Article

The influence of crowding on clinical practice in the emergency department

a b s t r a c t

Background: This study aimed to clarify the association between the crowding and clinical practice in the emer- gency department (ED).

Methods: This 1-year retrospective cohort study conducted in two EDs in Taiwan included 70,222 adult non-trau- ma visits during the day shift between July 1, 2011, and June 30, 2012. The ED occupancy status, determined by the number of patients staying during their time of visit, was used to measure crowding, grouped into four quar- tiles, and analyzed in reference to the clinical practice. The clinical practices included decision-making time, pa- tient length of stay, patient disposition, and use of Laboratory examinations and computed tomography (CT). Result: The four quartiles of occupancy statuses determined by the number of patients staying during their time of visit were b 24, 24-39, 39-62, and N 62. Comparing N 62 and b 24 ED occupancy statuses, the physicians’ decision- making time and patients’ length of stay increased by 0.3 h and 1.1 h, respectively. The percentage of patients discharged from the ED decreased by 15.5% as the ED observation, general ward, and intensive care unit admis- sions increased by 10.9%, 4%, and 0.7%, respectively. CT and laboratory examination slightly increased in the fourth quartile of ED occupancy.

Conclusion: Overcrowding in the ED might increase physicians’ decision-making time and patients’ length of stay, and more patients could be admitted to Observation Units or an inpatient department. The use of CT and labora- tory examinations would also increase. All of these could lead more patients to stay in the ED.

(C) 2017

Introduction

According to a task force report from the American College of Emer- gency Physicians in 2008, emergency department (ED) crowding is gen- erally defined as occurring “when there is no space left to meet the timely needs of the next patient who needs emergency care” [1]. In ad- dition, since the first article discussed ED crowding about 25 years ago [2], issues that pose a potential threat to patient health and public safety must be confronted urgently [3]. Various studies have been published in the past few decades regarding the association between ED crowding and patient outcomes, and most of them revealed an unfavorable result [4-8]. Sion et al. also found that increased mortality among admitted critically ill patients is associated with ED crowding [6]. Sun et al. de- scribed that periods of high ED crowding are associated with increased

* Corresponding author at: No. 123, Dapi Rd., Niaosong Dist., Kaohsiung City 833, Taiwan.

E-mail address: [email protected] (C.-J. Li).

inpatient mortality, length of stay (LOS), and costs for admitted patients [9]. ED crowding also affects patient satisfaction. Pine et al. described that poor ED service experience, measured by ED hallway use and prolongED boarding time, are adversely associated with ED satisfaction and even predict lower satisfaction with the entire hospitalization [10]. ED crowding was also associated with worse patient perceptions of clinician-patient communication, and poor communication may further increase the risk of adverse physiologic outcomes [11].

In addition to patient outcomes and satisfaction, ED crowding might also influence the clinical practice of emergency physicians (EPs). Some studies have discussed ED crowding in association with compromise of medical behavior [12-14]. One study that was conducted in Cincinnati suggested that ED crowding is associated with a delay in the reassess- ment of critically ill pediatric patients [15]. Another study also described that there is an association between ED crowding and delayed adminis- tration of analgesia in patients presenting with acute abdominal pain [16]. According to our experience, EPs might increase the use of diag- nostic tools such as laboratory examinations or imaging studies due to

http://dx.doi.org/10.1016/j.ajem.2017.07.011

0735-6757/(C) 2017

Table 1

Demographics of the patients.

From July 1, 2011, to June 30, 2012, all adult non-trauma patients who presented to the EDs during the day shift were included in the

1st quartile

2nd quartile

3rd quartile

4th quartile

P

value

analysis. The day shift in the two EDs comprised of 8 h corresponding

to the emergency physicians’ working hours, which was 7:00 to 15:00

Age

58.1

57.1

56.8

57.1

b0.001

and 7:30 to 15:30, respectively. The attending physicians should com-

+-18.57

+-18.88

+-19.07

+-19.17

plete all day-shift patients’ disposition (including discharge, hospital ad-

Male

8960

8496

9305

9021

b0.001

mission, or ED observation room admission).

