JMIR Publications

ABSTRACT

Background:

Frailty is prevalent among hospitalized patients and is associated with increased risk of complications, prolonged hospitalization, higher mortality and health care costs. Existing frailty screening tools, including the Clinical Frailty Scale (CFS) and Hospital Frailty Risk Score (HFRS), have limitations related to subjectivity, resource intensity or reliance on retrospective data, limiting use in routine acute care. Early identification of frailty is critical to support targeted intervention and resource allocation, yet scalable approaches for predicting frailty early during hospital admission remain limited. This paper evaluates whether routinely collected Electronic medical Record (EMR) data available within the first 72 hours of admission can be used to predict frailty in hospitalized patients.

Objective:

: Key objectives were to build a robust frailty prediction model based on the data available 72 hours after hospital admission, identify important predictors of frailty and evaluate the model’s performance at 24 hours after admission and discharge.

Methods:

A retrospective cohort study was conducted using routinely collected EMR data from the Royal Adelaide Hospital, including overnight admissions for patients aged 18 and above with a length of stay of at least 72 hours between July 2023 and December 2024. Frailty was defined using the HFRS (score>4) based on diagnosis codes available after the visit. Predictors were derived from EMR data available within 72 hours of admission, including demographics, medications, laboratory results, nursing assessments, and medical and allied health consult orders. Data were split into training (80%) and test (20%) sets. Multiple machine learning models were trained, with F2 score as the primary optimization metric. Model performance was evaluated using F2 score, recall, Area Under the Curve (AUC), and calibration.

Results:

The study included 26,824 admissions, of whom 9,427 (35%) were classified as frail. The final XGBoost model achieved an F2 score of 0.751, an AUC of 0.776, and recall of 0.871 on the test dataset, with good calibration. Key predictors included polypharmacy, laboratory results (including creatinine, albumin, sodium), allied health consult orders, and nursing assessment data. Model performance was lower at 24 hours post-admission and improved by discharge. Performance was reduced for Indigenous patients compared to non-Indigenous patients, while performance by gender was broadly similar.

Conclusions:

An EMR- based frailty prediction model using data available at 72 hours post-admission can identify frail patients early in their hospital stay, supporting targeted interventions and operational planning. Reduced performance in Indigenous patients highlights the need for culturally validated frailty indices and ongoing evaluation of equity in predictive modelling.