Emergency Severity Index v.4 Lectures


[Music] [Music] The purpose of this segment of the DVD is
to familiarize emergency physicians and emergency nurses with the rationale to
support your decision to implement a five-level triage system in your department. We must begin with discussing the
concept of triage in general. What is the real purpose of triage? The first, and always most important function of triage, is to identify which patients that need to
be seen immediately, and separate them from those patients that can safely wait
to be seen. In the current era of crowded and overcrowded
EDs, triage becomes more important everyday. In order to accomplish this critically important
task, triages nurses must be experienced emergency
department nurses who have demonstrated competence in the triage role. Triage decisions must be correct. It is never safe to under-triage as this compromises
patient safety. Over-triage is
also a problem and results in taking away a bed from the next patient who will
actually need it more. Triage acuity rating decisions must be right. Therefore, more and more attention
is being paid to triage systems in the United States. So, realizing that the most important purpose of triage is to enhance patient safety in
the waiting room by determining who can wait, we also begin to realize, triage
data can be used for other purposes. For example, hospital administrators will
now have a way of describing the acuity of your department, other than admission
rates. Administrators can use triage data to describe
not only the number of patients seen in the Emergency Department (ED), but also describe the acuity level of those patients, and changes in acuity over time and during different
periods of the day in your department. This data can also be used in real time as well. Do you need extra help tonight? Should you be going on bypass? Triage data can help
administrators make decisions to increase staffing levels and identify other
departmental resource needs. Once integrated in your department, a good
triage system will allow you to compare yourself
to other departments or to yourself over a designated time period. So, you can see, a good triage system provides
more data than just describing the volume that you see in your department. So next, we ask who else uses our triage data? Actually, many agencies use emergency
department triage data. State and local health departments, as well as government policy makers, use our data to help
identify trends in ED visits. Our triage data help shape the need for public policy changes. In the current era of
bioterrorism threats, real-time data can be used to help identify possible
bioterrorism events. For example, a sudden increase in young, healthy Level 3 patients with respiratory complaints, post
exposure at a public event, could raise the suspicion of a possible bioterrorism event. Finally, the Centers for Disease
Control are primarily responsible for monitoring trends in emergency department care. The CDC conducts the National Hospital Ambulatory Medical Care Survey, also called NHAMCS. This survey is conducted every year by the
Department of Health and Human Services, by the CDC, the national center for health statistics. Emergency departments across the
United States are sampled. Surveyors select a number of EDs and come out and abstract many different data points from
your ED records. Data points include demographics, chief complaints, treatments, medications administered,
as well as the triage acuity rating. Historically, the CDC has used
a 4-level triage system to categorize ED visit acuity. This data is reported yearly to
describe the “state” of EDs. In a report published by the CDC in 2002,
the CDC reported an alarming 17% decrease in the number
of patients with emergent complaints seen from 1997-2000. This surprised many emergency clinician
leaders. None of us, you included, probably felt
this decrease in acuity. The report was limited by the quality of
data that was abstracted. The CDC uses a 4-level triage
system, while most EDs used a 3-level triage system. The problem the CDC
faces when attempting to describe emergency care is the lack of a standardized
triage system in the United States. Unlike Australia, Canada, and the United
Kingdom, currently in the US, EDs use whatever triage system they choose. Previously, 3-level triage systems have been
the standard. Unfortunately, as we
will discuss, 3-level systems do not use standardized criteria and have been
shown to be unreliable and not valid. In other words, there are no standardized
definitions for 3-level triage acuity rating systems. Consequently, data used from
3-level systems are meaningless and not helpful in describing a particular
department’s acuity status. It is also not a safe triage system when triage
nurses will routinely triage patients differently. So, the CDC is limited by our lack of the
use of a standard triage system in the United States. So, it is up to us, individual EDs, to begin
