Development and Internal Validation of the Immobility Harm Risk Score for Predicting Inpatient Immobility-Associated Outcomes: A Retrospective Cohort Study
Development and Internal Validation of the Immobility Harm Risk Score for Predicting Inpatient Immobility-Associated Outcomes: A Retrospective Cohort Study
ES评分 0
| DOI |
10.20900/agmr20260005 |
| 刊名 |
AGMR
|
| 年,卷(期) |
2026, 8(1) |
| 作者 |
|
| 作者单位 |
Department of Medicine, Indiana University Health System, Indianapolis, IN 46204, USA
|
| 摘要 |
Background: Hospitalized adults face multiple immobility-related adverse outcomes, yet existing risk tools address single outcomes with modest discrimination. A unified multi-outcome framework is needed. Methods: Retrospective cohort study at a large academic health system (2013–2023). Adults with complete mobility assessments (AM-PAC “6-Clicks” or Barthel Index; N = 9842) were included. Seven regularized logistic regression models predicted functional decline, venous thromboembolism (VTE), falls, pressure injuries, prolonged length of stay, post-acute discharge, and 30-day readmission using admission-time predictors. Models were developed (80%) and validated (20%) with cross-validation. Performance was compared with Braden, Morse, Padua, Charlson, and LACE scores. Results: Outcome incidences ranged from 1.1% (VTE) to 33% (prolonged stay). After recalibration, the Immobility Harm Risk Score (IHRS) demonstrated good discrimination (AUC 0.72–0.82) with excellent calibration (slopes 0.96–1.01). IHRS outperformed legacy instruments by 0.04–0.16 AUC units with positive net reclassification indices (+0.19 to +0.59). Decision curve analysis confirmed clinical benefit across relevant threshold ranges. Temporal validation showed stable discrimination (ΔAUC < 0.02). Conclusion: The IHRS provides a practical, multi-outcome risk stratification framework at admission. This single-center study requires external validation; performance is strongest in patients aged ≥ 65 years.
|
| Abstract |
Background: Hospitalized adults face multiple immobility-related adverse outcomes, yet existing risk tools address single outcomes with modest discrimination. A unified multi-outcome framework is needed. Methods: Retrospective cohort study at a large academic health system (2013–2023). Adults with complete mobility assessments (AM-PAC “6-Clicks” or Barthel Index; N = 9842) were included. Seven regularized logistic regression models predicted functional decline, venous thromboembolism (VTE), falls, pressure injuries, prolonged length of stay, post-acute discharge, and 30-day readmission using admission-time predictors. Models were developed (80%) and validated (20%) with cross-validation. Performance was compared with Braden, Morse, Padua, Charlson, and LACE scores. Results: Outcome incidences ranged from 1.1% (VTE) to 33% (prolonged stay). After recalibration, the Immobility Harm Risk Score (IHRS) demonstrated good discrimination (AUC 0.72–0.82) with excellent calibration (slopes 0.96–1.01). IHRS outperformed legacy instruments by 0.04–0.16 AUC units with positive net reclassification indices (+0.19 to +0.59). Decision curve analysis confirmed clinical benefit across relevant threshold ranges. Temporal validation showed stable discrimination (ΔAUC < 0.02). Conclusion: The IHRS provides a practical, multi-outcome risk stratification framework at admission. This single-center study requires external validation; performance is strongest in patients aged ≥ 65 years.
|
| 关键词 |
prediction model; immobility; hospital-acquired conditions; functional decline; electronic health records; deconditioning
|
| KeyWord |
prediction model; immobility; hospital-acquired conditions; functional decline; electronic health records; deconditioning
|
| 基金项目 |
|
| 页码 |
- |
Chris Gale, Eric Gan, Joshua Mijares, Manoj Bhagwat, Neil Jairath.
Development and Internal Validation of the Immobility Harm Risk Score for Predicting Inpatient Immobility-Associated Outcomes: A Retrospective Cohort Study [J].
Advances in Geriatric Medicine and Research.
2026; 8; (1).
- .