Related Experiment Video
Updated: May 11, 2026
Assessment of Dependence in Activities of Daily Living Among Older Patients in an Acute Care Unit
Published on: September 30, 2020
A simple admission-based risk score for hospital-associated disability in older adults
Hiroyuki Umegaki1,2, Hirotaka Nakashima3, Yosuke Yamada3
1Department of Community Healthcare and Geriatrics, Graduate School of Medicine, Nagoya University, 65 Tsuruma-Cho, Showa-Ku, Nagoya, Japan. umegaki@med.nagoya-u.ac.jp.
Aim:
To develop a simple admission risk stratification rule to identify older inpatients at risk of hospital-associated disability (HAD).
Findings:
A four-item rule (age ≥ 80, CFS ≥ 5, MMSE ≤ 23, emergency admission) stratified HAD risk effectively (AUC 0.80). Low scores identified patients with very low risk, whereas high scores identified those at markedly increased risk.
Message:
This bedside rule supports targeted use of comprehensive geriatric assessment and tailored management while avoiding unnecessary intervention in low-risk patients. Hospital-associated disability (HAD) is a frequent complication in hospitalized older adults and is associated with unfavorable outcomes. Because preventive interventions require substantial resources, efficient identification of high-risk patients at admission is required. We developed a simple admission-based clinical decision rule for early risk stratification using routinely available admission information.
Methods:
In a multicenter retrospective cohort study, adults aged ≥ 65 years admitted to acute care hospitals between October 2019 and March 2025 were included. HAD was defined as a ≥ 5-point decline in the Barthel Index from admission to discharge. Candidate predictors included age, admission type, Clinical Frailty Scale (CFS), and Mini-Mental State Examination (MMSE). A multivariable logistic regression model derived a simplified point-based score; discrimination was assessed by AUC, calibration by bootstrap-corrected calibration, and clinical utility by decision curve analysis with 2,000 bootstrap resamples.
Results:
Among 1,292 patients (mean age 82.9 ± 6.7 years, 58.2% women), 26.2% developed HAD. The final score (0-8 points) included emergency admission, frailty (CFS ≥ 5), cognitive impairment (MMSE ≤ 23), and age ≥ 80 years. Discrimination was good (AUC 0.796; 95% CI 0.771-0.820) with adequate calibration. HAD incidence increased from < 5% in the lowest-risk group to ~ 50% in the highest-risk group, and decision curve analysis showed net clinical benefit across relevant thresholds.
Conclusions:
This bedside rule supports targeted use of comprehensive geriatric assessment and tailored management while avoiding unnecessary intervention in low-risk patients.