to use a valid and reliable triage system. And, both professional
organizations, the American College of Emergency Physicians and the
Emergency Nurses Association have recognized the need to standardize triage
systems in the US. A joint task force between the 2 organizations
was initiated in 2003 and remains active today. The committee was charged with reviewing the
evidence for all 5-level systems and making a recommendation to both
organizations. A comprehensive literature review
on all 5-level systems was conducted and published in the Journal of
Emergency Nursing in February 2005. The task force recommended the
following policy that was approved by both organizations in 2003. “The American College of Emergency Physicians
and the Emergency Nurses Association believe that the quality
of patient care would benefit from implementing a standardized ED triage scale
and acuity categorization process. Based on expert consensus of currently available
evidence, ACEP and ENA support the adoption of a reliable, valid
5-level triage system”. Clearly, the need for
EDs to adopt a 5-level triage system with established reliability and validity has
been recognized. While the group stopped short of
recommending a particular system, the published paper suggests either the
Canadian Triage and Acuity System (CTAS) or the Emergency Severity Index (ESI) are acceptable scales. So, the need for 5-level is clear from
a National perspective. We realize many of you may still be
using a 3-level system. So, you must evaluate
the following questions: Can you manage your waiting room with 3
triage levels? Can you easily describe the acuity of your
patients in the waiting room AND in the treatment area with only 3
choices? When faced with 10, 20, or
30 patients waiting to be seen in the typically over-crowded, or crowded ED, can you differentiate using only 3 levels, “really sick”, “sick” and “not so sick”? Probably not. If you don’t have any waits in your department
this seems a moot point. However, it’s not. It is critical be able to report acuity levels
for every ED patient including those not “triaged” by the triage
nurse. For example, the trauma
patients, cardiac arrest and all patients arriving by ambulance who may bypass
triage. If these patients do not receive an acuity
score, they are not included in your case mix data and you therefore under-estimate
your departmental acuity. Remember the CDC report. You want to be able to capture the true acuity of your total patient population. So, back to the original questions,
can you tell your hospital administrator the real acuity in
your department at any minute? And, why do we care so much now? Well, there is a crisis in the ED! We are all acutely aware of the overcrowding issue, but so is most of the public. This is now a dated cover of the
US News and World Report. We all recognize we are overcrowded and this makes the triage decision
so much more important. But, why are we overcrowded? I don’t need to tell you the answer to that
question. Hospital closures, down-sizing, the nursing
shortage and the aging population have contributed each in their own important
way to our current over-crowded state. Finally, we are always the safety net for
anyone who cannot obtain health care. The prediction is the number of individuals
without health care is only expected to increase. We are always open and
must always provide care. We must be prepared to accommodate
large numbers of patients. The time to use a valid and reliable
5-level triage system is now. So, what systems are we using
in the United States? This data is the DEEDS data,
published in 2001 by the ENA, and reflects approximately 1/3 of all
EDs in the US. As you can see, at that
time, most EDs were still using 3-level systems, with a few using 4-level and
even fewer using 5-level. We expect that number will continue
to dramatically shift towards more and more hospitals
using 5-level systems. Of importance is the CDC data from 2001
compared with 2002 data. This slide demonstrates the
importance of reporting a triage score for every ED patient. As can be seen in the 2001 data in the blue bar, approximately 25% of visits did not have a triage acuity level assigned. When reviewing the 2002 data, that number
dramatically decreased and you can clearly see that the patients
with no triage acuity score reported actually reflect sicker patients. There was a larger increase in the number
of level 1-3 patients when compared with the
increase in the number of level 4 patients. Stated more clearly, the patients that did
not receive an acuity score were sicker. So take credit for all of your patients. In general, the data
demonstrates the continued large increase in the number of ED visits –
110 million visits. So, we must then ask,
what are we overcrowded with? Is it “sick” or “not so sick” patients? The general public perception may be that
we are seeing a lot of “not so sick” patients. While there may be institutional differences,
