Article, Emergency Medicine

Use of an emergency department by nonurgent patients

Use of an emergency department by nonurgent patientsB

William E. Northington, MDa,*, Jane H. Brice, MD, MPHb, Bin Zou, MD, MPHb

aUniversity of Pittsburgh Affiliated Residency in Emergency Medicine, Pittsburgh, PA 15213, USA

bDepartment of Emergency Medicine, University of North Carolina Hospitals, Chapel Hill, NC 27599, USA

Received 7 August 2003; revised 28 April 2004; accepted 10 May 2004

Presented at the American College of Emergency Physicians Research Forum, Philadelphia, Pa, October 2000; the Third Annual Southeastern Regional Society for Academic Emergency Medicine Conference, New Orleans, La, March 2000; the Fifth Annual University of North Carolina Emergency Medicine Research Forum, Chapel Hill, NC, April 2000; and the University of North Carolina John B. Graham Student Research Day, Chapel Hill, NC, January 2001.


Objectives: The objectives of this study are (1) to determine whether patients seeking emergency department (ED) nonurgent care have primary care providers (PCP) or know of other care sources and

  1. to determine the reasons why they choose to use the ED.

Methods: A cross-sectional survey in a university ED was administered to self-referred nonurgent patients for 6 weeks. Use of a PCP, knowledge and attempts to seek other care, past use of the ED, urgency self-report, time of visit, and reasons for choosing an ED were recorded.

Results: Of the 563 approached subjects, 314 were eligible and 279 agreed to participate. One hundred

fifty-seven (56%) had PCPs. For 183 (66%) subjects, the ED was the only place they knew to go for their present problem, and 75 (27%) reported that they depended on the ED for all medical care. Of those patients with a PCP, 73 (47%) rated the ED better for unscheduled care. Eighty-one (52%) subjects thought their PCP would be more efficient and 66 (42%) thought their PCP would be cheaper.

Conclusions: Although most ED nonurgent patients were not dependent upon the ED, the majority was

unaware of other places to go for their current health problem. Even those patients with a PCP sought care in the ED because the ED was believed to provide better care despite its perceived increase in both waiting time and cost.

D 2005

B This study was supported by the National Institutes of Health student summer research fellowship and an unrestricted educational grant from the Aspect Medical Systems.

* Corresponding author. Tel.: +1 412 401 0178; fax: +1 412 647 4670.

E-mail address: [email protected] (W.E. Northington).


There has been a steady increase in the use of emergency departments (ED) since the mid 1940s [1]. Some might dismiss this fact as obvious, citing the increase in visits as a result of overall population growth. However, these increases in ED use have been much more drastic than the increase in population [1-3]. Currently, it is estimated that

0735-6757/$ – see front matter D 2005 doi:10.1016/j.ajem.2004.05.006

patient visits to an ED for nonurgent complaints are between 33% and 50% [1,2,4-6]. The increased patient load may cause emergency providers to allot less time to care for each patient and to decrease time for health teaching, counseling, and helping patients with chronic health problems to find appropriate resources to help manage their Disease states [1]. Given that previous reports found that patients presenting to the ED report longer waits, more costly care, and poor follow-up care [6], the question remains why patients decide to use the ED for nonurgent complaints. Furthermore, as EDs around the country face bovercrowdingQ issues, what should be done with this increasing nonurgent population? The purpose of this study is to further investigate the use of the ED by self-referred nonurgent patients (defined as those low-acuity patients that meet Emergency Severity Index [ESI] triage categories of 4 or 5), including (1) whether nonurgent patients presenting to an ED have primary care providers (PCP), (2) what reasons factor into the decision to come to an ED for a nonurgent complaint, (3) if nonurgent patients know of other sources of care for their complaints,

  1. if there is a difference in regards to the previous 3 aspects when comparing the patient population that presents between 9:00 am and 5:00 pm (normal business hours) to those patients who present between 5:00 pm and 1:00 am, and
  2. if these nonurgent patients have insurance.


