Currently Viewing:
The American Journal of Managed Care November 2014
The Correlation of Family Physician Work With Submitted Codes and Fees
Richard Young, MD, and Tiffany L. Overton, MPH
Population Targeting and Durability of Multimorbidity Collaborative Care Management
Elizabeth H.B. Lin, MD, MPH; Michael Von Korff, ScD; Do Peterson, MS; Evette J. Ludman, PhD; Paul Ciechanowski, MD, MPH; and Wayne Katon, MD
Have Prescription Drug Brand Names Become Generic?
Alfred B. Engelberg, JD
Will Medicare Advantage Payment Reforms Impact Plan Rebates and Enrollment?
Lauren Hersch Nicholas, PhD, MPP
Variation in Hospital Inpatient Prices Across Small Geographic Areas
Jared Lane K. Maeda, PhD, MPH; Rachel Mosher Henke, PhD; William D. Marder, PhD; Zeynal Karaca, PhD; Bernard S. Friedman, PhD; and Herbert S. Wong, PhD
Medical Cost Burdens Among Nonelderly Adults With Asthma
Emily Carrier, MD, and Peter Cunningham, PhD
The Role of Retail Pharmacies in CVD Prevention After the Release of the ATP IV Guidelines
William H. Shrank, MD, MSHS; Andrew Sussman, MD; and Troyen A. Brennan, MD, JD
Care Coordination Measures of a Family Medicine Residency as a Model for Hospital Readmission Reduction
Wayne A. Mathews, MS, PA-C
Medication Adherence and Readmission After Myocardial Infarction in the Medicare Population
Yuting Zhang, PhD; Cameron M. Kaplan, PhD; Seo Hyon Baik, PhD; Chung-Chou H. Chang, PhD; and Judith R. Lave, PhD
Currently Reading
Reasons for Emergency Department Use: Do Frequent Users Differ?
Kelly M. Doran, MD, MHS; Ashley C. Colucci, BS; Stephen P. Wall, MD, MS, MAEd; Nick D. Williams, MA, PhD; Robert A. Hessler, MD, PhD; Lewis R. Goldfrank, MD; and Maria C. Raven, MD, MPH
Service Setting Impact on Costs for Bevacizumab-Treated Oncology Patients
Nicole M. Engel-Nitz, PhD; Elaine B. Yu, PharmD, MS; Laura K. Becker, MS; and Art Small, MD
Influence of Hospital and Nursing Home Quality on Hospital Readmissions
Kali S. Thomas, PhD; Momotazur Rahman, PhD; Vincent Mor, PhD; and Orna Intrator, PhD

Reasons for Emergency Department Use: Do Frequent Users Differ?

Kelly M. Doran, MD, MHS; Ashley C. Colucci, BS; Stephen P. Wall, MD, MS, MAEd; Nick D. Williams, MA, PhD; Robert A. Hessler, MD, PhD; Lewis R. Goldfrank, MD; and Maria C. Raven, MD, MPH
Frequent emergency department (ED) users gave similar reasons for using the ED rather than a clinic compared to other patients, including concerns around convenience, access, and quality.
To examine patients’ reasons for using the emergency department (ED) for low-acuity health complaints, and determine whether reasons differed for frequent ED users versus nonfrequent ED users.

Study Design
Prospective cross-sectional survey.

Patients presenting to an urban public hospital for low-acuity health complaints were surveyed about their reasons for visiting the ED rather than a private doctor’s office or clinic. Patients with 3 or more visits to the study hospital ED over the past year were classified as frequent ED users. Multivariable logistic regression was used to determine if frequent ED users gave different reasons for ED use than nonfrequent ED users, while controlling for differences in other baseline patient characteristics.

940 patients, including 163 frequent ED users, completed the study questionnaire. Commonly cited reasons for using the ED were that coming to the ED was easier than making a clinic appointment (82.3% agreed); the problem could not wait (78.8%); they didn’t know how to make a clinic appointment (66.7%); they felt the ED provided better care (56.7%); and they believed the clinic would cost more (54.8%). After controlling for other patient characteristics, there were no significant differences found in reasons for ED use given by frequent versus nonfrequent ED users.

Frequent ED users gave similar reasons for using the ED for low-acuity health complaints compared with nonfrequent ED users. Access, convenience, cost, and quality concerns, as well as feeling that ED care was needed, were all commonly cited as reasons for using the ED.

