Effect of the emergency department deciding to admit patients to internal medicine to resolve overcrowding in emergency medical centers

Article information

J EMS Med. 2025;.jemsm.2025.00213

Publication date (electronic) : 2025 December 10

doi : https://doi.org/10.35616/jemsm.2025.00213

Gun Bea Kim

Department of Emergency Medicine, National Health Insurance Service, Ilsan Hospital, Ilsan, Korea

Correspondence to: Gun Bea Kim Department of Emergency Medicine, National Health Insurance Service, Ilsan Hospital, 100 Ilsan-ro, Ilsandong-gu, Goyang 10444, Korea E-mail: drkgb74@nhimc.or.kr

Received 2025 October 28; Revised 2025 November 13; Accepted 2025 November 20.

Abstract

INTRODUCTION

Overcrowding in emergency medical centers (EMCs) remains a persistent concern both domestically and internationally and requires ongoing attention [1,2]. In response to this issue, South Korea implemented legislation on December 3, 2017, under Article 33-2 of the Emergency Medical Act and Article 20-2 of the Enforcement Rules, mandating that regional and district EMCs limit the proportion of patients staying in the emergency department for more than 24 hours to <5%.

One major contributor to overcrowding in EMCs is the influx of non-emergency patients, which impedes the timely care of critically ill patients. However, resolving this issue requires policy support, public awareness, and a broad consensus regarding the appropriate use of EMCs, making it a long-term challenge. Given the adverse effects of overcrowding on the prognosis of critically ill patients, EMCs worldwide are making efforts to reduce both overcrowding and prolonged patient stays [3-8].

Since the end of 2020, due to a shortage of internal medicine residents, our hospital has implemented a policy allowing emergency physicians to make admission decisions for internal medicine patients during weekday night-time hours for approximately 7–9 weeks each year. During these periods, rapid disposition of critically ill patients was consistently observed. Since March 2023, the decision to admit patients from the EMC to the Department of Internal Medicine (DIM) has been made during weekday night-time hours. This study aimed to evaluate the impact of implementing an EMC-issued admission policy for internal medicine patients on EMC length of stay (LOS) by comparing patient data before and after the policy was introduced. Based on these findings, we intend to expand the scope of this investigation in the future.

METHODS

This retrospective study included patients aged ≥18 years who were admitted to the DIM of our hospital’s EMC. We excluded all patients who were not admitted to that department. The study encompassed two periods: January 1, 2019, to December 31, 2019, and March 1, 2023, to February 29, 2024.

During the pre-intervention period (PrIP), the DIM made admission decisions for all patients who were admitted to the DIM. During the post-intervention period (PoIP), the DIM made admission decisions for patients admitted to the DIM from 7:30 AM to 4:30 PM on weekdays, and during all other hours the EMC made the admission decisions for patients who were admitted to the DIM.

Through a review of medical records, data such as sex, age, date and time of admission, Korean Triage and Acuity System (KTAS) classification level, severity of illness, department issuing the admission order, and LOS in the EMC, including nursing reception time, initial consultation time, admission order issuance time, and departure time were collected.

The triage of emergency patients at our hospital is performed using the KTAS, a five-level national standardized triage system utilized across the Republic of Korea. KTAS is an adaptation of the Canadian Triage and Acuity Scale and is designed to standardize the assessment of patient severity and waiting priority in the emergency department.

The variable severity of illness used in the tables is derived from severe illness diagnosis codes applied upon the patient’s discharge or admission. These codes specifically identify conditions designated by the Korean Ministry of Health and Welfare that require high-level emergency care resources and are associated with high mortality or major morbidity. For a comprehensive list of the diagnosis codes defining the severity of illness, please refer to Supplementary Table 1.

Statistical analyses were performed using SPSS (version 23.0; IBM Corp.). Categorical variables were expressed as counts (%), and continuous variables were presented as mean ± standard deviation. The chi-square test and t test were used for categorical and continuous variables, respectively. Statistical significance was set at P<0.05.

RESULTS

General characteristics of the patient group

During the study period, 90,158 patients visited the EMC of our hospital: 47,709 patients (52.9%) during the PrIP and 42,449 patients (47.1%) during the PoIP. Of these, 21,573 patients (23.9%) aged ≥18 years and admitted to DIM were included in the study, resulting in 10,164 patients (47.1%) during the PrIP and 11,409 patients (52.9%) during the PoIP (Fig. 1). Among the patients admitted during the PrIP, 5,114 (50.3%) were men and 5,050 (49.7%) were women. Among patients admitted during the PoIP, 5,817 (51.0%) were men and 5,592 (49.0%) were women. No statistically significant differences were observed between the two groups. The number of patients aged ≥60 years was significantly higher during the PoIP. The number of patients classified as having KTAS levels 1–3 (corresponding to moderate to severe acuity) and those with severe illnesses was also significantly higher during the PoIP (Table 1).