Study design and population sample

Using a cross-sectional study design, a convenient sample of patients presenting to the University of North Carolina ED who were triaged as nonurgent by trained nursing staff were interviewed by trained research assistants. Patients triaged to the acute ED and minor trauma areas as nonurgent between 9:00 am and 1:00 am, the hours for which research assistants were available, during the study period of June 23, 1999, to August 8, 1999, were invited to participate.

Exclusion criteria included those persons who were intox- icated, pregnant, mentally impaired, younger than 18 years, a contract case (those cases that the ED had a contractual obligation to see, such as prisoners or hospital employees), suspected victim of abuse, or those referred by a physician.


University of North Carolina Hospital is a tertiary care, level 1 trauma center with an annual patient census of 65,000 triaged patients that serves a suburban/rural popu- lation in central North Carolina. The ED is organized into 4 separate areas: (1) acute ED, (2) pediatric ED, (3) minor trauma, and (4) urgent care. The acute ED is the only area open 24 hours, minor trauma is open 12:00 pm to 12:00 am, the pediatric ED 9:00 am to 11:00 pm, and urgent care 9:00 am to 9:00 pm. Any patients who would be triaged to minor trauma, urgent care, or pediatric ED are sent to the acute ED when one of these areas is closed. Board- certified emergency medicine physicians staff the acute ED and minor trauma. The pediatric ED only sees patients 16 years or younger and is staffed by pediatricians. Urgent care is staffed by internal medicine physicians and serves mainly as a walk-in clinic for minor medical complaints. Internal medicine physicians often refer their patients to urgent care when the patient has an acute medical problem and cannot be seen by the physician. Patients who were triaged to the urgent care area were not included in the study population because most of these patients are referred to this area by their PCP or medicine clinic. All patients presenting to the ED are triaged by a trained member of the nursing staff to the appropriate area of care. This study received an expedited approval from the University of North Carolina Committee for the Protection of Human Subjects in Research.

Triage Scale

Patient urgency was assessed using the ESI triage algorithm, a validated scale on which the nursing staff

Fig. 1 Questions on the patients’ completed questionnaire.

Fig. 2 Interview questions asked by the trained research assistant.

has received extensive on-going education and evaluation [7-12]. Those patients assigned a triage ranking of 4 or 5 were considered nonurgent. The 4 and 5 ratings are given to those patients who are triaged as Very low risk. These patients are responsive, oriented, in no acute distress, have stable vital signs, and estimated to require no more than one resource (lab, test, or consult).

Survey tool

The survey tool was a 2-part instrument. The principal investigator piloted the tool with actual patients in the ED on 3 separate occasions before the start of the study to refine the survey instrument. The first part of the instrument was a brief patient-completed questionnaire, shown in Fig. 1, which assessed the demographics as well as the degree of

Fig. 3 Parameters of each visit recorded by the research assistant.

tigator. A biostatistical consultant using SAS Research Data Management (SAS Institute, Cary, NC) performed the statistical analysis. Categorical data were analyzed using v2 and continuous data with a t test.



Demographics of

patients triaged 4

or 5

All patients triaged as nonurgent in the acute ED and minor trauma

Excluded patients

Included patients


36.1 years

36.4 years

37.4 years

(SD F 17.6)

(SD F 17.9)

(SD F 14.9)


53.3% male

55.0% male

55.2% male

58.6% Caucasian

58.8% Caucasian


34.1% African

Not recorded

37.3% African



7.3% Other

3.9% Other

worry and urgency (as rated by the subject) that led the patient to come to the ED. A brief interview conducted by trained research assistants was then administered to assess the patient’s reasons for choosing to seek care in an ED by asking the questions listed in Fig. 2. The research assistant also recorded the specific parameters of each patient visit as detailed in Fig. 3.

Table 3 survey results

Where do you usually go for care?

Primary care physician 43.7%

Only ED 27.1%

Clinic 16.2%

Other 11.2%

Health department 1.8%

Other than an ED, do you know of any other places you could have gone for this problem today?

