Key points are not available for this paper at this time.
Frail, hospitalized patients are at a greater risk of further health issues. Frailty levels can worsen during hospitalization, with exercise being one of the best strategies to manage frailty. However, there is a lack of synthesized evidence to indicate whether exercise interventions are effective at managing frailty in this unique setting. This systematic review and meta-analysis determined if exercise interventions either preserve or improve frailty levels among inpatients. Sources were searched in May 2024 (updated August 2025) and included Scopus, EMBASE, MEDLINE, CINAHL, and Academic Search Premier. Studies were included if they conducted an exercise intervention in middle-to-older adult (≥45-years) inpatients and assessed frailty as an outcome using any validated frailty assessment tool. The between-group effect size of change in frailty (Intervention vs. Control) was calculated as the standardized mean difference (SMD) using a random effects model. Five studies including n=810 intervention (age: 82.3±3.5 range: 79-87.6 years; 45±1% reported females) and 900 control (age: 81.6±4.1 range: 76.0-87.1 years; 44±1% reported females) participants were included. Improvements in patients’ frailty were observed in n=3/5 studies. The meta-analysis exhibited a favorable effect on frailty between the intervention and control groups (SMD: 0.716, 95%CI: 0.119–1.312) with significant heterogeneity between studies ( I 2 =92%, p 0.196) . Despite limited available evidence, results indicate exercise training interventions may be effective in improving frailty. This supports the inclusion of exercise interventions beyond usual hospital care to improve frailty. • Exercise interventions are a viable and feasible strategy for managing frailty in hospitalized patients. • Aerobic or combined exercise programs are effective at improving frailty assessed via the frailty phenotype or frailty index. • High adherence & retention support intervention efficacy but improving recruitment and eligibility may enhance implementation.
Shivgulam et al. (Thu,) studied this question.