Fig. 1.

Flowchart of enrolled patients.

Table 1.

General characteristics

Length of each stage of care for patients admitted to the DIM in the period before and after the EMC made inpatient admission decisions

The time from the visit to the decision for admission, the time from the decision for admission to departure from the EMC, and the total time from the visit to departure from the EMC decreased significantly during the PoIP. However, when comparing the PrIP and PoIP, the time from the visit to the first consultation and the time from the first consultation to the decision for admission significantly increased (Table 2).

Table 2.

Comparison of time intervals of patients admitted to the DIM according to the decision of the EMC

Length of each stage of care during the period when the EMC issued admission orders to the DIM

The time from the visit to the decision for admission, the time from the decision for admission to departure from the EMC, and the total time from the visit to departure from the EMC significantly decreased during the PoIP (Table 3).

Table 3.

Comparison of time intervals of patients admitted to the DIM according to the decision of the EMC during PoIP

Length of each stage of care for patients with severe illness admitted to the DIM

When the EMC began issuing admission orders to DIM, the time from the visit to the first consultation with the internal medicine department increased, although this change was not statistically significant. Meanwhile, the time from the first consultation to the decision for admission, the time from the visit to the decision for admission, the time from the decision for admission to departure from the EMC, and the time from the visit to departure from the EMC significantly decreased (Table 4).

Table 4.

Comparison of time intervals of patients admitted to the DIM with SIDC according to the decision of the EMC

Length of each stage of care for patients with severe illness admitted to the DIM during the PoIP

When the EMC issued admission orders to the DIM, the time from the visit to the admission decision and the time from the visit to departure from the EMC decreased significantly. However, the difference in time from the decision for admission to departure from the EMC between the groups was not statistically significant (Table 5).

Table 5.

Comparison of time intervals of patients admitted to the DIM with SIDC according to the decision of the EMC during PoIP

Length of each stage of care according to the KTAS level of patients admitted to the DIM

For patients classified as KTAS levels 1–3, the times from the visit to the decision for admission, from the decision for admission to departure from the EMC, and from the visit to departure from the EMC decreased significantly during the PoIP (Table 6).

Table 6.

Comparison of time intervals of patients classified as KTAS levels 1–5 admitted to the DIM according to the decision of the EMC

Comparing the PrIP and the PoIP, the following observations were made:

• For patients classified as KTAS level 1, the time from the visit to the first consultation increased; however, this increase was not statistically significant. The time from the first consultation to the decision regarding admission orders decreased significantly.

• For patients classified as KTAS level 2, the time from the visit to the first consultation increased, and the time from the first consultation to the decision to admit decreased significantly.

• For patients classified as KTAS level 3, the time from the visit to the first consultation decreased; however, the difference was not statistically significant. The time from the first consultation to the decision for admission decreased significantly (Table 6).

• For patients classified as KTAS level 4, the time from the visit to the decision for admission increased during the PoIP, whereas the time from the decision for admission to departure decreased. However, the total time from the visit to departure increased significantly. Additionally, the times from the visit to the first consultation and from the first consultation to the decision for admission significantly increased (Table 6).

• For patients classified as KTAS level 5, the time from the visit to the decision for admission significantly decreased during the PoIP. However, the time from the decision for admission to departure increased, and the time from the visit to departure decreased, but the differences were not statistically significant. When comparing the PrIP and PoIP, the time from the visit to the first consultation increased, although not significantly. However, the time from the first consultation to the decision to admit decreased significantly (Table 6).

Time at each stage of care according to the KTAS level for patients admitted to the internal medicine department during the PoIP

For patients classified as KTAS level 2, the time from the decision for admission to departure increased during the PoIP; however, this difference was not statistically significant. For patients classified as KTAS level 5, the time from the decision for admission to departure was significantly longer. Time decreased significantly for all other variables (Table 7).

Table 7.

