Implementation Handbook
The Emergency Severity Index (ESI) is a five-level emergency department (ED)
triage algorithm that provides clinically relevant stratification of patients
into
five groups from 1 (most urgent) to 5 (least urgent) on the basis of acuity
and resource needs.The Agency for Healthcare Research and Quality (AHRQ) funded
initial work on the ESI.
A well-implemented ESI program will help hospital emergency departments rapidly identify patients in need of immediate attention, better identify patients who could safely and more efficiently be seen in a fast-track or urgent care center rather than the main ED, and more accurately determine thresholds for diversion of ambulance patients from the ED.
Select for information on how to request print
copies of the handbook or DVDs.
Contents
Note from the Director
Preface
Copyright Notice
1: The Evolution of Triage
2: Triage Acuity Systems
3: Introduction to the Emergency Severity Index
4: ESI Level 2
5: Expected Resource Needs
6: The Role of Vital Signs in ESI Triage
7: Implementation of ESI Triage
8: Evaluation and Quality Improvement
9: Practice Cases
10: Competency Cases
Appendixes
Appendix A: Frequently Asked Questions and Post-Test Materials
for Chapters 3-8
Appendix B: ESI Triage Algorithm, v. 4
Chapter 1. The Evolution of Triage
The purpose of emergency department (ED) triage is
to prioritize incoming patients and to identify those
patients who cannot wait to be seen. The
experienced triage nurse is able to rapidly and
accurately identify the small percentage of patients
requiring immediate care. The triage nurse is then
challenged to sort the remaining large number of
patients who do not require immediate treatment
and can wait for physician evaluation. The number
of patients presenting to emergency departments is
increasing, and this trend is not likely to change. As
EDs are struggling to cope with overcrowding there
is a critical need for a valid, reliable triage acuity
rating system in order to sort these incoming
patients more rapidly and accurately.
This chapter explores the evolution of triage in the
United States and describes the dominant triage
systems currently in use in EDs. A discussion follows
of why the acuity ratings scales currently in place in
most emergency departments are no longer
adequate to meet the needs of the 21st century in
light of recent trends in patient demographics, ED
utilization, and other factors affecting patient flow
through the ED.
Triage History
The word "triage" is derived from the French verb
"trier," to "sort" or "choose." Originally the process
was used by the military to sort soldiers wounded in
battle for the purpose of establishing treatment
priorities. Injured soldiers were sorted by severity of
their injuries ranging from those that were severely
injured and deemed not salvageable, to those who
needed immediate care, to those that could safely
wait to be treated. The overall goal of sorting was to
return as many soldiers to the battlefield as quickly
as possible.
Changes in the health care delivery system forced
U.S. emergency departments to consider alternative
ways of handling an increase in the number of
incoming patients during the 1950s and early 1960s.
In the late 1950s, physician practice began to
change. Physicians moved away from solo practice;
the days of house calls and the family doctor
became nearly obsolete. Physicians formed office-based
group practices that offered regular office
hours with appointments. Emergency departments
became the principal provider of primary medical
care when doctors' offices were closed, principally
during evenings and weekends. At the same time,
more physicians entered specialties rather than
general practice. Emergency departments started to
experience a large increase in volume. The increased
volume was a result of use of the ED by patients
with lower acuity problems. Emergency departments
recognized they needed a method to sort patients
and identify those needing immediate care. This
provided the impetus to put ED triage systems into
place. Physicians and nurses who had used the triage
process effectively in the military first introduced
triage into civilian EDs. The transition of the triage
process from the military to U.S. emergency
departments was extremely successful.
Thompson and Dains (1982) identified the three
most common types of triage systems:
- Traffic director.
- Spot-check.
- Comprehensive triage.
Traffic director is the simplest type of system. A nonclinical
employee greets the patient and directs the
patient to a treatment area or the waiting room
based on their initial impression. By 2002, this type
of system no longer worked effectively.
The second type of triage is a spot-check triage
system, appropriate for a low volume emergency
department where it is not cost effective to always
have an RN at triage since patients do not need to
wait. Instead, a registration person greets the patient
and pages the triage nurse when a patient presents.
The RN then determines patient acuity based on a
brief triage assessment. Patient assessment is a
nursing function that cannot be delegated to less
qualified personnel.
