Document resource
Objective To examine whether Functional Independence Measure (FIM) scores assessed during the early phase of hospitalisation, before the initiation of rehabilitation, independently predict home discharge among older adults admitted to an acute general medical ward.Design Retrospective observational cohort study.Setting The General Medicine Department of a municipal general hospital in Japan.Participants Of 1168 community-dwelling adults aged ≥65 years admitted from home, the study included 744 who received rehabilitation and underwent early-phase FIM assessment before rehabilitation.Main outcome measures The primary outcome was the discharge destination (home vs facility). The predictive value of the total FIM score (FIM-T), motor score (FIM-M) and cognitive score (FIM-C) was evaluated using multivariable logistic regression and receiver operating characteristic (ROC) curve analyses.Results Overall, 494 patients (66.4%) were discharged home, 175 were discharged to facilities, and 75 died in hospital. The admission FIM was measured during the early phase of hospitalisation, with a mean assessment time of 4 days after admission. Early-phase FIM scores were significantly higher in the home discharge group than the facility discharge group (FIM-T: 65.8 vs 44.8; FIM-M: 38.1 vs 23.1; FIM-C: 27.7 vs 21.8; all p<0.001). After adjustment for age, clinical variables and social factors, higher FIM-T, FIM-M and FIM-C scores remained independently associated with home discharge (adjusted OR per 1-point increase for FIM-T: 1.06). ROC analysis showed moderate discriminative ability (areas under the curve of 0.75 for FIM-T and FIM-M, and 0.71 for FIM-C).Conclusions Early-phase FIM scores, including both motor and cognitive domains, independently predict home discharge among older adults in acute general medical wards and may support individualised discharge planning in routine practice. These findings may be most generalisable to acute care hospitals where early-phase functional assessments are performed at the initial rehabilitation evaluation, which, in our setting, occurred a mean of 4 days after admission.