history of hepatitis or a chronic liver condition, independent of poverty and insurance status. Methods: Retrospective cohort study using the National Health Interview Survey (2008-2018) linked to the National Death Index through December 31, 2019. Adults 18 years or older reporting a history of hepatitis or a chronic liver condition, eligible for mortality linkage, and without baseline liver cancer were included. The FVI was defined as the proportion of 9 self-reported deficits present (fair/poor self-rated health, ADL or IADL difficulty, work limitation; and difficulty walking, climbing, standing, carrying, or grasping), modeled per 0.10-unit higher score and by deficit count (0, 1, 2, or ≥ 3). Survey-weighted Cox models estimated hazard ratios (HRs) for all-cause mortality, and survey-weighted modified Poisson models estimated 5-year mortality risk ratios. Results: Among 9291 adults (1151 deaths; 59,732 person-years; mean follow-up of 6.4 years), 64.6% had no functional deficit. In the fully adjusted model, each 0.10-unit higher FVI was associated with higher mortality (HR: 1.19, 95% confidence interval [CI]: 1.15-1.22), corresponding to an HR of 1.56 (95% CI: 1.44-1.68) per 1 - SD higher FVI. Versus no deficit, mortality increased with increasing burden: 1 deficit (HR: 1.37, 95% CI: 1.05-1.80); 2 deficits (HR: 2.49, 95% CI: 1.86-3.33); and ≥3 deficits (HR: 3.29, 95% CI: 2.65-4.08). Weighted 5-year mortality increased from 3.2% with no deficit to 23.0% with ≥3 deficits. The association was nonproportional and strongest during the first 2 years (HR: 1.28, 95% CI: 1.22-1.35), but persisted beyond 5 years (HR: 1.12, 95% CI: 1.07-1.18). Multiple imputation yielded similar results (HR: 1.20, 95% CI: 1.17-1.23). Poverty predicted mortality before but not after FVI adjustment; uninsurance did not. Conclusion: Among community-dwelling US adults with hepatitis or chronic liver disease, a survey-based functional index was strongly and gradually associated with all-cause mortality, capturing risk not explained by poverty or insurance status. Brief functional assessment may identify high-risk patients well outside the transplant clinic.