Usage Patterns of Emergency Medical Services in Korea: Analysis of Patient Flow
Introduction
Emergency medical services (EMS) in Korea predominantly rely on public healthcare, which has led to disparities in the supply and demand for these services. Regional variations in the quality of EMS further exacerbate these disparities. Establishing an efficient emergency medical delivery system and ensuring equitable provision of EMS are crucial for addressing these issues. Regionalization strategies have been employed to improve patient outcomes and reduce treatment times by incorporating EMS bypass systems. A key component of these strategies is the establishment of medical treatment zones, which serve as the basis for resource allocation in the EMS system. Analyzing patient flow within these zones can provide valuable insights for effective resource allocation and policy development.
The patient origin method, which examines the distribution of patients using emergency medical institutions, is one approach to establishing medical treatment zones. This method can reveal usage patterns of EMS based on patient flow. Two key indicators used in this method are the Relevance Index (RI) and the Commitment Index (CI). RI measures the percentage of patients from a specific area who visit a particular hospital, indicating regional preferences. CI measures the percentage of patients from the local area who use the hospital, reflecting the hospital’s local commitment. Previous studies have used these indices to classify regional usage patterns, but these studies may not fully reflect recent changes in the EMS environment post-2010.
The National Emergency Department Information System (NEDIS), established in 2003, has been expanded to include 461 emergency medical institutions nationwide as of 2016. This study utilizes NEDIS data to calculate RI and CI indices for both emergency and critical emergency patients, followed by a cluster analysis to identify regional usage patterns.
Methods
Ethical approval for this study was obtained from the Regional Ethical Review Board in the National Emergency Medical Center, Seoul. The study analyzed anonymous patient visit data from the NEDIS for the year 2016, focusing on patients who visited emergency rooms. Patients with critical emergency illnesses were identified using severe Korea Classification of Diseases-7 (KCD-7) codes. The addresses of patients and emergency medical institutions were categorized by city, province, and district.
The RI and CI were calculated using the following equations:
RIij = Oij / Oi. CIij = Oij / O.j.
Where Oij is the number of patients from area i who visited hospital j, Oi. is the total number of patients from area i, and O.j is the total number of patients who visited hospital j. RI indicates the proportion of patients from a specific area who visit a particular hospital, while CI indicates the proportion of patients from the local area who use the hospital.
The number of clusters for the cluster analysis was determined using the NbClust package in R, which applies the majority rule to select the optimal number of clusters based on various indicators. The K-means method was used for cluster analysis, and the non-parametric Kruskal-Wallis test and Mann-Whitney test were used to examine differences between clusters.
Results
The study analyzed 8,389,766 cases (92.3% of total cases) from 411 emergency medical institutions, excluding 699,752 cases due to missing or unknown addresses. Additionally, 837,623 severe emergency cases from 409 institutions were analyzed. Thirty-six areas were excluded due to the absence of emergency medical institutions or lack of data transmission to NEDIS.
For total emergency department (ED) patients, the optimal number of clusters was determined to be three. Cluster 1 included 54 regions (25.2%) with low RI and high CI, indicating an outflow of patients to other areas. Cluster 2 included 58 regions (27.1%) with low CI, indicating an inflow of patients from other regions. Cluster 3 included 102 regions (47.7%) with high RI and high CI, indicating self-sufficient areas with balanced patient flow.
For critical ED patients, the optimal number of clusters was two. Cluster 1 included 129 regions (60.3%) with high CI, indicating an outflow of critically ill patients to other areas. Cluster 2 included 85 regions (39.7%) with low CI, indicating an inflow of critically ill patients from other regions.
Discussion
The study highlights regional disparities in EMS usage patterns in Korea. The cluster analysis revealed three distinct patterns for total ED patients and two patterns for critical ED patients. These findings can inform the establishment and selection of EMS areas and vulnerable EMS areas, ensuring more equitable resource allocation.
The results align with previous studies that have classified regional usage types based on RI and CI. However, this study provides a more comprehensive analysis by incorporating recent data from the NEDIS, reflecting changes in the EMS environment post-2010. The study also identifies specific regions with high outflows of critically ill patients, which should be prioritized for additional EMS resources.
The reliability of the NEDIS data was confirmed by the high input and completion rates of emergency room data, as reported in the national emergency medical institution evaluation. Future studies should explore determinants of RI and CI, such as patient demographics, disease classifications, and characteristics of emergency medical institutions, to further refine resource allocation strategies.
Conclusion
This study provides a detailed analysis of EMS usage patterns in Korea, using NEDIS data to calculate RI and CI indices and perform cluster analysis. The findings reveal distinct regional usage patterns for both total and critical ED patients, offering valuable insights for the establishment and selection of EMS areas and vulnerable EMS areas. By identifying regions with high outflows of critically ill patients, the study highlights areas in need of additional EMS resources. The results can inform policy development and resource allocation strategies to improve the equity and efficiency of EMS in Korea.
doi.org/10.1097/CM9.0000000000000062
Was this helpful?
0 / 0