Comprehensive triage, the most advanced system,
has continued to evolve in the United States. It is
supported by the Emergency Nurses Association
(ENA) Standards of Emergency Nursing Practice:
The emergency nurse triages each patient and
determines the priority of care based on physical,
developmental and psychosocial needs as well as
factors influencing access to health care and
patient flow through the emergency care system.
Triage is to be performed by an experienced ED
nurse who has demonstrated competency in the
triage role. The goal is to rapidly gather "sufficient"
information to determine triage acuity (ENA, 1999,
p. 23).
Though it is recommended that comprehensive
triage is to be completed in 2 to 5 minutes, Travers
(1999) demonstrated at one tertiary center ED that
this goal was only met 22 percent of the time.
Triaging pediatric and elderly patients has been
found to take more time than other patients. The
level of detail necessary for comprehensive triage
can be difficult for the experienced nurse to
complete in a short timeframe such as 2 to 5
minutes. The triage nurse is expected to obtain a
complete history, take vital signs and complete
department-specific screening questions. Sufficient
information must be obtained to make the correct
triage decision. Under-triage in the era of ED
overcrowding can compromise patient safety.
Emergency nurses must question whether we have
set unrealistic standards for ourselves and whether
the distinction between a comprehensive triage
assessment and initial assessment remains clear. A
comprehensive triage system can lead to a backlog
of patients waiting to be seen by the triage nurse. In
an attempt to facilitate the flow of patients through
high-volume emergency departments and to ensure
that no patient waits to be seen by a triage nurse,
two-tier or two-step triage systems have evolved. An
experienced triage nurse greets the patient and
decides whether the patient can safely wait for
further assessment and registration or whether they
should go directly to the patient care area. The
decision is based on chief complaint and an "across-the-room assessment."
The introduction of triage systems into emergency
departments in the 1960s, 1970s, and 1980s had a
number of clear benefits for patients and for the
department. Some of the benefits included:
- Each patient being greeted by an experienced triage nurse.
- A patient who cannot wait to be seen is immediately identified.
- First aid is provided.
- A registered nurse is available to meet the emotional needs of the patient and family.
Return to Contents
Triage Acuity
Today most emergency departments in the United
States use some type of triage acuity system. A triage
acuity system is used to communicate to the clinical
staff in the department which patient can safely
wait and which patient needs to be seen
immediately.
In 2001, the Emergency Nurses Association surveyed
U.S. emergency departments about the type of triage
acuity scale used by their department (MacLean,
2002). The survey included responses from 1,380
emergency department managers, which represent
approximately 27 percent of all EDs in the United
States. Sixty-nine percent of the emergency
departments used a three-level scale, 12 percent used
a four-level scale, 3 percent used either the
Australasian or Canadian five-level scale, and 16
percent did not answer the question or used no
triage acuity rating scale.
More recent data reflect a
trend towards five-level triage. In 2003, the National
Center for Health Statistics found that 47 percent of
EDs used three-level triage systems, while 20 percent
used four-level and 20 percent used five-level
systems (personal communication, Catharine Burt,
November 1, 2004).
The commonly used three-level
scale includes these acuity levels:
Patients are rated as
emergent if they have a problem that poses an
immediate life or limb threat (ENA, 2001). Patients
considered urgent are those that require prompt
care, but can wait up to several hours if necessary.
Nonurgent patients have conditions that need
attention, but time is not a critical factor.
As emergency departments and the health care
system have continued to change, the value of the
existing acuity rating scales have come under
increasing scrutiny. This scrutiny led to research
which found traditional triage models inadequate.
In particular, emergency medicine and emergency
nursing leaders question the reliability and validity
of the three-level acuity-rating scale being used by
the majority of EDs in the United States. The
definitions of emergent, urgent, and nonurgent are
unclear, not uniform and are often hospital
dependent and nurse dependent. Wuerz, Fernandes,
and Alarcon (1998) measured the interrater and
intrarater agreement of three-level triage. Agreement
was measured with the kappa statistic, which ranges
from 0 (no agreement) to 1 (perfect agreement).