Nowhere else 65.6%

Clinic 17.2%

Other 6.5%

Primary care physician 5.4%

Different ED 3.9%

Health department 1.4%

Did you seek any other sources of care before coming to an ED today?

No 74.9%

Other 14.3%

Clinic 6.1%

Primary care physician 3.6%

Health department 0.4%

Different ED 0.4%

On a scale of 1 to 10, how urgent is your condition today?

Median 5.0 Mean 5.6

On a scale of 1 to 10, how worried are you about your condition today?

Median 5.5

Mean 5.7

Why did you choose the ED as your source of care today?

Better care 76.1%

Urgency 73.6%

Immediacy 68.6%

Payment flexibility 41.9%

Expediency 39.7%

Research assistant training

Six research assistants were trained during a 1-hour session to conduct the survey. The principal investigator observed each research assistant during 2 actual patient encounters. Both the research assistant and principal investigator scored the encounter separately. Later, the 2 encounters were compared using the principal investigator’s assessment as the gold standard. In all cases, there was 100% concordance between the research assistant and the principal investigator.

Statistical analysis

All the data were coded and entered into Microsoft Excel (Microsoft, Seattle, Wash) by the principal inves-


During the study period, 563 patients were evaluated for exclusion criteria. Of these, 284 were excluded for the following criteria: 50 were younger than 18 years, 4 were pregnant, 28 were intoxicated, 46 were mentally impaired, 26 were non-English speaking, 7 were suspected victims of abuse, 85 were referred to the ED by their physician, 35 refused to participate, and 3 left before they had been

Table 2 Expanded demographics of included patients

Primary care physician 56.3% Yes

Chief complaint 36.2% Orthopedic (those reported over 5%) 10.0% Neck/back pain

9.7% Laceration

8.2% Minor medical 7.5% Motor vehicle crash 7.5% Infection

Insurance 35.8% Private

31.5% Uninsured

9.7% Medicaid

6.1% Medicare

5.0% HMO

11.8% Other

Wait time for physician 66.2 min (SD F 49.1) Range, 1-330 min

Total in time in ED 175.6 min (SD F 106.0) Range, 33-877 min

uled care, whereas other sources were better for overall waiting time and overall cost of the visit.

Table 4 How does the ED compare to your other source of care in terms of the following?

ED better (%)

Other source better (%)

No difference (%)

Do not know/have no other

source (%)

Waiting time





Quality of care
















Ease of







discharged. There was no significant difference between excluded and included patients in regards to age or sex.

Patient demographics

Two hundred seventy-nine patients completed the survey and interview. Patient demographics are recorded in Table 1. No significant difference existed between the included patients and the overall population of the ED who received a triage rating of 4 or 5 in regards to age, sex, or race. Expanded demographics for patients included in the study are listed in Table 2.

Survey results

The survey results, shown in Tables 3 and 4, reveal that the majority (56.3%) of the questioned population did indeed have a PCP. However, of this majority, only 43.7% received their usual care from a PCP, and 27% were reliant on an ED for all their medical needs.

Furthermore, 74.9% of patients came directly to the ED for care, with 65.6% knowing of no where else that they could seek appropriate care for their complaint at that given time. When queried as to why they decided to seek care in an ED, 76.1% believed that they would receive better care and 73.6% cited that their complaint was urgent.

Most of the population surveyed (68.5%) had some form of insurance, with most (35.8%) possessing private insur- ance. When comparing the population that came to the ED from 9:00 am to 5:00 pm (normal physician office hours) to the group that entered from 5:00 pm to 1:00 am, there was no significant difference in regards to (1) having a primary care physician, (2) whether they sought care/advice before proceeding to an ED, (3) reason for seeking care in an ED, or (4) insurance status.

When asked how worried and how urgent they felt their complaint is, the means were 5.7 and 5.6 on a 10-point scale. Finally, when asked to compare the ED with other sources of care they had encountered in the past, this population felt that an ED is better for obtaining unsched-


Do nonurgent patients presenting to the ED have PCPs?