Urgent

49.8%

3521

19.6%

49.6%

3644

21.3%

52.0%

4335

24.2%

52.5%

4505

26.2%

b0.001

In both EDs, three attending physicians were in charge in the day shift. As all study sites were teaching medical units, residents assisted

Hospital

Northern

2125

5853

13,885

17,098

b0.001

in the treatment of ED patients under an attending physician’s supervi-

11.8%

34.1%

77.6%

99.4%

sion. Overall, 76 full-time attending physicians were involved in this

Southern

15,859

88.2%

11,287

65.9%

4014

22.4%

101

0.6%

study, 59 and 17 for each medical center.

the lack of time for complete physical examinations and giving patients instruction, but EPs could also decrease the use of diagnostic tools to fa- cilitate patient disposition. To clarify this problem, we investigated the association between ED crowding and three clinical parameters: clinical efficiency, Diagnostic tool use, and patient disposition, which have been used to evaluate the clinical practice of EPs before [17]. The aim of this study was to clarify how ED crowding influences the clinical practice of EPs.

Materials and methods

Study design

This was a retrospective 1-year cohort study approved by the insti- tutional review board of the Chang Gung Medical Foundation. All pa- tients’ and physicians’ records and information were anonymized and de-identified before analysis.

Study setting and participants

This study was conducted in two EDs of the tertiary referral medical centers located in northern and southern Taiwan separately. The total bED capacity of these two EDs were 80 and 60, and total beds of obser- vation room were 160 and 148, respectively. An observation room was set up for short stay to follow up clinical changes or status upon hospital admission. The inpatient beds are N 3500 and 2500 in the two hospitals.

Measures

The ED occupancy status was used to measure crowding. Patients were grouped into four occupancy statuses based on the number of pa- tients staying in ED during their time of visit and divided into quartile [4, 18]. The main analysis involved the differences of clinical practice in ref- erence to these four occupancy statuses. The patient demographic and clinical information were drawn from the ED administrative database. All ED visits were classified into different disease acuities based on Five Level Taiwan Triage and Acuity Scale , which is a commonly used triage system formulated by the Department of Health in Taiwan [19]. According to the TTAS, patient acuity were determined based on their initial vital sign (heart rate, blood pressure, respiration rate, oxy- gen saturation) and chief complaints. For example, a patient presenting dyspnea with unstable vital sign would be determined as triage 2, or even triage 1, if immediate resuscitation is needed. Based on these criteria, patients identified as triage levels 1 and 2 should be seen imme- diately or within 10 min, respectively, and are defined as urgent. Pa- tients with triage levels 3, 4, and 5 should be assessed within 30, 60, or 120 min, respectively, and are classified as non-urgent. The patient diagnosis were divided into six categories, i.e., neuromuscular (ICD-9- CM: 320-389 and 430-438), gastrointestinal (ICD-9-CM: 520-579),

genitourinary (ICD-9-CM: 580-629), respiratory (ICD-9-CM: 460-

519), cardiovascular (ICD-9-CM: 390-429 and 439-459), neoplasm (ICD-9-CM: 140-239) diseases. The clinical practices were divided into three parts, including clinical efficiency, patient disposition, and diagnostic tool use. The clinical efficiency was further divided into

Fig. 1. The distribution of 4 quartile ED occupancy in different diagnosis (p b 0.001) Abbreviation of the diagnosis: neuromuscular disease (Neuro), gastrointestinal disease (GI), genitourinary disease, respiratory disease (Resp), cardiovascular disease (CV), disease of neoplasm (Neoplasm) and others.