Triage nurses and emergency medical technicians
(EMTs) at two hospitals were asked to rate the acuity
of five scripted patient scenarios using a three-level
scale. Six weeks later participants were asked to
again rate the same scenarios. Only 24 percent of
participants rated all five cases the same in both
phases. The overall kappa statistic for severity rating
was 0.35, which shows poor agreement among
nurses.
Rapid, accurate triage of patients is key to successful
emergency department operations in the 21st
century. In particular, the triage nurses' initial acuity
categorization is critical. Under-categorization
(undertriage) leaves the patient at risk for deterioration while waiting. Initial overcategorization
(overtriage) uses scarce resources,
limiting availability of an open ED bed for another
patient who may require immediate care. For these
reasons, the initial triage categorization by the triage
nurse must be as accurate as possible. Accurate triage
categorization can only be accomplished by the use
of a reliable and valid triage acuity system in which
all ED nurses have been adequately trained. Initial
triage categorization is not as important in small,
low volume emergency departments where there is
often no wait to be seen. Unfortunately, this is not
the case for most EDs throughout the United States.
However, an important benefit of using a valid and
reliable triage system is the ability to use triage data
to describe ED casemix. Therefore, using a valid and
reliable triage system is also important in low-volume
EDs.
Return to Contents
Recent Trends Affecting
Emergency Departments
Many opposing forces affect our ability to provide
quality care and maximize patient flow through the
ED. Emergency department overcrowding is a well-documented
problem in the United States today;
patient volumes continue to rise for many reasons
and this trend is not likely to change in the near
future (Adams & Biros, 2001; Derlet, Richards, &
Kravitz, 2001; Taylor, 2001). The American Hospital
Association (2002) reported 90 percent of hospital
emergency departments perceive they are at or over
operating capacity. This translates into longer
waiting times to be seen and longer lengths of stay
in the ED. The average waiting time to be seen by an
emergency physician in 2001 was 49 minutes,
which represented an increase of 11 minutes from
1997 (McCaig & Ly, 2002).
Factors contributing to the increase in ED patient
volumes and waiting times include a decrease in the
number of U.S. emergency departments, aging of
the general population, longer lengths of ED stays,
an inability to move admissions into the hospital
because of a decreased number of inpatient beds due
to hospital closings and downsizing, an increase in
the number of uninsured patients, poor access to
primary care, and a nursing shortage which often
leaves open beds unable to be used due to lack of
nursing staff. The impact of these issues on triage
will be discussed in detail.
The number of visits to emergency departments in
the United States is continuing to grow. The
National Hospital Ambulatory Medical Care Survey:
2002 Emergency Department Summary reports an
estimated 110 million visits were made to
emergency departments in 2002 (McCaig & Burt,
2004). This represents an increase of 23 percent
between 1992 and 2002, with an average of 38.9
visits per 100 persons in 2004.
The highest rate of ED visits is by persons age 75
and older. This rate is approximately 61.1 visits per
100 persons (McCaig & Burt, 2004). The U.S. Census
Bureau (1996) reports that the number of persons in
the 65 to 74 age group and in the 75 and older
category will continue to grow rapidly. In 1990 there
were approximately 10 million persons in the 75
and older age group. This number is projected to
grow to 23 million by 2030. One in eight Americans
was 65 and older in 1994; by 2030 this ratio will
change to about one in five. This age group has the
highest number of emergency department visits;
thus, it is expected that EDs will see a continuing
increase in the number of visits by the elderly
population each year.
There were approximately 39 million uninsured
persons in the United States in 2001 and that
number is continuing to rise (U.S. Department of
Health and Human Services, 2002). Individuals may
be uninsured because they lack access to a group
plan or are unable to afford the cost of health
insurance. The number of immigrants with health
insurance is low (Velianoff, 2002). Many of these
individuals are using and will continue to use
emergency departments for primary care.
The actual number of emergency departments in the
United States has continued to decline (McCaig &
Ly, 2002). Over the 3-year period from 1997 to 2000,
the number of hospital emergency departments
decreased from 4,005 to 3,934. As the demand for
ED services continues to increase, the number of
annual visits to each emergency department has
increased 14 percent on average.