Am J Manag Care. 2014;20(11):e506-e514
Frequent emergency department (ED) users are of concern to insurers, healthcare systems, and policy makers because they account for a disproportionately large share of ED visits and costs. We studied patients’ reasons for using the ED for lowacuity health complaints, comparing frequent versus nonfrequent ED users.
  • Frequent and nonfrequent ED users gave similar reasons for using the ED.
  • Common reasons for ED use included concerns about convenience, timely access, quality, and costs.
  • Health systems must examine whether reasonable alternative options to the ED exist and are responsive to patients’ needs and desires, rather than penalizing so-called “inappropriate” ED use.
The number of emergency department (ED) visits in the United States has grown dramatically over the past 15 years, from 93.1 million annual visits in 1996 to 129.5 million visits in 2011.1 Rather than being equally distributed across the population, approximately 5% of patients who use the ED are responsible for 25% of all ED visits.2 Termed frequent users, these patients are of interest to policy makers due to the high healthcare costs they incur both in the ED and in other parts of the healthcare system.

Prior studies have examined basic characteristics of frequent ED users, using definitions of frequent use varying from 2 to 20 or more ED visits per year.3 Research has shown that, in contrast to common assumptions, frequent ED users are sicker than the general population and often have multiple health problems.2,4,5 Furthermore, they tend to have higher ED triage acuities and hospital admission rates than other patients.2,6,7 Most frequent users are insured, and they utilize outpatient clinics in addition to the ED.2,8-15 Studies have identified many correlates of frequent ED use including specific medical conditions, mental illness, and type of insurance, among others.2,3,16

Amid the literature on frequent ED users, however, there has been almost no prior research exploring reasons that frequent ED users give for using the ED. One qualitative study of a small group of frequent ED users identified themes such as lack of timely primary care provider availability and feeling that ED services were needed.17 Multiple past studies have examined reasons for ED use among patients more generally, regardless of level of prior ED use. These studies have paid particular attention to patients deemed to have visited the ED for so-called “nonurgent” or “inappropriate” reasons,18-23 even though no consensus exists on how to define these visits and such categorizations are fraught with difficulty.24-26

Some of the more common reasons found in prior research for why patients use the ED were believing their problem required ED care,19,22,23,27-31 lacking accessible primary care alternatives,19,21,22,32 not being able to make expedient outpatient appointments,19,23,27,33-35 and trusting the ED more than other sites.22,31,36 Studies have not explored differences in reasons for ED use given by frequent versus nonfrequent ED users, leaving a significant gap in the existing literature and providing no clear direction for policy makers. In the current study, we used a cross-sectional survey of low-acuity patients presenting to an urban public hospital ED to determine patients’ reasons for ED use and to examine whether frequent ED users report different reasons compared to nonfrequent users.


Study Design and Setting

This study uses results from a cross-sectional survey conducted as part of a larger trial.37 In brief, the trial examined the impact of an intervention offering ED patients with low-acuity complaints the opportunity to forgo their ED visit and instead receive same-day care in a primary care clinic in the same hospital building as the ED. The study was conducted at Bellevue Hospital Center, a public safety net hospital in New York City, whose ED has 100,000 patient visits yearly. The study was approved by the institutional review boards at New York University School of Medicine and Bellevue Hospital Center.

Patient Enrollment and Classification

Patients were considered for inclusion if they presented to the ED with selected presenting complaints that a layperson would be expected to recognize as low-acuity. These were: sore throat; medication refill; nontraumatic joint or back pain; symptoms of a simple urinary tract infection; symptoms of a cold or upper respiratory infection; or a stable chronic medical illness of headache, hypertension, or diabetes. All patients were 23 years or older, as younger patients are served by a pediatric ED.

Patients arriving by ambulance were excluded, as were patients with a temperature greater than 38.3°C. Patients were also excluded if the triage nurse felt the individual required ED care (for example, if the nurse felt the patient had complex coexisting conditions that made transfer to the clinic unsafe or was otherwise not clinically stable for transfer to the clinic). Though, in practice, the triage nurses rarely overruled study enrollment, study investigators felt that including this provision of triage nurse discretion was important in providing an additional layer of safety to the intervention. Enrollment was conducted from January 2007 to January 2008 on weekdays from 9 am to 3 pm when the hospital’s primary care clinic was open and available for patients. Patients could only be enrolled in the study once.