Fig. 2. The distribution of 4 quartile ED occupancy in different day of the week (p b 0.001) Abbreviation of the day of the week: Monday (Mon), Tuesday (Tue), Wednesday (Wed), Thursday (Thu), Friday (Fri), Saturday (Sat), and Sunday (Sun).

decision-making time of EP (the time interval between the patient reg- istration and the EP completing the Disposition decision) [20] and ED LOS (the time interval between patient registration and patient leaving ED). Patient disposition was categorized into ED discharge, ED observa- tion (admission to ED observation unit), general ward admission, inten- sive care unit (ICU) admission, and ED mortality. Finally, ED diagnostic tool use was assessed based on the Diagnostic investigations ordered by physicians, including computed tomography (CT) scans and labora- tory examination (e.g., complete blood count, blood chemistry, Urine analysis, stool analysis, or influenza screen test).

Data analysis

Patients’ age was reported as means with standard deviations (SDs) and was analyzed using analysis of variance (ANOVA). Because the

distributions of decision-making time and ED LOS were not normal, we used medians with interquartile ranges (IQRs) and the nonparamet- ric Kruskal-Wallis test to describe and evaluate their associations with ED crowding. All other variables were reported as numbers with per- centages and analyzed using the ?2 test. To analyze the associations be- tween clinical practice indicators and ED crowding while adjusting for potential confounding factors (age, sex, disease acuity, medical setting, different diagnosis category, and day of the week, and month of patient visit), multinomial logistic regression with disposition of discharge as dependent variable was selected for patient disposition, and binomial logistic regression for diagnostic tool use. Effects were estimated in terms of adjusted odds ratios (aORs) and the corresponding 95% confi- dence intervals (CIs). Significance testing was two-sided, and the signif- icance threshold was set at P b 0.05. The SPSS version 12.0 (SPSS, Chicago, IL) was used for all statistical analyses.

Fig. 3. The distribution of 4 quartile ED occupancy in different month (p b 0.001) Abbreviation of the month: January (Jan), February (Feb), March (Mar), April (Apr), May (May), June (Jun), July (Jul), August (Aug), September (Sep), October (Oct), November (Nov), and December (Dec).

Table 3

The association between ED crowding and patient disposition by logistic regression

Disposition/level of ED crowding

1st quartile

2nd quartile 3rd quartile 4th quartile

Ref.

aOR

95% CI

aOR

95% CI

aOR

95% CI

Observation in ED

1

2.0

1.94-2.17

3.2

2.95-3.39

5.2

4.79-5.68

Admission to

ward

1

1.5

1.39-1.56

2.0

1.82-2.10

2.5

2.31-2.74

Admission to ICU

1

1.5

1.33-1.76

2.0

1.67-2.35

2.7

2.22-3.34

ED mortality

1

0.9

0.59-1.46

0.7

0.37-1.17

1.0

0.50-1.85

Regression adjust for age, sex, disease acuity, medical setting, different diagnosis category, and day of the week, and month of patient visit.

Fig. 4. The median of EP decision making time and patient ED length of stay of 4 quartile ED occupancy. Note: Decision making time (hr) represents the time interval between patient registration and emergency physician completing disposition decision. ED length of stay (hr) represents the time interval between patient registration and leaving ED. The two types of data were presented as median with interquartile ranges (IQRs) and nonparametric Kruskal-Wallis tests were used to evaluate the differences. Both P b

0.001 for decision making time and ED length of stay. [Decision making time: b24: 1.6 (1.2); 24-39: 1.7 (1.3); 39-62: 1.8 (1.6); N 62: 1.9 (1.6); ED length of stay: b 24: 3.8

(21.1); 24-39: 4.0 (23.2); 39-62: 4.4 (25.6); N 62: 4.9 (28.0); Data are median (IQR)].