in general, most emergency clinicians believe we are seeing
sicker patients. But, can we prove it? Do you use a triage system that can generate
data that administrators and policy makers can actually use? If you are using a 3-level system, the
answer is definitely not. So, what makes a triage system useful and a “good” triage system? It must be reliable and valid and triage of
any particular patient cannot depend either on who the triage nurse is that
day, or on who the attending physician is that day. Without a reliable triage system, you cannot
describe your case mix and you certainly cannot
compare yourself to other EDs. So, let’s talk about two crucial concepts to choosing a triage system: reliability and validity. Reliabilty of a triage system means the consistency
or agreement. In other words, do two nurses assign the same triage score to an individual patient? This is described as inter-rater reliability. Triage systems also measure intra-rater reliability. For this question we ask, will I assign the
same triage score to the same patient next week? Inter-rater reliability can be measured with
different statistics, but is frequently measured with
the kappa statistic. A kappa of 0 indicates no agreement,
and 1 indicates perfect agreement. Generally, a kappa score of 0.7 or greater
indicates very good agreement. A triage system must
also be valid. Validity is an accuracy term. For example, when we measure
pain, we want to make sure we are measuring pain and not anxiety. Measuring validity of a triage system
is a little more challenging and proxy measures of
acuity have been used. Typically hospitalization, number of resources
used and death have been used to measure validity of
triage systems. A triage system
would be deemed not valid if a high proportion of low acuity patients were
admitted to the hospital as typically, patients with low acuity complaints are
discharged home from the ED. Now that we understand these terms let’s
look at what we know about 3-level systems. The late Dr. Wuerz examined these
concepts by asking triage nurses to rate written patient scenarios using a 3-level
system. In phase I, kappa scores=.347. In phase II, Dr. Wuerz measured intra-rater
reliability and found even more concerning results, kappa scores were
between .145 to .554, very poor. Even worse, only 24% of nurses rated all their
cases the same during both phases. Does this mean the triage nurses were not
good? NO! It means that they were using a very poor
triage system. It is our job as emergency leaders to give triage nurses the best tools that we can to
enable them to do their job well. How could they possibly have good inter-rater
reliability when 3-level systems do not have definitions, guidelines, or rules? We can do better. So, in closing, using a 3-level system, you
cannot reliably describe what is going on in your waiting
room right now. How can you make
critical decisions like when to go on bypass? You don’t really know who is and who is
not stable in your waiting room. You can’t describe your acuity on a day-to-day or hour-to-hour basis. And you certainly can’t predict staffing
needs or urgent care hours, or anything else. So, it is time for 5-level triage. And, the CDC has also recognized this. As we mentioned before, the NHAMCS survey is performed annually by the CDC. Beginning in 2005, it will collect data using a 5- not 4-level system. So, you may wonder, what other 5-level systems
are available? There are actually four systems in the literature. The Australasians have been
using 5-level triage the longest and began in the mid-1990s. The Canadians modified the
Australasian Triage System (ATS) and rapidly followed with
the Canadian Triage Acuity System, CTAS, in the mid-1990s. Emergency clinicians in both countries are
required to use the scale and report data to the national ministries of health. The triage ratings are a rich data source
in these countries, and decisions are made with this information. In the United Kingdom,
the Manchester Triage Scale is used. This is a chief complaint driven algorithm. The triage nurse must first choose one of
52 chief complaints and then evaluates specific key elements. There are very good web sites available for
both the CTAS and ATS. The fourth triage system, the Emergency Severity
Index or ESI, was developed by Dr. David Eitel and the late Richard Wuerz,
both emergency physicians. Their research involved a team of emergency
nurses and physicians in the development and refinement of the algorithm. With proper training, the algorithm is easy to use. It also has demonstrated very good
reliability and validity. A training handbook was published in 2003. There are actually many publications that have reported the reliability and validity of ESI. Today, we will present the results of only
a few studies. Original work by Dr. Wuerz investigated
the inter- and intra-rater reliability of written scenarios. Weighted kappa scores ranged between 0.68-.87
