Prediction of inpatient falls and key predictors using machine learning applied to electronic health records: a retrospective cohort study in a tertiary hospital in Türkiye

Clinical Snapshot

Key Findings

• P Value: p<0.001 for model comparisons

• Effect Size: AUC 0.821 for random forest quantile classifier

• Primary Outcome: Inpatient falls occurrence

• Nnt Or Sensitivity: Sensitivity 0.799 (95% CI 0.734-0.867), Specificity 0.722 (95% CI 0.691-0.755)

• Confidence Interval: 95% CI 0.781 to 0.861

Clinical Application

Requires sophisticated electronic health record infrastructure and machine learning implementation capabilities Relevant to Australian tertiary hospitals with advanced EHR systems; aligns with ACSQHC National Safety and Quality Health Service Standards for preventing falls Adult inpatients in tertiary hospital settings with comprehensive electronic health records

Abstract

OBJECTIVE: Traditional fall risk tools are often inaccurate and burdensome. This study aims to develop a predictive model for inpatient falls and identify the most influential variables using machine learning applied to electronic health record data. DESIGN: A retrospective cohort study. SETTING: A large tertiary university hospital in Türkiye. PARTICIPANTS: Adult patients (≥18 years) hospitalised in a university hospital between January 2017 and June 2023. PRIMARY OUTCOME MEASURES: Occurrence of inpatient falls recorded in incident reporting systems. RESULTS: A total of 518 fallers were identified and compared with 3121 non-fallers. Fallers were significantly older (median 68.5 vs 64 years, p<0.001), had longer hospital stays (16 vs 12 days, p<0.001) and higher comorbidity burden (p<0.001). They were also more likely to receive high fall-risk medications (86% vs 76%, p<0.001) and had a higher prevalence of mental disorders (26.3% vs 17.2%, p<0.001). The random forest quantile classifier demonstrated the highest discrimination (area under the curve (AUC) 0.821, 95% CI 0.781 to 0.861), with balanced sensitivity (0.799, 95% CI 0.734 to 0.867) and specificity (0.722, 95% CI 0.691 to 0.755). Pairwise comparisons using DeLong's test showed significantly higher AUC compared with several models (adjusted p=0.001), while differences with eXtreme Gradient Boosting were not significant (p=0.240). Key predictors included comorbidity burden, age, number of fall-risk-increasing medications and laboratory variables such as eosinophil and basophil counts, blood urea nitrogen, sodium and red blood cell count. CONCLUSION: Machine learning models using electronic health data can predict inpatient falls and reveal key risk factors. The random forest quantile classifier offers a promising approach for improving fall risk prediction in imbalanced clinical datasets.

References

1. 1.Baris, V. K., & Hudaverdi, B. (2026). Prediction of inpatient falls and key predictors using machine learning applied to electronic health records: a retrospective cohort study in a tertiary hospital in Türkiye. BMJ Open. https://doi.org/10.1136/bmjopen-2025-113384