Results

During this one-year study, there were 70,222 ED visits. The four oc- cupancy statuses according to the number of patients staying in the ED at their visit time by quartile were b 24, 24-39, 39-62, and N 62. Each quartile contained visits of 17,984 (25.6%), 17,140 (24.4%), 18,350

(26.1%), and 16,748 (23.9%) patients, respectively. The patients’ rele- vant demographic factors (age, sex, disease acuity, and medical set- tings) (Table 1) and the distribution of the six diagnosis categories (Fig. 1) in the four study groups were statistically different. ED crowding was most frequent on Tuesday, followed by Monday and Wednesday (Fig. 2). Considering the difference of ED visits of each month, ED crowding was more severe in March, followed by February and April, than it was in other months (Fig. 3).

There was an association between ED crowding and decision-mak- ing time and LOS in the ED. The decision-making time increased by

0.1 h and the LOS in the ED increased by 0.2 h in the second quartile of ED occupancy, 0.2 and 0.6 h in the third quartile, and 0.3 and 1.1 h in the fourth quartile compared with the first quartile of ED occupancy (Fig. 4). Compared with the first quartile of ED occupancy, in the second quartile of ED occupancy, the percentage of patients who were discharged from the ED decreased by 8.8%, by 11.1% in the third quartile, and by 15.5% in the fourth quartile, with increases of 7.4%, 7.5%, and

Table 2

Patient disposition and diagnostic tool use in the emergency department

10.9%, respectively, in ED observation; 1.4%, 3.3%, and 4%, respectively, in general ward admission; and 0.2%, 0.5%, and 0.7%, respectively, in ICU admission. Although there was a statistical difference in the use of diagnostic tools in the four study groups, the difference was less obvious (Table 2).

After controlling for potential confounding factors (age, sex, disease acuity, medical setting, different diagnosis category, day of the week, and month) with a regression model, we found that ED crowding was still related to patient disposition. With the increase of occupancy, the incidence of ED observation, general ward admission, and ICU admis- sion increased, but were not related to mortality in the ED (Table 3). There was a slight increase in the use of CT and laboratory examinations in the fourth quartile of ED occupancy (Table 4).

Discussion

We studied three clinical parameters that were associated with ED crowding, including EP practice efficiency, ED diagnostic tool use, and patient disposition. We evaluated EP practice under different ED occu- pancy statuses. Increased ED occupancy was associated with particular days of the week and months, and it was also related to clinical practice adjustment. First, ED crowding was associated with the efficiency of clinical practice. With increasing ED occupancy, EPs might spend more time on decision making and patients might have a longer LOS in the ED. These might cause a delay in patient treatment. Rishi et al. reported that delayed administration of antibiotics for pneumonia was associated with ED crowding [21]. This delay in treatment was also observed in an- other retrospective study from a single, urban, tertiary center, where delayed resuscitation effort was found in crowded EDs [22]. These de- lays in treatment might influence the patients’ prognosis. Sun et al. reported that periods of ED crowding are associated with increased in- patient mortality [9]. Sion et al. also found that increased mortality among admitted critically ill patients is associated with ED crowding [6]. On the other hand, increased LOS in the ED may be a result of ED crowding and may also contribute to it. Further, according to previous studies, the rotation rate and crowding of different inpatient depart- ments can also affect ED LOS [23,24]. In this study, due to limitations of the data, the crowding status of inpatient departments was not considered, and this might have influenced the result.

1st quartile

2nd quartile

3rd quartile

4th quartile

P

value

Second, ED crowding also influenced patient disposition. Increased ED occupancy was associated with greater ED observation and hospital

Disposition

Discharge

8009

6114

5973

4993

b0.001

admission, including admission to the general ward and ICU. This

44.5%

35.7%

33.4%

29.0%

ED observation

4103

5180

5432

5800

22.8% 30.2% 30.3% 33.7%

Table 4

General ward

5295

5273

5854

5741

The association between ED crowding and diagnostic tool use by logistic regression

admission

29.4%

30.8%

32.7%

33.4%

ICU admission

525

534

608

626

Diagnostic tool 1st 2nd quartile 3rd quartile 4th quartile

2.9%

3.1%

3.4%

3.6%

use/level of ED quartile

Ref. aOR 95% CI aOR 95% CI aOR 95% CI

ED mortality

52

39

32

39

crowding

0.3%

0.2%

0.2%

0.2%

Diagnostic

CT

2427

2340

2589

2634

b0.001

CT

1

1.0 0.95-1.08 1.1 0.99-1.16 1.1 1.03-1.25

tool use

13.5%

13.7%

14.5%

15.3%

Laboratory exam

1

1.0 0.91-1.01 1.0 0.96-1.09 1.1 1.03-1.20

Regression adjust for age, sex, disease acuity, medical setting, different diagnosis category, and day of the week, and month of patient visit.