among a group of nurses. We conducted our own evaluation of inter-rater reliability at Northwestern by
retrospectively reviewing 402 actual patient triages at our facility that was not an
original research site. We reported a weighted kappa of .89. As we mentioned,
more data is available and most studies have been published in the journal,
Academic Emergency Medicine. In addition to reported reliability, validity of the ESI has also been reported in the literature. This paper by Eitel et al reports
hospitalization as a measure of validity. As you can see from the chart, there is
not much variability in admission rates for each triage category, between the 7
different EDs shown. Almost 100% of level 1 patients are hospitalized,
somewhere between 60 and 80% of level 2 patients are
hospitalized, 30-50% of ESI Level 3 patients were hospitalized and almost 0 level
4 and 5 patients were hospitalized. While there is some variability, this work
demonstrates good validity of the ESI triage system. An individual hospital could use their own
data to help administration make real-time decisions. If an ED waiting room had 20 level 3
patients and 5 level 2 patients, it would be easy to predict that between 9 and 15
of the level 3 patients would eventually be admitted and between 3 and 5 of the
level 2 patients would be admitted. Administrators can use this data to plan
ahead and hopefully prevent a bed crunch crises much earlier and ideally
implement actions to help diffuse the situation. Finally, this chart depicts the
ability of EDs using the ESI system to benchmark or compare themselves to
other EDs. Case mix is depicted on the y axis and triage
level is on the x axis. It is clear from examining this graph, that the 17th street ED sees a very different
case mix with a much lower acuity. And, indeed this is the case. The facility is more of an urgent care walk-in
than a full service ED. So, in summary, we have emphasized
how critically important the triage role is today in the era of overcrowding. The use of a poor triage system has major
safety implications. In addition to the first basic purpose of triage, we have discussed how hospital
administrators and public health agencies use our triage data whether we realize
it or not. So, it is time to start using a reliable and
valid triage system not only to improve patient safety, but to provide better data. [Music] The purpose of this segment is to teach you
how to use the Emergency Severity Index. I will review the algorithm in detail and
will incorporate case examples to clarify important points. At the end of this segment, you should be
able to use ESI to classify any patient that presents to your emergency
department. Following this segment, you will have the opportunity
to apply what you have learned to patient scenarios. The Emergency Severity Index uses both acuity
and expected resource consumption to determine triage priority. Acuity or severity of illness or injury is
used to assess the ABCs: airway, breathing and circulation. The triage nurse
evaluates the potential for life, limb or organ threat to determine how soon the
patient needs to be seen. Acuity is always assessed initially. After determining
there is no immediate life threat, expected resource consumption is evaluated. Based on the patient’s chief complaint and
brief triage assessment, the triage nurse estimates the number of resources a patient is expected to consume in order for a disposition decision to be reached. In other words, what is it going to
take to get this patient out of the emergency department – whether that is
admitted, discharged, or transferred. I will talk more about resources later in
this segment. ESI is a 5-level triage system; each of
the levels is clearly defined and mutually exclusive. A patient can’t be a little bit of this
and a little bit of that – they must be assigned to one level. There are no 3.5s. ESI allows for rapid sorting of patients upon presentation into one of these five levels. Triage using ESI differs from a complete assessment in respect to the amount of information
that is obtained. For triage with ESI, only the information
necessary to assign an ESI level is gathered. The triage nurse is gathering only enough
subjective and objective information to assign
a triage acuity rating. In this era of ED overcrowding,
rapid sorting at triage is critical. A triage assessment should, on
average, take between 2 and 5 minutes. As Paula stated earlier, triage data needs
to be valid and reliable. One of the ways to ensure that this requirement is met is by having an experienced
emergency department nurse at triage – a nurse who has attended an ESI
educational program and whose competency has been validated. A triage nurse
needs the knowledge and critical thinking skills to function effectively in this role and
rapidly sort patients. An inexperienced triage nurse will make unsafe
decisions with any triage system. To understand how ESI works, it is important
