Abstract
The objective of a study was to examine the systemwide effects of prehospital emergency medical services (EMS) resources quantitatively on crash outcomes throughout the entire South Korean freeway system. Latent class cluster and binomial probit regression models were combined to achieve this objective. In the cluster-based binomial probit regression, surrogate measures for prehospital EMS resources were obtained by combining medical service portals, freeway heliport maps, and freeway network log data in the crash data set. As a result, eight latent class clusters of crashes were determined on the basis of features associated with EMS resources, province, roadway, and traffic conditions at the scene of the crash. On-scene and recovery times were commonly significant in increasing the probability of fatal crashes in both entire groups and in each group of crashes, while the nearest ramp location and number of nearby EMS facilities significantly affected fatal crashes for a certain group of crashes. The findings provide meaningful insights that can enhance EMS training programs for initial medical aid and postcrash traffic management on all provincial freeways. Supplemental nearby EMS facilities and access points to them are needed particularly in South Korea’s southeastern and central province freeway sections, respectively. This research is the first data-driven study to assess systemwide EMS resources for the entire South Korean freeway system by using multiple data sources. It would contribute to informed decision making for future EMS provision.
Get full access to this article
View all access options for this article.