Emergency Overcrowding is Seasonal in a Hospital in Hainan Province, Southern China

Background Emergency overcrowding is a serious problem for hospital staff in China. Immigrant patients caused emergency overcrowding in a hospital in southern China. However, the demographic characteristics of this population have not been determined. Methods Data on all emergency visitors, including patients who stayed in emergency department (ED) observation units or critical care units were collected and saved in an Excel sheet. Patient characteristics, including place of origin, type of disease, visiting time, and factors associated with prolonged stay in the ED, were analyzed. Results November 2017


Background
Overcrowding is a serious problem in developed countries (1)(2)(3)(4)17) and in China (5) and is primarily due to the large number of emergency visitors, shortage of emergency departments (EDs), nurses, and physicians, and limited access to health services. Overcrowding can lead to poor patient outcomes and increased length of stay (17). Some studies addressed these problems in China. A study (2017) evaluated 18 tertiary hospitals from Beijing and showed that 10 EDs had more than 100,000 patients annually (6). In each hospital, more than 11,000 patients stayed in the observation unit with a mean length of stay (LOS) of more than 5 days, and only 20% of them received inpatient care (6). More than 260,000 patients visited the ED of Shanghai Ruijin Hospital; of these, 2702 patients stayed in the ED for more than 48 hours from September 2009 to August 2010 (7), and (7). One study from Zhejiang province showed that access block (period from admission to the ED to admission to an inpatient unit) was a major contributor to prolonged LOS in the ED (8).
Our hospital is located in Sanya, a tropical city in Hainan Province in southern China. Many retired people from mainland migrate to Hainan in winter and, for this reason, are known as "snowbirds" (9)(10). Seasonal migrations cause overcrowding in the ED of our hospital. The objective of this study is to investigate the epidemiology of ED visitors and factors affecting the LOS in the ED.

Study setting
This retrospective cohort study was conducted from August 2014 to July 2019 and was approved by the Institutional Review Board of our hospital. The study was conducted in an adult ED at a tertiary hospital with 660 inpatient beds and approximately 30,000 emergency patients annually. The ED has four subdepartments: outpatient clinic, observation unit (10 beds), critical care unit (10 beds), and ward (11 beds).
One or two junior physicians work in each sub-department and one senior physician is on duty. Patients presenting to the ED were triaged by experienced nurses according to guidelines based on the Emergency Severity Index (ESI), which grouped patients into ve triage levels (Table 1) (11). High-acuity patients (levels 1 and 2, and part of level 3) were referred to observation units or critical care units, whereas lowacuity patients (part of level 3 and levels 4 and 5) were referred to emergency outpatient clinics.

Study Design
All the emergency visitor's data were collected and saved in an Excel sheet, including patient age, gender, place of origin, visiting month, and type of disease. The LOS in the ED observation unit or critical care unit was also calculated. Other factors that might favor a higher LOS, including age, visiting season, and type of disease, were analyzed.
Preliminary diagnosis was de ned in each department according to the medical condition (medical diseases or surgical diseases). Medical diseases included respiratory, digestive, neurological, cardiovascular, endocrine and rheumatic, emergency, dermatological, and other conditions. Surgical diseases included mild trauma, orthopedic, urological, general surgery, and other conditions. Patient destination after discharge from the observation unit and critical care unit was divided into four groups: inpatient admission, transfer to other hospitals, death in the ED, and discharge after recovery or discharge without receiving advanced therapy.

Statistical analysis
Parametric data were presented as means standard deviation (SD) and frequencies. The categorical variables were compared using chi-square or Fisher's exact tests, where appropriate. Analysis of variance was performed to assess differences in the LOS in the ED among different groups and Student's ttest to analyze differences between two groups. A Cox proportional hazards model was constructed to evaluate the factors related to LOS. A two-tailed P < 0.05 was considered statistically signi cant. All analyses were  Table 2. The characteristics were different between Hainan and other provinces. The number of patients with older age, and respiratory diseases, was smaller and higher of mild trauma patients in Hainan Province (Fig. 4).  (Fig. 4).
The risk factors assessed by multivariate Cox analysis are shown in Table 3. The factors associated with prolonged LOS in the ED were the month of admission (November to March) and the presence of respiratory or digestive diseases. The LOS was shorter in patients with acute myocadiac infarction needing PCI.

Discussion
To our knowledge, this study is the rst to evaluate emergency crowding by immigrants in southern China. The number of patients visiting the ED and their LOS were higher from November to March of next year. The demographic characteristics were different between local and immigrant patients.
In recent years, because of air pollution and severe weather, a large number of retired people migrate to the southern coast in warmer months (9-10). Hainan Province is the most popular destination and attracted a large number of immigrants after 2009 (9-10). There were more than one million immigrants in Hainan Province in 2015, which was double the number in 2010 (9-10).
The period with the highest number of patients-November to March of next year-was designated the "medical boom season" by hospital staff. The months from November to April are often the coldest in mainland China, especially in northeast China. Our sample included many people from the north and northeast, including Heilongjiang, Beijing, Jilin, and Liaoning. Previous studies showed that many retired people, especially those with chronic diseases, migrated from these provinces to Hainan (9-10, 12,14). It has been reported that the good weather conditions in Hainan can improve stroke recovery, blood pressure, and chronic conditions, including pulmonary diseases (12)(13)(14)(15)(16). This fact could explain why relatively more patients with respiratory and cardiovascular diseases visited the ED in Hainan.
Emergency crowding in our hospital was not worse than the problem reported in larger cities in China, including Beijing, Shanghai, and Hangzhou (6-8). The mean LOS in the ED was 1.2 days, and only 50% of the patients visiting this department received inpatient care. However, in Beijing, some patients stayed more than 5 days in the observation room, and only approximately 10% could wait for admission. In Shanghai, 12% were admitted to hospital wards, and some patients had to stay in the ED for extended periods, even more than 1 month. Similar results were found in Zhejiang. The problem of emergency overcrowding in China seems to be worse than in other countries. In Australia and the United Kingdom, the threshold of access block is 8 and 4 hours, respectively (3,4).
The number of patients with respiratory, digestive, or cardiovascular diseases who were not admitted to the ward and had to spend a long time in the ED was higher than that of patients with other conditions. This result may be due to the limited capacity of specialized wards and other factors. Previous studies showed that specialized wards refused to admit patients with multiple organ dysfunction syndrome (6-8). Our results indicated that patients requiring emergency operations were transferred from the resuscitation room faster than other patients, and this may be because the surgical patients in our hospital were given priority to inpatient care.
The present study makes a signi cant contribution to the problem of emergency overcrowding. In the "medical boom season," hospitals need to hire additional staff and provide more ward beds, and the government should take measures to improve local hospitals' capacity to serve for more patients. The present study has limitations. First, this study was performed in a single center and, therefore, did not re ect the situation in the entire province. Second, data were incomplete because we did not determine the distances traveled by migrants and commodity factors affecting the long stay in the ED. The patients discharged from hospital after recovery and those discharged without receiving advanced therapy were not distinguished.

Conclusions
The results indicated that emergency overcrowding in a hospital in southern China was seasonal and was caused by an increasing number of migrants. Several patients had to stay in the ED for a long time, especially those with special commodity diseases. Hospitals should take appropriate measures to confront problems related to emergency overcrowding. Hospital, which waived the requirement to obtain informed consent from the subjects.

Consent for publication
Not applicable.

Availability of data and materials
The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
We, the authors, declare that we have no competing interests.