to first look at the conceptual model of the algorithm. You will notice 4 decision points: A, B, C and D. To use the algorithm, the triage nurse
always starts with decision point A – is this patient dying? If the answer is yes, the patient meets
ESI Level 1 criteria If the answer is no, the nurse moves
on to decision point B. But let’s look closely at decision Point A
– Is this patient dying? The criteria used to answer this question
are: does this patient require immediate life saving interventions? If the answer is yes, the patient is
assigned ESI level 1. What is meant by immediate life saving intervention? Does the patient have an obstructed or partially
obstructed airway? Are they unable to protect their own airway? Are they apneic, or were they intubated prior
to arriving at the hospital? Or, does the triage nurse feel they need to be
immediately intubated or placed on noninvasive ventilatory support because
they are they in severe respiratory distress? Or, is their oxygen saturation less than
90%? Is this a patient who is pulseless? Is the triage nurse concerned that the
pulse rate, rhythm or quality is an immediate threat to life? Does this patient
require immediate cardiac pacing, cardioversion, defibrillation or immediate
intravenous access and large amounts of fluid or blood given? Is this a patient who
requires immediate medications to reverse a threat to life or limb? Does this
patient have an acute change in mental status that requires immediate life saving
intervention? For example, are they hypoglycemic and require
IV glucose, a heroin overdose who needs a reversal agent,
a subarachnoid hemorrhage that cannot protect their own airway? For decision point A, the triage nurse is
identifying the patient who on the AVPU scale meets the P or the U – a painful
stimulus is required for response or they are unresponsive. Any patient assigned to ESI level 1 is physiologically
unstable and requires immediate life
saving interventions. These patients
need immediate care by both an ED MD and one or more ED RNs – In some
EDs this may be a formal team response like the code team or trauma team. The hospitalization rate for ESI Level 1 patients
is very high – most will be admitted to intensive care units – a few will
die in the ED and a few will be discharged or may leave against medical advice
following treatment. Some examples of ESI level 1 patients:
a cardiac or respiratory arrest, an overdose with a respiratory rate of 8, someone in severe respiratory distress with agonal or gasping respirations, acute shortness of breath with
an oxygen saturation of less than 90%, anaphylactic shock, a critically injured trauma patient – for
example, a gunshot wound to the abdomen with a palpable blood pressure of 88, chest pain, who’s pale, diaphoretic with a BP 90/palp, patient with a chief complaint of dizziness,
with a recent loss of consciousness, who presents with a HR of 40, patients with chest palpitations with a HR of 180, unresponsive with a strong odor of alcohol, severe stroke who needs airway protection. For each of these examples, the patient needs immediate, aggressive life saving
interventions. The question is will this intervention save
this person’s life? Interventions include: intubation, surgical airway, continuous positive airway pressure (CPAP) or BiPAP, ventilation with a bag valve mask device,
defibrillation, cardiovarsion, external pacing, chest needle decompression, significant volume replacement with crystalloid
or colloid, immediate administration of medications such
as vasopressors, glucose, and the control of major arterial bleeding. There are many interventions that may be performed routinely in the emergency
department for diagnostic and therapeutic purposes that are not life saving. These include: diagnostic tests such as EKG, lab studies, administration of supplemental oxygen, placing the patient on a cardiac monitor,
obtaining IV access, and the administration of medications that are important
but not immediately life saving – these may include ASA, nitroglycerine, pain
medications, antibiotics, and heparin. These interventions are NOT considered life
saving and do NOT meet Level 1 criteria. It is important to clearly understand the
need for immediate life-saving interventions. If you do not, you will end up over-categorizing
ESI Level 2 patients and the meaning of ESI Level 1 will be lost. When you communicate
with your physician and nurse colleagues that you have a patient that meets ESI Level 1 criteria, everyone will know this is the sickest patient in the department and both the
doctor and nurse should drop what they are doing. The patient requires
immediate life-saving interventions that cannot be administered only by the
nurse. This is not a patient that the nurse can put
on the monitor, get a line, and ECG and then let the doctor know about the
patient. To summarize, decision point A – is this
patient dying? If the answer is yes, the patient meets ESI level 1 criteria.