This study found that most nonurgent patients who use the ED for care do, in fact, have a PCP. However, if most of these patients have a PCP, the question arises as to why they decide to seek out the ED rather than their PCP. It is understandable why those who admitted that the ED is their only source of care came to the ED; it serves the purpose of a PCP. Yet, a large proportion of the patients who stated that they would normally seek out their PCP for medical problems, on these occasions, have ended up in an ED. Other studies have reported similar findings, noting that roughly half (48%-61%) of the nonurgent patients present- ing to an ED have a PCP [13,14]. However, reports looking at the ED population as a whole (both urgent and nonurgent) have varied results ranging from 39% to 69% of patients having a PCP [5,13,14].

What reasons factor into the decision to come to an ED for a nonurgent complaint?

When prompted as to why they chose to come to an ED for care, the 3 most common reasons were (1) belief that they could receive better care at an ED, (2) urgency of their complaint, and (3) immediacy. Patients citing better care included those who did not believe that other sources of care would have the resources needed to properly care for their complaint (eg, a patient who fell and hurt his arm and feared that he broke it, but knew that his PCP did not have an x-ray machine at his office comes because the ED has the radiographic capabilities). Patients citing urgency, a somewhat expected response, believed that their conditions warranted prompt medical attention and feared that delaying care would lead to an adverse outcome. Related to the feeling of urgency is immediacy, the belief that the ED would see patients quickly and without an appointment.

If patients believe that an ED would provide better care, it is understandable why they would come directly to the ED, regardless of the availability of other sources of care. Similarly, patients who feel they are in need of care urgently and/or immediately may also come directly to an ED instead of other sources, because they know that an ED will have adequate sources to treat their complaint–whether it be the ability to run diagnostic tests or obtain specialty consults in a timely fashion.

Although mentioned significantly less frequently, patients also cited payment flexibility and expediency as contributing to the decision to seek care in an ED. The mention of payment flexibility reflects the population’s

knowledge that most EDs do not make patients pay at the time of care, but instead will send bills later. This allows the uninsured who may not have money to seek care from a PCP the opportunity to receive prompt medical attention without the concern of paying at the time of care, placing the ED in the position of a safety net for healthcare for the uninsured. Expediency refers to the fact that an ED may be more convenient for the patient, whether it be because the ED is open 24 hours or the patient has a hectic schedule and has problems making and keeping scheduled appointments with a PCP.

Examining other studies that have surveyed the reasons that nonurgent patients present to an ED most have found similar answers. Habenstreit [5] postulates that those patients who lack PCPs and rely exclusively on the ED for care do so because of early socialization, habit, and convenience. Although it may be true that many of these individuals do not exercise selective judgment when deciding to present to an ED with a nonurgent complaint, other studies have found that nonurgent patients cite the following reasons for their choice of care: (1) high acuity of current ailment [1,13], (2) immediate care available in an ED [1,3,4,13,15,16], (3) easy accessibility [1,3,5,16], (4) quality of care received in an ED superior to other sources [1,5,6], (5) inability to get an appointment with PCP [4,6,16], (6) unfamiliarity with other sources of care [6,16], (7) knowledge that they would not be billed at time of visit [5,16], and (8) overall convenience [1,5].

Do nonurgent patients know of other sources of care for their complaints?

Most patients in this study did not know of anywhere else to go for their problem on that occasion. However, only 27.1% were completely dependent upon the ED as their sole provider of healthcare. Thus, most patients had received care from a health provider outside an ED and may be expected to know of another source of care besides the ED, although that source may not be appropriate for their immediate healthcare needs.

It has been suggested that many nonurgent patients use an ED as a backup for care, when their PCP is unavailable [5]. But between 9:00 am and 5:00 pm, the same hours that a typical PCP’s office would be open, an ED sees its greatest number of nonurgent patients [1,17]. It is possible that patients underestimate the level of care that a PCP can provide (or similarly, they overestimate the severity of their complaint). By definition, patients who were triaged as nonurgent were nonacute cases that should have fallen within the scope of a PCP’s care [8]. However, if the PCP does not have the proper resources (radiograph, expedient laboratory results, etc) or time to see the patient, the outpatient PCP system cannot provide adequate care for nonurgent patients that seek immediate care.

