Acute Care for Elders Risk Score: A Practical Machine Learning-Based Tool for Screening High-Risk Older Inpatients

Area of Science:

• Geriatric Medicine
• Artificial Intelligence in Healthcare
• Clinical Risk Prediction

Background:

• Older hospitalized patients are at high risk for adverse events.
• Existing risk assessment tools may not fully capture this risk.
• Early identification of high-risk patients is crucial for timely intervention.

Purpose of the Study:

• To develop and validate the Acute Care for Elders (ACE) Risk Score.
• Integrate Clinical Frailty Scale (CFS) with clinical and laboratory data.
• Predict adverse outcomes in older inpatients.

Main Methods:

• Retrospective cohort study using a machine learning framework (AutoScore).
• Included 21,757 hospital admissions of patients aged ≥65 years.
• Developed a parsimonious model with 5 variables: CFS, albumin, C-reactive protein, hemoglobin, and pre-admission medications.

Main Results:

• The ACE Risk Score demonstrated superior predictive performance (AUC 0.837) compared to CFS alone (AUC 0.798) and age (AUC 0.630).
• Higher ACE scores correlated with increased risks of safety events, readmissions, longer hospital stays, and rapid response team activations.
• The score effectively predicted a composite outcome of in-hospital delirium, pressure ulcers, falls, and mortality.

Conclusions:

• The ACE Risk Score is a practical, interpretable, and scalable tool for early identification of high-risk older inpatients.
• Utilizes data available on the first day of admission to support timely, targeted geriatric interventions.
• Facilitates broader implementation of risk-guided care strategies for older adults in hospital settings.