|
Abstract
|
Introduction
Nowadays, a leading cause of mortality and morbidity of elderly is vehicle crash. Given the vulnerability of the senior victims in crashes, the decisions of emergency vehicle services as well as triage to shock trauma centers become extremely crucial for this cohort to receive timely transport service and proper trauma care. However, many who could be saved by shock trauma centers are either not transported via emergency medical services (EMS) or treated by non-trauma hospitals.
Method
This paper explores possible ways to improve the EMS and trauma triage decision processes. First of all, transportation-related data sources, such as traffic volumes and time-dependent vehicle speeds, have been integrated with EMS and hospital records, offering measurements of exposure to crash risks. The critical transportation information is typically missing from the field triage process but becomes readily available in real-time to support decision-making. Then the integrated data has been employed to construct machine learning models with the new predictors.
Results
Predictions of EMS transport and triage to shock trauma centers have been analyzed using a Maryland dataset with records of over 55,000 elderly patients. Compared to benchmark models without transportation information as predictors, the models trained with integrated data exhibit superior prediction accuracy.
Conclusions
This paper is among the first to develop and demonstrate a methodological framework of integrating transportation-sector data with health-related data to support various decision-making scenarios in transportation safety, emergency responses, and trauma-care triage. It has empirically demonstrated that integrated transportation and health data significantly improves the recall rates of decisions of EMS and trauma triage and showcased that under triage, a major issue for high-risk trauma patients, can be significantly addressed.
Language: en |