Laboratory

12,881

11,986

12,505

12,250

b0.001

examination

71.6%

69.9%

69.9%

71.2%

finding confirmed our hypothesis that stress from ED crowding will alter EPs’ disposition when making decisions. With less time to evaluate, manage, and instruct patients in a crowded ED, physicians tend to keep patients in the ED longer or just admit them to make sure that adverse outcomes do not occur. This situation was also demonstrated in other studies. A retrospective study that was conducted in Canada showed that among patients with acute stroke or transient ischemic attack, as ED crowding worsens, the likelihood of hospitalization increases [14]. Again, this disposition adjustment might lead to a vicious circle that eventually worsens ED crowding.

Finally, ED crowding might increase the use of diagnostic tools in the ED. In a previous study, ED crowding was associated with increased health care costs, and increased LOS in the ED and inpatient admissions contributed to cost [25]. In this study, CT and laboratory examination were selected as indicators for two reasons. First, CT is the most fre- quently used imaging tool for further confirmation of a diagnosis, with a relatively high expense compared with X-ray imaging. Second, both CT and laboratory examinations take a longer time to perform in the ED than other diagnostic methods. Based on the results of this study, in the fourth quartile of ED occupancy, the use of CT and laboratory ex- amination was increased. One possible explanation for this was that the EP might order more examinations in an extremely stressful, crowded ED. The increase in examinations can not only cause health care costs to surge but also can prolong the patients’ LOS in the ED, which could worsen ED crowding [26].

ED crowding might increase EPs’ decision-making time and patients’ LOS in the ED, and more patients could be admitted to observation units or an inpatient department. The use of CT and laboratory examinations would also increase. All of these could lead more patients to stay in the ED. Therefore, crowding begets more crowding in a cyclical pattern. Since the medical facility cannot be expensed without limitation, one possible resolution is to increase medical staff on particular days of the week or months. This might release the EPs’ stress and improve patient dispositions, which might improve ED crowding.

Limitations

This study has several limitations. First, the two study hospitals belonged to the same healthcare system, which might limit the general- izability of the conclusions; trauma patients were not included in this study because trauma surgeons rather than emergency physicians were in charge of these patients in the selected hospital, resulting to dif- ficulty in getting these data. Second, data with regard to the technical quality and appropriateness of clinical care were not available in the electronic system, which limits the evaluation of the effects of medical care factors on patient outcomes. Finally, only a limited collection of confounding factors was considered, and the crowding status of inpa- tient departments was not considered; therefore, the results of this study should be interpreted cautiously.

Data sharing

No additional data available.

Conflicts of interest statement

No conflict of interest for all authors.

Funding sources/disclosures

This study was supported in part by Research Grants from the Kaohsiung Chang Gung Memorial Hospital (CMRP-G8C0361).

Acknowledgments

The authors gratefully acknowledge the support by research grants from the Kaohsiung Chang Gung Memorial Hospital.