At the same time, the actual number of hospital
beds across the country has decreased. For example,
the American Hospital Association reports that
between 1994 and 1998 the number of inpatient
beds nationwide dropped 8 percent (Shute &
Marcus, 2001). As a result emergency departments
are experiencing difficulty moving admitted patients
into the hospital, at times creating gridlock.
Hospitals are making changes to cope with the
volume. For example, systems are being put into
place to clean rooms more efficiently and physicians
are being asked to make rounds and discharge patients earlier in the day. Despite these efforts, the
average emergency department length of stay for
both admitted and discharged patients is increasing.
Anecdotal reports of patients staying in an ED for
days are no longer uncommon.
The nursing shortage is another factor that has
impacted emergency department overcrowding.
Most emergency departments are facing serious
staffing issues and are increasingly turning to new
and/or inexperienced ED nurses. The average ED RN
is very experienced but is 45 years of age, working
harder, and concerned about the increased volume.
For many, the solution is leaving for a position that
is less stressful and offers more control over their
own assignment. In-house nursing shortages directly
affect the ED, as some open beds cannot be filled
due to the unavailability of a nurse to staff the bed.
Emergency departments are in a unique and
challenging position with regard to controlling
patient flow in and out of the unit. As opposed to
inpatient units that don't admit patients when they
are full, EDs have generally been thought of as units
that are always open, with a potentially limitless
capacity for patients. Most emergency departments
have little control over when admissions can be
transferred from the ED to their assigned inpatient
bed. One option for the overcrowded ED is to try to
control the "front end," or the number of patients
presenting for care. Some hospitals have the ability
to close to ambulance traffic for a period of time,
which is known as "going on diversion" or bypass.
This is a strategy EDs can use when they are
overcrowded and unable to safely care for any
additional patients. This strategy may buy an
emergency department time to deal with the
patients already in the department; however, it is
not a panacea for the problem of overcrowding. Due
to their remote location some hospitals do not have
the option to divert ambulances. Diversion is not an
absolute solution, since 75 percent of patients arrive
at the emergency department by means other than
ambulances (McCaig & Ly, 2002).
Clearly a busy emergency department can lead to
delays in care. One problem related to the increase
in volume is emergency departments seeing an
increased number of patients who are choosing to
leave prior to a medical screening exam (Derlet,
2002). The patient may recognize that the wait to
see a physician is significant and they decide to
leave without being seen. While some of these
patients may have less urgent conditions and suffer
no ill effects by leaving the ED, others may be at risk
for serious consequences by not receiving treatment.
Those patients who stay may endure long waits and
suffer adverse events.
In June 2002, the Joint Commission on
Accreditation of Healthcare Organizations (JCAHO)
released a Sentinel Event Alert (JCAHO, 2002).
JCAHO identified emergency departments as the
source for more than half the reported sentinel
event cases of patient death or permanent disability
due to delays in treatment. In 31 percent of the
cases, overcrowding was identified as a contributing
factor.
The causes and effects of ED overcrowding are
complex and difficult to define, and researchers
continue to develop metrics to measure them
(Derlet, Richards & Kravitz, 2001; Weiss et al., 2004).
Many models identify increasing patient acuity as a
major factor in ED overcrowding, and in some
studies researchers have used triage ratings to
represent ED patient acuity (Derlet & Richards, 2000; Liu, Hobgood & Brice, 2003). It is even more
important to move beyond defining overcrowding
and examining the effects of overcrowding on
patient outcomes. The triage decision is an
important element to be examined. Due to the
prevalence of overcrowding, many EDs are actually
beginning to implement protocols that involve a
physician or nurse practitioner role at triage. The
benefits and cost-effectiveness of this arrangement
need to be studied.
In many ways emergency departments today are
facing the same major issues seen in the late 1950s
and early 1960s. At that time EDs were dealing with
an increase in volume. No method was in place to
identify the patient who needed to be seen
immediately from the one who could wait safely.
Patient safety was a major concern. One solution
was the introduction of basic triage principles into
the emergency department. Today, EDs are once
again facing the issues of overcrowding and finding
that some of the triage solutions put into place in
the 1950s are no longer effective. The current state
of overcrowding threatens patient safety and has
caused an increased focus on triage. The triage
process, use of standing orders, and a physician or
nurse practitioner role at triage are all important
concepts that need to be examined to optimize
safety of the triage process. Attention to adequate
training of triage nurses is another critical element
that requires attention.