Finally, nonurgent patients may seek out care in the ED because they truly believe that their case is an emergency.

Patients do not have the clinical experience to accurately assess the severity of their medical complaint; this is complicated by the fact that a problem often seems much more urgent when it is happening to one’s self. There is no way to effectively encourage this group of patients to seek care from other sources.

Is there a difference in regards to (1) having a PCP, (2) the reasons factoring into the decision to seek care in an ED for a nonurgent complaint, and

  1. knowing of other sources of care,when comparing the patient population that presents between 9:00 AM and 5:00 PM (normal business hours) to those patients who present between 5:00 PM and 1:00 AM?

This study found no differences among the population that presented between 9:00 am and 5:00 pm and that which entered the ED from 5:00 pm to 1:00 am in regards to (1) having a PCP, (2) the reasons factoring into the decision to seek care in an ED for a nonurgent complaint, and (3) knowing of other sources of care. It was originally conjectured that there would be a difference in these 2 populations in regards to having a PCP–the population that came during 9:00 am to 5:00 pm was expected to be less likely to have a PCP than the population that sought care from 5:00 pm to 1:00 am, hours when a PCP would normally be unavailable. This postulation stemmed from the assumption that those patients who had a PCP would seek care from that provider when available (9:00 am to 5:00 pm), and would come to an ED when the PCP’s office was closed and unavailable. Those patients without a PCP, on the other hand, would be just as likely to seek care from an ED at all hours. This study did not support that conjecture.

Do these nonurgent patients have insurance?

More than two thirds of the study population had some form of insurance. Previous studies examining the nonur- gent population presenting to the ED have mixed findings; some have similar findings, showing that more than 57% of the nonurgent population are insured [16,18], whereas another demonstrated that most of the nonurgent population is uninsured [4]. If most of this population is indeed insured, perhaps it is in the best interest (from a revenue standpoint) for the ED to see and treat these patients.

Impact of nonurgent patients on the ED

In much of the previous literature examining ED over- crowding, a common proposed solution has been to eliminate nonurgent patients from the ED. However, as the face of the nation’s healthcare system changes, these patients will continue to seek care in the ED until they are denied of care or until they are unwilling to endure the long waits and overcrowded conditions that currently exist. Another poten-

tial solution to overcrowding is to increase resources in the ED to accommodate this increased patient load.

Williams [14], in a study investigating the costs of visits to EDs, found that the costs of caring for nonurgent patients in the ED are relatively low–much lower than commonly believed. In addition, a recent study examining the cost effectiveness and clinical outcomes of seeing all presenting patients (not going on diversion) found that the revenue generated from the increasED patient volume was greater than the cost incurred for the additional staffing needs [19]. Furthermore, there was no significant increase in errors or adverse clinical outcomes with this increased patient volume [19]. These studies suggest that it may be in the best interest of the ED to treat these nonurgent patients.

According to the ESI triage level definitions, those patients triaged as 4 or 5 (the definition of nonurgent for this study) should have an expected time to disposition of 2 hours or less [8], which is consistent with this study. Furthermore, these nonurgent patients should fall under the scope of care offered by a midlevel provider. Thus, if physical resources (ie, appropriate patient care areas) permit, overcrowded EDs should strongly consider looking at the cost-analysis of hiring additional midlevel providers to provide healthcare for this nonurgent patient population. Most of this population is insured and has a quick turnover time in the ED.


