Abstract
INTRODUCTION: The Glasgow Admission Prediction Score (GAPS) is a triage tool that was developed in the United Kingdom to predict hospital admission based on available patient data upon arrival at emergency departments (EDs). Despite being associated with length of hospital stay, 6-month mortality, and readmission, its validity has only been assessed in the UK. Its generalizability therefore remains unclear. This study aimed to assess the predictive performance of the GAPS for short- and intermediate-term outcomes in a Japanese ED setting. METHODS: We conducted a retrospective cohort study of ambulance-transported ED visits between December 2020 and September 2023 at a Japanese general hospital. After excluding pediatric patients and those with missing or inconsistent data, 22,179 encounters were analyzed. GAPS scores were calculated upon ED arrival and stratified into tertiles. The outcomes included 30-day mortality, 30-day emergency re-transportation, 100-day hospital readmission, and length of inpatient stay. Kaplan-Meier and Cox proportional hazards analyses were performed. RESULTS: Higher GAPS scores were significantly associated with less favorable outcomes. Each 1-point increase was linked to a 10.3% increase in 30-day mortality risk (hazard ratio [HR] = 1.10; 95% confidence interval: 1.10-1.11), or a four-fold increase per 15-point rise (HR = 4.32). Similar associations were observed for re-transportation (HR = 1.03) and hospital readmission (HR = 1.09). Higher GAPS scores were also associated with longer hospital stays (HR for discharge = 0.98). CONCLUSION: GAPS presents a practical tool for predicting ambulance-transported ED encounter outcomes in Japan, although its broader applicability warrants further research.