While all of these issues are
important, the selection of a reliable and valid triage
system is a fundamental decision to help begin to address safety at triage. Current triage acuity systems
are inadequate given the complex issues facing EDs.
There is a need to replace the traditional triage
acuity system with a research-based, valid, and
reliable system. The Emergency Severity Index (ESI),
which is introduced in Chapter 3, can provide EDs
with a reliable, valid triage system. The ESI is a triage
system that accurately identifies those patients who
need to be seen immediately from those patients
who can safely wait to be seen. The ESI is discussed
in detail in subsequent chapters of the handbook.
Return to Contents
References
Adams JG, & Biros MH (2001). The endangered safety
net: Establishing a measure of control. Academic
Emergency Medicine 8(11):1013-5.
American Hospital Association (2002). Emergency
department overload: A growing crisis. Medical Benefits
19(10):8.
Derlet RW (2002). Overcrowding in emergency
departments: Increased demand and decreased
capacity. Annals of Emergency Medicine 39(4):430-2.
Derlet RW, & Richards JR (2000). Overcrowding in the
nation's emergency departments: Complex causes and
disturbing effects. Annals of Emergency Medicine 35(1):63-8.
Derlet RW, Richards JR, Kravitz RL (2001).
Frequent overcrowding in U.S. emergency departments.
Academic Emergency Medicine 8(2):151-5.
Emergency Nurses Association (1997). Triage: Meeting the
challenge. Park Ridge, IL: Author.
Emergency Nurses Association (1999). Standards of
emergency nursing practice (4th ed.). Des Plaines, IL:
Author.
Emergency Nurses Association (2001). Making the right
decision: A triage curriculum (2nd ed.). Des Plaines, IL:
Author.
Joint Commission on Accreditation of Healthcare
Organizations (JCAHO) (2002). Delays in treatment.
Sentinel Event Alert 26:1.
Liu S, Hobgood C, & Brice JH (2003). Impact of critical
bed status on emergency department patient flow and
overcrowding. Academic Emergency Medicine 10(4):382-5.
MacLean S (2002). 2001 ENA national benchmark guide:
Emergency departments. Des Plaines, IL: Emergency
Nurses Association.
McCaig LF & Burt CW (2004). National hospital
ambulatory medical care survey: 2002 emergency
department summary. Advance Data from Vital and Health
Statistics 340. Hyattsville, MD: National Center for
Health Statistics.
McCaig LF & Ly N (2002). National hospital ambulatory
medical care survey: 2000 emergency department summary.
Advance Data from Vital and Health Statistics 326:1-31.
Hyattsville, MD: National Center for Health Statistics.
Shute N & Marcus M (2001). Code blue crisis in the ER.
U.S. News and World Report 131(9):54-61.
Taylor TB (2001). Emergency services crisis in 2000—The
Arizona experience. Academic Emergency Medicine 8(11):1107-8.
Thompson J & Dains J (1982). Comprehensive triage.
Reston, VA: Reston Publishing Company, Inc.
Travers D (1999). Triage: How long does it take? How
long should it take? Journal of Emergency Nursing 25(3):238-40.
United States Bureau of the Census (1996). Current
population reports, Special Studies, 65+ in the United States
(pp. 23-190). Washington, DC: U.S. Government Printing
Office.
United States Department of Health and Human Services.
(2002). Health insurance coverage improves for
American children. HHS News 02(04):1.
Velianoff, G. (2002). Overcrowding in the emergency
department: The health care safety net unravels.
Nursing Clinics of North America 37(1):59-65.
Weiss SJ, Derlet R, Arndahl J, Ernst AA, Richards J,
Fernandez-Frankelton M, et al. (2004). Estimating the
degree of emergency department overcrowding in
academic medical centers: Results of the national ED
overcrowding study (NEDOCS). Academic Emergency
Medicine 11(1):38-50.
Wuerz RC, Fernandes CM & Alarcon J (1998).
Inconsistency of emergency department triage.
Emergency department operations research working
group. Annals of Emergency Medicine 32(4):431-5.
Return to Contents
Proceed to Next Section