Patients were enrolled regardless of previous ED utilization and were then divided into “frequent users” and “nonfrequent users” for the analysis using ED visit data obtained from hospital administrative databases. We categorized patients as frequent users if they had made ≥3 visits to the study hospital ED over the prior year (including the study enrollment visit). No agreed-upon standard for how many prior visits defines “frequent” ED use exists in the literature or practice; we chose the 3 visits per year because this threshold has been commonly used in previous studies.2 Though all baseline visits were for low-acuity health complaints, prior ED visits could have been for any reason.

Data Collection and Measures

Research associates administered a baseline questionnaire to all patients. This questionnaire permitted the collection of demographic and health information (eAppendix A, available at and included a series of 10 questions about patients’ reasons for visiting the ED that day rather than a private doctor’s office or clinic (Figure). Questions were developed via consensus by a group of emergency and primary care clinicians based on review of prior literature and observations from clinical practice. The questionnaire was pilot-tested in the ED prior to beginning the study to ensure question clarity and patient comprehension. Each question gave a reason why one might use the ED rather than a clinic and asked patients whether they agreed or disagreed that this reason influenced their own decision to use the ED that day. Questionnaires were administered verbally. Dual-headset translation phone systems were used for non-English speaking patients. Written informed consent was

obtained from all patients.

Data Analysis

Simple means and proportions are presented for baseline characteristics of frequent and nonfrequent ED users. Bivariate relationships between frequent use and reason for ED use were examined using χ2 tests. Multivariable logistic regression was used to examine the independent effect of being a frequent ED user on likelihood of agreeing with each reason for ED use while controlling for other differences in baseline patient characteristics. Control variables included age, sex, race/ethnicity, language, education, employment, insurance, chief complaint, overall health, whether the patient reported having a personal doctor, and prior use of the hospital’s primary care clinic. These control variables were chosen because prior studies have found these factors to be related to the likelihood of using the ED for care and of being a frequent ED user.2,21 All control variables were based on self-report except for prior primary care clinic use, which was ascertained from hospital administrative databases.

Exploratory factor and principal components analysis found no condensed set of factors that adequately captured the correlation and variance among individual reasons for ED use (full correlations are shown in eAppendix B). Therefore, 10 separate multivariable logistic regressions were conducted, each using a different reason-for-ED-use question as the dependent variable. All control variables were retained in the multiple logistic regression models based on theory and to avoid omitted variable bias.38,39 Model goodness of fit was confirmed with Hosmer-Lemeshow tests. All analyses were conducted using SAS 9.2 (Cary, North Carolina).


Of the 1404 patients eligible for the study, 439 declined to participate and 965 agreed and were enrolled. Twenty-five patients were excluded from the analysis because they had not completed any of the questions regarding reasons for ED use. Thus 940 patients were included in the current analysis, including 163 who had made ≥3 ED visits over the prior year and were categorized as frequent users.

The mean age was 49 years for frequent users and 46 years for nonfrequent users (Table 1). Approximately 60% of patients in each group were male; most patients were Latino or black. More frequent than nonfrequent users identified English as their primary language (59.5% vs 41.8%). Frequent users were more often uninsured (73% vs 45.8%), yet were more likely to have a personal doctor (50.9% vs 32.4%) and to have used the hospital’s primary care clinic in the past year (52.8% vs 17%).

Overall, the most commonly endorsed reason for using the ED was that coming to the ED was easier than making a clinic appointment, with 82.3% of patients agreeing with this statement (Table 2). Patients also frequently agreed that they would have gone to a clinic but their problem could not wait until tomorrow or the next day (78.8%) and that their problem could have been treated in a clinic but they did not know how to make an appointment (66.7%). More than half also agreed that patients would get better care in the ED (56.7%), a clinic would want them to pay today (54.8%), and a clinic visit would cost more (53.1%). Somewhat fewer patients reported that a clinic could not treat their problem, that they needed tests that a clinic could not do, or that they needed to be admitted to the hospital. Frequent and nonfrequent ED users gave similar reasons for using the ED (Table 2). The only significant differences observed in bivariate analyses were that frequent users were less likely to think they needed tests that a clinic could not do (37.9% vs 50.1%), less likely to feel that their problem could have been treated in a clinic but that they did not know how to make an appointment (57.8% vs 68.6%), and less likely to say that a clinic would want them to pay today (46.3% vs 56.7%). After adjusting for differences in baseline patient characteristics, no significant differences were observed between frequent and nonfrequent ED users in reasons for using the ED (Table 3).


Copyright AJMC 2006-2020 Clinical Care Targeted Communications Group, LLC. All Rights Reserved.
Welcome the the new and improved, the premier managed market network. Tell us about yourself so that we can serve you better.
Sign Up