If the answer is no, the triage nurse moves on to Decision point B – is this a
patient who shouldn’t wait? There are three questions the triage nurse
needs to use to identify the patient who meets ESI
Level 2 criteria: Is this a high-risk situation? Is this patient newly confused, lethargic
or disoriented? Is this patient in severe pain/distress? Let’s look at each of these questions in more
detail. Is this a high-risk situation? This is a situation where the triage nurse
feels it would be unsafe for the patient to wait for more than a few minutes for a bed. The patient is presenting with
symptoms of a condition that could easily deteriorate or with a condition that’s
treatment is time sensitive or they have signs or symptoms of a condition that
has the potential for major life or organ threat. Again, this is where the
experienced triage nurse is so important. Using their knowledge of anatomy,
physiology, pathophysiology, emergency medicine, and their clinical experience,
the triage nurse uses their critical thinking skills to identify high risk situations. The determination of high-risk is based on
a brief patient interview or on gross observations of the patient or the triage nurse’s
6th sense or intuition. Some examples of “high risk’ situations include:
the patient with a cardiac history who presents complaining of chest pain who is
physiologically stable, rule out pulmonary embolus in a patient with
multiple risk factors, again physiologically stable, a newborn with a fever – we will discuss these criteria in detail later, rule out ectopic pregnancy, neutropenia with a fever,
suicidal/homicidal, a needlestick in a healthcare worker. Moving on – let’s talk about: is this patient confused/lethargic or disoriented? The triage nurse needs to identify the patient with an acute change in level of
consciousness. Is this a situation where the brain is structurally
or chemically compromised? The triage nurse has to determine that the patient is physiologically stable and does not require
immediate life saving interventions. Examples of new onset confusion, lethargy
or disorientation. Let’s talk about those. New onset of confusion in an elderly patient
– the family reports that this 86 year old female is usually awake, alert and oriented, she took a nap and now woke up confused. A 30 year old, with a known brain tumor, whose
wife reports that today he is confused. Adolescent who was found confused and disoriented The third question with decision point B: is
this patient in severe pain or distress? This is probably the most difficult concept
for new users of ESI to fully understand. Many patients present to the emergency department
complaining of pain and so it is important for triage nurses
to assess a patient’s pain on presentation. Pain is a subjective phenomenon that has been
defined as whatever the patient says it is. Triage nurses are required to assess and document
a patient’s pain upon presentation to the
ED. Patients should be asked to rate
their pain using a research-based pain intensity rating, scale such as the visual
analogue scale or the Wong Baker phases. The triage nurse asks the patient if
they are currently in pain – if the answer is yes, they are asked to rate their pain
on a scale of 0-10, with 0 being no pain and 10 being the worst pain you can
imagine – how would you rate your pain? At the same time, the triage nurse is
completing a brief triage assessment, asking about chief complaint, past medical
history, medications and allergies, and obtaining a set of vital signs. If the patient rates their pain as 7 out of 10 or greater, the triage nurse can assign the patient to
ESI level 2 – IF and ONLY if – the triage nurse can do nothing to relieve their
pain at triage AND the patient is in need of immediate interventions to relieve
their pain. The test has to be whether the triage nurse
would give their last open bed to this patient! Not every patient who rates their pain as
7 out of 10 or greater will be assigned to ESI level 2 – for
many patients the nurse can provide comfort measures to relieve their pain,
such as immobilization, ice, elevation, and distraction. In some EDs, the triage nurse works under
triage protocol that provide for the administration of PO pain
medication to certain patients. If the patient rates their pain as 9 out of 10, but has had the pain for a few days, is
laughing at triage or eating chips, the nurse in good conscience will not give the
patient the last open bed. Remember, it is the patient in severe pain
that cannot be addressed at triage who will be given your
last open bed. Patients who would
meet this criteria include: the patient with a suspected or known kidney
stone who cannot sit still, who is nauseous and needs IV pain medication, the patient with a severe burn, the oncology patient whose pain regimen is
no longer working and they present to the ED for pain management, the patient with a possible dislocated shoulder who is crying, diaphoretic and
clearly in excruciating pain, the patient with a suspected compartment syndrome. Let’s move on and talk about distress – any patient who is experiencing
significant physiological or psychological distress should be assigned ESI Level 2. Some examples of this include: a sexual assault victim,
the combative patient, the homicidal or suicidal patient, the bipolar patient who is manic the acute grief reaction, the known alcoholic with signs of minor head trauma. Frequently, the triage nurse struggles with
assigning ESI Level 2 to any of the patients we just mentioned. However, it is important not to get emotionally
involved in this decision. All of the patients we just discussed are