This study had limitations. Despite attempting to encounter all nonurgent patients who presented to the ED during the study period, only a little over a third of all nonurgent patients were approached to be included. This percentage is due to the time constraints on the research assistant, the high number of nonurgent patients, and the quick turn around time on certain patients. However, the sample group that was surveyed was demographically similar to the whole population that entered the ED during the study period. Also, the ED at University of North Carolina is set up in such a way that the main triage desk serves the ED and the urgent care clinic. The urgent care clinic is run by the internal medicine service and treats minor medical problems–almost all of the patients seen there are triaged as nonurgent. However, many internal medicine physicians tell their patients that if they have problems and cannot reach them, they should go to the urgent care clinic for treatment. In this way, the urgent care clinic operates much like a walk-in medical clinic except that some nonurgent patients thinking they will be seen in the ED may be triaged to the urgent care clinic. Finally, one of the shortcomings of the triage system was that some of the minor trauma cases (particularly lacerations) that were accurately triaged as nonurgent were serious that they may have exceeded the limitations of some PCPs.


Among the study population, most nonurgent patients are insured and do not depend upon the ED for routine care. In fact, most have a PCP and would usually consult him/her for medical care. Most of these nonurgent patients, however, are unaware of other sources of adequate care for their current healthcare needs and thus seek out the ED. These patients also use the ED for their nonurgent complaints because they believe that the ED will provide better care in an immediate fashion for what they believe to be urgent complaints.


  1. Pisarcik G. Why patients use the emergency department. J Emerg Nurs 1980;6:16 – 21.
  2. Lang T, Davido A, Diakite B, et al. Non-urgent care in the hospital medical emergency department in France: how much and which health needs does it reflect? J Epidemiol Community Health 1996;50:456 – 62.
  3. Canizaro P. Management of the non-emergent patient. J Trauma 1971;11:544 – 51.
  4. Grumbach K, Keane D, Bindman A. Primary care and public emergency department overcrowding. Am J Public Health 1993;83: 372 – 8.
  5. Habenstreit B. Health care patterns of non-urgent patients in an Inner city emergency room. N Y State J Med 1986;86:517 – 21.
  6. Burnett M, Grover S. Use of the emergency department for nonurgent care during regular business hours. CMAJ 1996;154:1345 – 51.
  7. Wuerz R, Milne L, Eitel D, et al. Reliability and validity of a new five- level triage instrument. Acad Emerg Med 2000;7:236 – 42.
  8. Gilboy N, Tanabe P, Travers D, Eitel D, Wuerz R. The Emergency Severity Index implementation handbook: a five-level triage system. Des Plaines, IL7 emergency nurses Association; 2002.
  9. Eitel DR, Travers DA, Rosenau AM, et al. The Emergency Severity Index triage algorithm version 2 is reliable and valid. Acad Emerg Med 2003;10:1070 – 80.
  10. Tanabe P, Gimbel R, Yarnold PR, et al. Reliability and validity of scores on the Emergency Severity Index version 3. Acad Emerg Med 2004;11:59 – 65.
  11. Tanabe P, Gimbel R, Yarnold PR, et al. The Emergency Severity Index (version 3) 5-level triage system scores predict ED resource consumption. J Emerg Nurs 2004;30:22 – 9.
  12. Bernstein SL, Verghese V, Leung W, et al. Development and validation of a new index to measure emergency department crowding. Acad Emerg Med 2003;10:938 – 42.
  13. Haddy R, Schmaler M, Epting R. Nonemergency emergency room use in patients with and without primary care physicians. J Fam Pract 1987;24:389 – 92.
  14. Williams R. The costs of visits to emergency departments. N Engl J Med 1996;334:642 – 6.
  15. Meggs W, Czaplijski T, Benson N. Trends in emergency department utilization, 1988-1997. Acad Emerg Med 1999;6:1030 – 5.
  16. Powers M, Reichelt P, Jalowiec A. Use of the emergency department by patients with nonurgent conditions. J Emerg Nurs 1983;9:145 – 9.
  17. Lynch J, Silverman M, Delaney K. Emergency room utilization: an initial look. Conn Med 1981;45:309 – 10.
  18. Gavaler J, Van Thiel D. The non-emergency in the emergency room. J Natl Med Assoc 1980;72:33 – 6.
  19. Lev R, Simmons C, Mankin S, et al. No bypass week: when there is a will, there is a way. Ann Emerg Med 2001;38:S42 – 3 [Abstract].

Leave a Reply

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