Effect of vital signs on triage decisions

Abstract

Study Objective: We sought to determine whether knowledge of vital signs changes nurse triage designations (TDs). We also sought to determine whether patient age and ability to communicate modify the effect of vital signs on triage decisions. Methods: We performed a prospective observational study, in 24 emergency departments, of nurse-assigned TDs of all ED patients undergoing triage. Nurses performed their typical triage routine, except that they chose 1 of 5 hypothetical TDs (call 911, ED <2 hours, physician's office 2 to 8 hours, physician's office 8 to 24 hours, or home care) before and after measurement of vital signs. The main outcome measure was the change of TD after knowledge of a patient's vital signs, with stratification on the basis of patient age and communication barriers. The secondary outcome was the final ED disposition. Results: Six hundred twenty-five experienced triage nurses at 24 different EDs collected data on 14,285 patients. TDs were downgraded (decreased in urgency) in 2.4% of patients, and 5.5% were upgraded (increased in urgency) after vital signs were known. Changes were more likely to occur in the young (≤2 years old; 11.4%) and the elderly (≥75 years old; 9.9%) than in those 3 to 74 years of age (7.5%). When nurses reported a communication barrier, a change in post-vital signs TD was also more common (11.2% versus 7.7%). The post-vital signs TD better predicted patient ED disposition. Conclusion: In this sample, 92.1% of the nurses' TDs were not affected by the knowledge of patient vital signs. For the other 7.9%, including many patients from vulnerable populations, the vital signs changed the nurses' assessments of the patients' triage designation. Methods of triage that do not determine vital signs may not adequately reflect the urgency of the patient's presentation. [Cooper RJ, Schriger DL, Flaherty HL, Lin EJ, Hubbell KA. Effect of vital signs on triage decisions. Ann Emerg Med. March 2002;39:223-232.]

Introduction

Most health care visits begin with medical data acquisition by someone other than the principal practitioner. This initial pre-encounter typically garners some combination of chief complaint, recent history, vital signs, and limited physical examination. These elements are collected to expedite the subsequent provider encounter, archive vital statistics (eg, blood pressure, weight, tobacco use), and assign a severity rating so that the order (and in some cases the setting) in which patients are seen can be determined (triage). There are hundreds of millions of such pre-encounters annually in the United States, but less than 15,000 of them have been formally studied.1, 9 We know very little about the importance of each data element to the process. We know more about the importance of checking smoking status than we do about the importance of obtaining vital signs.10, 11

In this article, we examine the use of the pre-encounter for the purpose of triage. The few existing studies demonstrate that in-person triage is highly subjective; has low interrater reliability among nurse, physician, and computer-aided methods; and is not particularly adept at identifying those who will be admitted to the hospital.1, 5 Because some popular triage instruments do not uniformly use vital signs and because telephone triage cannot use vital signs, we sought to examine how vital signs inform the triage decision.

Our objectives were to determine whether knowledge of vital signs changes nurse triage designations (TDs) and whether patient age and ability to communicate modify the effect of vital signs on triage decisions. We had several a priori hypotheses. First, we believed that knowledge of vital signs would lead to more changes in TD in the very young (≤2 years old) and the elderly (≥75 years old) than in those 3 to 74 years of age. Second, we believed that vital signs would be more likely to inform triage decisions (and thus produce greater changes in TD) in patients who had difficulty communicating because of a language barrier, thought disorder, or neurologic condition. Third, we believed that TDs made after obtaining vital signs would better predict patient disposition than designations made before obtaining vital signs.