Refusing Care to Emergency Department Patients: Evaluation of Published Triage Guidelines,☆☆,,★★

Presented at the Society for Academic Emergency Medicine Annual Meeting in Toronto, Ontario, Canada, May 1992.
https://doi.org/10.1016/S0196-0644(94)70042-7Get rights and content

Abstract

Study objective: To determine whether a set of published triage guidelines identifies patients who can safely be refused emergency department care. Design: Historical cohort study. Setting: A public hospital ED. Type of participants: All patients triaged during a one-week period who were not in the most acute triage category. Measurements: Two ED nurses, blinded to the study hypothesis, reviewed each triage sheet to determine whether the case met the published guidelines for refusing care. In addition, each ED record was reviewed for appropriateness; a visit was considered appropriate only if predetermined, explicit criteria were met and an emergency physician agreed that a 24-hour delay in care might have worsened the patient's outcome. Main results: Of the 106 patients who would have been refused care according to the triage guidelines, 35 (33%) had appropriate visits. Four were hospitalized. Conclusion: When tested in our patient population, the triage guidelines were not sufficiently sensitive to identify patients who needed ED care. Broad application of these guidelines may jeopardize the health of some patients. [Lowe RA, Bindman AB, UIrich SK, Norman G, Scaletta TA, Keane D, Washington D, Grumbach K: Refusing care to emergency department patients: Evaluation of published triage guidelines. Ann Emerg Med February 1994;23:286-293.]

Section snippets

INTRODUCTION

Emergency department overcrowding has become a national concern. Excess patient volume in urban EDs prolongs waiting times, delays treatment of severely ill patients, and requires diversion of ambulances to more distant hospitals.1, 2 ED directors at 65% of surveyed public and teaching hospitals reported that overcrowding had a negative impact on the quality of care they provided.1

ED overcrowding may stem from many causes: overburdened inpatient facilities, causing a backup of admitted patients

MATERIALS AND METHODS

This study used a historical cohort design, with data collected for a previous study of ED waiting times.26 The setting was San Francisco General Hospital (SFGH), a public hospital ED with approximately 78,000 patient visits annually. Because SFGH does not refuse ED care, information was available on the ED management of all patients. For each patient, we determined whether the patient would have been provided care or refused care if the published UCD triage guidelines had been in use, and

RESULTS

When the two nurse raters reviewed the 496 charts independently they initially disagreed as to whether the patient met the triage guidelines in 61 (12%) of cases. After discussion, the raters were able to achieve consensus on 487 (98%) of the patients.

The triage guidelines had limited ability to predict which ED visits would meet the explicit appropriateness criteria (Figure 4). Of the 106 patients who would have been refused care by the triage guidelines, 43 (41%; 95% confidence interval [CI],

DISCUSSION

This study calls into question the safety of applying the suggested triage guidelines. The published guidelines were somewhat difficult to apply in that two nurse raters disagreed on 12% of cases. Of greater concern, the triage guidelines lacked sensitivity. Forty-one percent of the patients meeting the triage guidelines for refusal of care received therapeutic or diagnostic interventions in the ED, suggesting that they needed prompt care.

There are several possible explanations for the

CONCLUSION

When tested in our patient population, the proposed triage guidelines were difficult to apply because of their ambiguities and were not sufficiently sensitive to allow their safe use. The limitations of our retrospective study suggest the need for caution in interpreting the results. Nevertheless, it would seem prudent to defer use of the triage guidelines unless they can be validated prospectively, in several EDs, with careful patient follow-up.

While the authors take responsibility for their

References (38)

Cited by (119)

  • On easy solutions

    2011, Annals of Emergency Medicine
  • ED patients: How nonurgent are they? Systematic review of the emergency medicine literature

    2011, American Journal of Emergency Medicine
    Citation Excerpt :

    The categorization was performed by 2 sorts of health professionals: a triage nurse or a physician. Among the 17 categorizations performed in the triage area, 15 were done by triage nurses (88.2%) [9-11,18,19,26-28,32,39,43,48,51,55] and 2 by ED physicians (11.8%) [28,49]. The 34 categorizations performed after the examinations were all done by ED physicians.

View all citing articles on Scopus

From the Prevention Sciences Group,* Division of Emergency Medicine, and Institute for Health Policy Studies, University of California, San Francisco and Division of General Internal Medicine,§ Department of Emergency Medicine, and Department of Family and Community Medicine, San Francisco General Hospital, San Francisco.

☆☆

This work was supported in part by the MEDTEP Research Center on Minority Populations (Agency for Health Care Policy and Research Grant #HS07373).

Address for reprints: Robert A Lowe. MD, MPH, FACEP, FACP, Center for Clinical Epidemiological and Biostatistics, University of Pennsylvania School of Medicine, 2L NEB/6095, 420 Service Drive, Philadelphia, Pennsylvania 19104-6095, 215-898-6726, Fax 215-573-5315

★★

Reprint no. 47/1/52553

View full text