indeed high risk. All of these patients need rapid assessment
and close attention, and if they leave the ED, the triage nurse will be accountable
for her decision. This is a good example of how triage category and process may differ. For example, you may assign a
psychiatric patient ESI Level 2, but they may not be placed in an
open bed for a period of time. But, the triage nurse may have called for
security standby, notified your psychiatric liaison,
or even initiated lab protocols. The patient remains high risk independent
of when they are assigned a bed. Take credit for the high acuity of these patients. Finally, we must also
discuss severe physiologic distress. Often patients in respiratory distress are
triaged as ESI Level 2. The patient may not require immediate life-saving
intervention, but timely intervention of oxygen may be required. In general, while
there are no time definitions for ESI, it is ideal for all ESI Level 2 patients to be
placed in a treatment area and evaluated by the emergency nurse within 10
minutes of arrival. We have covered high risk situations, new
onset confusion, lethargy and disorientation, and severe pain and/or distress. To review, the following patients
meet ESI Level 2 criteria: the patient with a history of renal colic
who presents with severe flank pain and is vomiting – there is nothing the triage
nurse can do to relive this patient’s pain; they need to be seen and have intravenous
pain medication administered, a patient with severe burns to both arms, a patient with a dislocated shoulder who rates the pain as 10+ and is diaphoretic and tearful, a psychiatric patient who presents to the
emergency department screaming obscenities, is in severe distress and meets
ESI level 2 criteria. ESI level 2 patients remain a high priority. The triage nurse and the emergency
department staff should be working together to facilitate rapid placement of the
patient into an open bed, which is not an easy situation in this era
of Emergency Department overcrowding! Patients assigned to ESI level 2 were identified
by an experienced triage nurse as someone who shouldn’t
wait! With ESI version 3,
about 25-35% of patients were identified as ESI level 2. 50-60% of these patients
were hospitalized and many required intensive care or telemetry beds. So, we’ve now addressed decision point A and
B – let’s move on to decision point C: how many different resources will this
patient consume? Research has demonstrated that experienced emergency department nurses are actually very
good at predicting how many different types of resources a patient will consume. This is based on the standard of care for
a given chief complaint or diagnosis. It is independent of the type of hospital, location of the hospital, physician on duty
or acuity of the department. Of course, there may be some regional variations
in physician ordering. For example, in some areas of the country routine
x-raying of injured toes may not be done – instead the
patient is treated based on the physical exam. Some institutions use the Ottawa Ankle Rules,
others do not. Another example is rapid strep screens – in
some areas the patient is treated based on the history and physical, in other
areas a rapid strep screen will be done to confirm the diagnosis of strep throat. All of these practice variations are ok,
and triage will accurately represent your individual departments’ acuity mix. Let’s look at the ESI decision point C: how many different resources will a patient consume? No resources is assigned ESI level 5, one resource – ESI level 4, two or more resources – ESI level 3. This graph
demonstrates the average number of resources that were actually consumed by
patients for each of the 5 triage levels. The data demonstrate the average
number of resources used decreased monotonically as a function of ESI level –
ESI level 3 used more resources than ESI level 4, ESI level 4 used more than ESI level 5. The triage nurse needs to count the number of different resources a
patient will consume – but what is and what is not a resource? If you look closely at the chart,
we have identified these for you. You will also notice when looking at the algorithm,
although you need to estimate required resources, you never have to count
beyond 2! Lab is a resource –
whether you do one blood test and a urine test or two blood tests, whether you
do one blood test and a culture – it still counts as lab – one resource. ECG is a resource; x-ray – whether you do one xray or 10 xrays, it still counts as
one resource. A CT scan, an MRI, an ultrasound, an angiogram – each of those
count as a resource. When a patient comes to the ED, it is expected
that they will have a history and physical exam – so these
are not resources. Their history
and physical exam should be appropriate for their chief complaint – so if they
have an eye complaint – they will have a slit lamp exam – not a resource. If the patient is female and presents with low abdominal pain, she will have a pelvic
exam – not a resource but part of the H&P for that chief complaint. Point of care
testing does not count as a resource. Examples include finger stick glucose or
pregnancy tests done in the ED. Inserting a saline or heparin lock does not
count as a resource, but the administration of intravenous fluid does count. The administration of IM or IV
medications does count as a resource. The administration of PO medications
does not count, neither does giving a tetanus immunization or handing a patient’s
prescription refill. Specialty consults count as a resource. The patient who needs to see psychiatry
will consume one resource, whereas the patient who
