Abstract
BACKGROUND: Frailty is a geriatric syndrome of decreased physiologic reserve and a risk factor for hospitalization and mortality. We hypothesized that an adapted survey-based frailty-related phenotype (aFRP) predicts hospitalization and mortality among HIV-infected and uninfected individuals in adjusted models but is uncommon among those achieving undetectable HIV-1 RNA. METHODS: Defined from self-reported domains of physical shrinking, exhaustion, slowness, and low physical activity in Veterans Aging Cohort Study (VACS) participants, aFRP was considered present with ≥3 domains and prefrailty with 1-2 domains. Cox survival analysis determined hazard ratios (HRs) for 5-year hospitalization and mortality risk adjusting for frailty states, demographics, health behaviors, comorbidities, and a validated risk index incorporating HIV-specific and general organ system biomarkers, the VACS Index. Model discrimination was assessed. RESULTS: Participants with complete data were included [6515/7324 (89%)]. Of these, 3.9% of HIV-infected individuals with HIV-1 RNA >400 copies per milliliter; 2.0% of HIV-infected individuals with HIV-1 RNA ≤400 copies per milliliter; and 2.8% of uninfected individuals met aFRP criteria (P = 0.01). After adjustment for other covariates, aFRP was associated with hospitalization (HR = 1.78; 95% confidence interval: 1.48 to 2.13) and mortality (HR = 1.75; 95% confidence interval: 1.28 to 2.40). C-statistics for the VACS Index for hospitalization (0.633) and for mortality (0.756) were higher than for aFRP (0.565 and 0.584, respectively). C-statistic for hospitalization improved modestly when VACS Index and aFRP were both included (0.646) and minimally for mortality (0.761). CONCLUSIONS: aFRP was independently associated with adverse health outcomes among HIV-infected and uninfected individuals. aFRP modestly improved prediction for hospitalization. However, the aFRP is rare among HIV-infected individuals with undetectable HIV-1 RNA.