BACKGROUND: The self-reported FRAIL scale is suitable for frailty assessment in the busy primary care practice setting. AIM: To estimate the association between phenotypic frailty assessed by FRAIL and subsequent health care costs in older men and women.
METHODS: Prospective study of 7947 community-dwelling adults (mean age 79.2 years, 52.6% female) enrolled in 4 cohort studies of older adults linked to U.S. Medicare claims. The primary predictor was the FRAIL scale categorized into 3 levels (robust, pre-frail, and frail). A multimorbidity index (Hierarchical Conditions Category score) and the Kim frailty indicator (approximating the deficit accumulation index) were derived from U.S. Medicare claims. Annualized total and sector-specific health care costs (U.S. 2023 dollars) for 36 months after the index examination were ascertained from claims. Generalized linear models with gamma variance and log link functions were used to estimate the association of FRAIL category with subsequent health care costs.
RESULTS: After accounting for claims-based indicators of multimorbidity and frailty, estimated annualized mean total health care costs for individuals categorized as robust were $9742 in women and $13,319 in men. Compared with robust individuals, adjusted mean annualized incremental costs for prefrailty were $4104 (95% C.I. 604–5604) in women and $1955 (95% C.I. 102–3808) in men and for frailty were $9028 (95% C.I. 6653–11402) in women and $6604 (95% C.I. 3168–11039) in men.
DISCUSSION AND CONCLUSION: Pre-frailty and frailty assessed using the self-reported FRAIL scale are associated with higher subsequent total health care costs, even after accounting for claims-based indicators of multimorbidity and frailty.
SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s40520-025-03272-5.