References

  1. Brent Asplin M, et al. Emergency Department Crowding: High-Impact Solutions. American College of Emergency Physicians; 2008.
  2. Andrulis DP, et al. Emergency departments and crowding in United States teaching hospitals. Ann Emerg Med 1991;20(9):980-6.
  3. Trzeciak S, Rivers EP. Emergency department overcrowding in the United States: an emerging threat to patient safety and public health. Emerg Med J 2003;20(5):402-5.
  4. McCusker J, et al. Increases in emergency department occupancy are associated with adverse 30-day outcomes. Acad Emerg Med 2014;21(10):1092-100.
  5. Richardson DB. Increase in patient mortality at 10 days associated with emergency department overcrowding. Med J Aust 2006;184(5):213-6.
  6. Jo S, et al. ED crowding is associated with inpatient mortality among critically ill pa- tients admitted via the ED: post hoc analysis from a retrospective study. Am J Emerg Med 2015;33(12):1725-31.
  7. Pines JM, et al. The association between emergency department crowding and ad- verse cardiovascular outcomes in patients with chest pain. Acad Emerg Med 2009; 16(7):617-25.
  8. Edmondson D, et al. The association of emergency department crowding during treatment for acute coronary syndrome with subsequent posttraumatic stress disor- der symptoms. JAMA Intern Med 2013;173(6):472-4.
  9. Sun BC, et al. Effect of emergency department crowding on outcomes of admitted patients. Ann Emerg Med 2013;61(6):605-611.e6.
  10. Pines JM, et al. The effect of emergency department crowding on patient satisfaction for admitted patients. Acad Emerg Med 2008;15(9):825-31.
  11. Chang BP, et al. Patient treatment in ED hallways and patient perception of clinician- patient communication. Am J Emerg Med 2016;34(6):1163-4.
  12. Michelson KA, et al. Pediatric emergency department crowding is associated with a lower likelihood of hospital admission. Acad Emerg Med 2012;19(7):816-20.
  13. Muller MP, et al. Hand hygiene compliance in an emergency department: the effect of crowding. Acad Emerg Med 2015;22(10):1218-21.
  14. Ben-Yakov M, et al. The association between emergency department crowding and the disposition of patients with transient ischemic attack or minor stroke. Acad Emerg Med 2015;22(10):1145-54.
  15. Depinet HE, et al. The effect of emergency department crowding on reassessment of children with critically Abnormal vital signs. Acad Emerg Med 2014;21(10): 1116-20.
  16. Kawano T, Nishiyama K, Hayashi H. Execution of diagnostic testing has a stronger ef- fect on emergency department crowding than other common factors: a cross-sec- tional study. PLoS One 2014;9(10):e108447.
  17. Wu KH, et al. The influence of physician seniority on disparities of admit/discharge decision making for ED patients. Am J Emerg Med 2012;30(8):1555-60.
  18. Verelst S, et al. Emergency department crowding in relation to in-hospital adverse medical events: a large prospective observational cohort study. J Emerg Med 2015;49(6):949-61.
  19. Ng CJ, et al. Validation of the Taiwan triage and acuity scale: a new computerised

    five-level triage system. Emerg Med J 2011;28(12):1026-31.

    Hall MK, et al. Impact of Point-of-care ultrasonography on ED time to disposition for patients with nontraumatic shock. Am J Emerg Med 2016;34(6):1022-30.

  20. Sikka R, et al. ED crowding is associated with an increased time to pneumonia treat- ment. Am J Emerg Med 2010;28(7):809-12.
  21. Hong KJ, et al. Association between ED crowding and delay in resuscitation effort. Am J Emerg Med 2013;31(3):509-15.
  22. McAlister FA, et al. Does reducing inPatient length of stay have upstream effects on the emergency room: exploring the impact of the general internal medicine care transformation initiative. Acad Emerg Med 2016;23(6):711-7.
  23. Lucas R, et al. Emergency department patient flow: the influence of hospital census variables on emergency department length of stay. Acad Emerg Med 2009;16(7): 597-602.
  24. Krochmal P, Riley TA. Increased health care costs associated with ED overcrowding. Am J Emerg Med 1994;12(3):265-6.
  25. Mills AM, et al. The association between emergency department crowding and anal- gesia administration in acute abdominal pain patients. Acad Emerg Med 2009; 16(7):603-8.

Leave a Reply

Your email address will not be published. Required fields are marked *