Frailty is increasingly recognized in older adults with lower respiratory tract infections (LRTIs), yet the optimal assessment tool and its prognostic value remain unclear. This study evaluates the prevalence, inter-tool agreement, and predictive performance of four frailty measures in this population.

Older adults hospitalized with LRTIs were prospectively enrolled. Frailty was assessed using the Clinical Frailty Scale (CFS), Fried Frailty Phenotype (FFP), FRAIL Scale (FS), and Short Physical Performance Battery (SPPB). The primary outcome was 1-year mortality; secondary outcomes included in-hospital mortality and decline in activities of daily living (ADL) at discharge and 1 year. Inter-tool agreement was assessed using Cohen’s kappa, and predictive performance was evaluated by the area under the receiver operating characteristic curve (AUROC).

Of 361 patients (median age 74 years; 61.5% male), frailty prevalence ranged from 18.0% (CFS) to 57.3% (SPPB), with fair to moderate inter-tool agreement. In-hospital and 1-year mortality were 1.4% and 11.5%, respectively. ADL decline occurred in 18.3% at discharge and 21.6% at 1 year. All tools were associated with 1-year mortality, but only the FFP and SPPB were associated with ADL decline.

The FFP demonstrated the the most consistent discriminative performance, with relatively high AUC and a good trade-off between sensitivity and specificity. The CFS exhibited high specificity (0.823–0.881), whereas the SPPB demonstrated high sensitivity (0.724–1.000) across different outcomes. None of the tools adequately predicted 1-year ADL decline (AUC < 0.600).

Frailty is prevalent and prognostic in older adults with LRTIs. The FFP provided consistent prediction of mortality and short-term ADL decline, the CFS was effective for confirming high-risk patients, and the SPPB was useful for early risk screening. The limited prediction of long-term functional decline highlights the need for LRTIs-specific frailty assessment instruments.