Factors influencing hospital admission and associated hospital costs of traffic victims admitted to an emergency department

Abstract

Abstract Objective Use of health care services and hospitalisation associated with traffic accidents impose an economic burden on society. This study analyses the determinants of hospitalisation and hospital costs associated with road traffic accidents in Belgium, using an emergency and hospital care dataset. Participants Traffic victims ( N =4645) admitted to the emergency department of the university hospital Brussels. Methods A logistic regression analysis and a generalised linear model (GLM) were used to analyse the probability of hospitalisation and costs respectively, controlling for roadway user categories, demographic (gender, age, and individual socioeconomic status (SES)) and clinical (nature, location, and severity of injury) characteristics. Results 20.3% of the traffic victims who went to the emergency department were hospitalised. The probability of hospitalisation, controlled for confounding factors, was significantly higher in victims aged 0–16 years (OR 2.46 (95% CI 1.74–3.49)) and ≥60 years (OR 1.52 (95% CI 1.06–2.17)) compared to those in age category 30–44 years. Motorcyclists, controlled for demographic and clinical factors, were significantly less likely to be hospitalised compared to pedestrians (OR 0.61 (95% CI 0.39–0.94)). Fractures and internal injuries were associated with the highest probability to be hospitalised. The GLM-analyses revealed that, controlled for confounding factors, men, older age and low SES patients were associated with higher hospital costs. The median hospital cost was €3273 (IQR €1733–€8891, 2011 euro price level) for inpatients. Conclusion In general, most of injury literature report ‘unit costs’ for fatally, severely and slightly injured traffic victims. This study demonstrates that other criteria such as traffic victim characteristics (gender, age, SES) and injury characteristics (nature, location, severity) need to be considered in order to give a more accurate picture of the probability of hospitalisation and associated medical costs.