Comparison of time intervals of patients classified as KTAS levels 1–5 admitted to the DIM according to the decision of the EMC during PoIP

DISCUSSION

Chronic overcrowding of EMCs is a concern not only in South Korea but also worldwide. Overcrowding arises from multiple factors, including visits by non-emergency patients, delays in determining the disposition of patients with high-severity conditions, and the increasing burden of comorbidities in older adults. Efforts to resolve these challenges continue both domestically and internationally, including policies aimed at limiting LOS in EMCs [3-8].

Aging populations are expanding rapidly in South Korea and around the world. In 2018, individuals aged ≥65 years accounted for 14% of the South Korean population, marking its transition to an aging society; by 2025, this proportion is projected to exceed 20%, classifying South Korea as a super-aged society [9]. Older adults commonly have multiple comorbidities that contribute to higher disease severity and increased hospitalization needs [10-12]. In addition, because our institution is located near numerous nursing hospitals and long-term care facilities, we frequently receive transfers of patients whose clinical status has deteriorated rapidly.

Overcrowding at our hospital’s EMC, largely driven by delays in patient departure, limits our ability to manage critically ill emergency patients in the local community and undermines timely access to care. Various mitigation strategies have been implemented elsewhere, including expediting disposition decisions for older adults with complex conditions [13-18]. At our hospital, this approach began at the end of 2020 and has been progressively expanded based on its observed effectiveness. This strategy facilitates faster disposition for both critically ill patients and those with mild illnesses. By reducing the LOS in the EMC, it increases capacity to accept patients transferred from 119 and nearby regional hospitals, ultimately improving emergency care availability and patient safety within the community.

During PoIP, there was a clear increase in patient age and illness severity, reflected in higher KTAS scores and a greater number of patients with severe illnesses (Table 1). First, for patients admitted to the DIM during the PoIP, the times from the visit to the first consultation and from consultation to the admission decision increased significantly, likely because of the greater overall severity of illness. However, the reduction in time from visit to departure between groups, when considered with PoIP results (Table 3), indicates a meaningful decrease in total LOS (Table 2). Second, for patients with severe illnesses during the PoIP, the times from the visit to the decision for admission, from decision to departure, and from the visit to departure all decreased significantly. Notably, overall LOS was shortened primarily through the reduction of decision-to-departure time rather than the time to reach the admission decision (Table 4). This finding highlights the effectiveness of the intervention in facilitating the movement of admitted patients out of the EMC. Third, for patients classified as KTAS levels 1–3, the times from the visit to the admission decision, from the decision to departure, and from the visit to departure significantly decreased. When considered with PoIP findings (Table 7), these results suggest that issuing admission orders directly from the EMC helped shorten the time to admission decisions and subsequently decreased total LOS (Table 6). Fourth, for patients with KTAS level 4, the time from the visit to the admission decision increased significantly during the PoIP. Although the decision-to-departure time decreased, the total LOS increased, likely reflecting increased EMC workload due to higher patient acuity. For KTAS level 5 patients, the time from the visit to the admission decision decreased significantly, while the time from the decision to departure increased; overall LOS decreased but without statistical significance (Table 6). Finally, when LOS was evaluated by KTAS level during the PoIP, KTAS level 2 patients showed a non-significant increase in decision-to-departure time, while KTAS level 5 patients showed a significant increase. Nonetheless, across all KTAS levels, issuing admission orders from the EMC significantly shortened both the time from the visit to the admission decision and the overall LOS (Table 7).

Several limitations should be noted. First, during the PoIP, the number of patients classified as having moderate-to-severe illnesses (KTAS levels 1–3) increased significantly, which may have attenuated the observed reductions in LOS. Second, when the EMC began issuing admission orders, our institution had recently been designated as a regional EMC. During this period, multiple departments undertook intensive efforts to improve the LOS for critically ill patients, which may have introduced selection bias. Third, this study was conducted at a single institution within a specific region, limiting generalizability to other hospital settings.

In conclusion, issuing admission orders to the DIM directly from the EMC appears to be an effective strategy for alleviating chronic overcrowding. Nevertheless, successful implementation requires careful consideration of the characteristics of each internal medicine specialty and the availability of sufficient hospital personnel. Future studies should evaluate patient outcomes according to illness severity to clarify how these interventions influence the prognosis of critically ill patients. Overall, reducing EMC LOS is a complex challenge shaped by numerous operational and clinical factors, yet the issuance of admission orders by the EMC contributes meaningfully to shortening LOS, particularly for patients with moderate or high severity. Given that national EMC assessments emphasize the LOS of critically ill patients, improvements in LOS for these groups may hold substantial significance.

Supplementary Material

Article information Continued