needs to be seen by surgery and social service will use 2 resources. Emergency physicians as part of their plan
of care may contact the patient’s primary care physician by telephone – this does not count as a resource. Finally, let’s look at procedures. Simple procedures such as inserting a urinary catheter or nasogastric tube count as one resource each. Conscious sedation counts as 2 resources. Simple dressings and crutch walking, slings and splints do not count as resources. The ESI research team is asked
many questions about this – nurses comment that they take a lot of
time and should be counted. Remember, we are talking about a triage acuity
rating system not a work load measure. And think, what if crutch walking counted? All patients with a sprain
would now be ESI Level 3. They would require an xray and crutch walking. By including everything we do, we lose the ability to discriminate among the large
number of lower acuity patients. So, don’t worry about it. The definitions are
there to provide standardization. You do not get to decide what is and is not
a resource. You must follow the algorithm. By doing this, you help maintain the
reliability and validity of the algorithm. As I previously mentioned,
patients who require no resources are assigned ESI level 5. Let me give you some examples
of this type of patient: a healthy 10 year old with poison ivy, a healthy 52 year old who ran out of his
blood pressure medicine yesterday, a 22 year old involved in a car accident 2 days ago and wants to be checked out, nothing hurts, a 46 year old with a cold. ESI level 4 patients are also stable and
can safely wait for hours to be seen. This is an ideal group of patients for mid
level providers to care for in a fast track or express care setting. ESI level 4 patients require a history and
physical exam and consume one resource. Some examples of ESI level 4 patients: a healthy 19 year old with a sore throat and fever – one resource: a rapid strep
screen, a healthy 29 year old with a UTI, denies
vaginal discharge – one resource: lab – needs a urine, urine culture and a urine
pregnancy test which together are one resource, a healthy 43 year old with a stubbed toe, “I think I broke it” – one resource: an xray. If your facility does not routinely xray these
patients, then the patient would be an ESI level 5. A healthy 12 year old with a minor thumb laceration
– needs suturing – an ESI level 4 Now, let’s move on to ESI level 3 – 30-40% of the patients seen in Emergency
Departments. These patients require an in-depth evaluation,
and because of this, have a long length of stay in the ED. These patients will require a
minimum of 2 resources. You must then address Decision point D: what
are the patient’s vital signs? The nurse needs to consider the vital signs
when assigning triage acuity. Are they outside the accepted parameters for age? If they are outside of those
parameters, the triage nurse can up-triage the patient to ESI level 2. Vital signs outside the accepted
parameters do not automatically up-triage a patient – instead the nurse
should consider the vital signs and make a decision. This is a really
important point to stress. The triage nurse does NOT have to up-triage
every adult patient with a heart rate of 100 or greater. Decision Point D also includes
temperature for children less than 36 months. The ESI triage research team is following
the American College of Emergency Physicians’ practice guidelines. The infant,
0-28 days, brought to the ED for a fever of 38.0 degrees C or 100.4 F
– this baby should be assigned to at least ESI level 2, regardless of how good
they look or what the chief complaint was. If the neonate’s condition on
presentation suggests that they are in need of immediate life saving interventions,
they should be assigned to ESI level 1. For the infant 1 to 3 months, the triage
nurse should also consider assigning the baby to ESI level 2 if their temperature
is 38.0 degrees C or 100.4 degrees F. For the 3 month to 36-month old
child with a temperature above 39.0 C or 102.2 F, the triage nurse should consider assigning to ESI level 3, if in addition to fever, their immunization history
is incomplete or they have no obvious source of a fever. For the 24 month old
presenting to the ED with a fever of 103 F, and the mom reports the child sees a
pediatrician regularly and woke up from a nap pulling on his ear, the child would
be assigned ESI level 5. Let’s review frequently asked questions: Do I have to upgrade a patient’s triage level if the pain rating is 7 out of 10 or greater? The answer is no, you don’t have
to. Let’s review some examples of patients who
might be assigned ESI level 3, 4 or 5 due to pain. ESI level 3: fractured ankle, abdominal pain, most migraines, ESI level 4: sprained ankle, toe, abscess ESI level 5: a toothache. Do I have to upgrade the patient’s triage level if their heart rate is 104? For the adult with a heart rate of 104, the triage nurse should consider this
as part of the assessment. The patient does not have to be up-triaged to ESI level 2. If the patient is always confused are they automatically assigned to ESI level 2? No – ESI level 2 is for those patients with a new onset confusion, lethargy or disorientation. Does ESI identify time to reassessment for each triage level? No, ESI does not do this and this is a key
difference between ESI and other 5-level triage acuity rating systems. The ESI triage research group has purposefully
not identified reassessment times, but has left that to individual emergency departments to incorporate into their triage policy. We urge you to use caution – in this era
of ED overcrowding, it is very difficult for busy
triage nurses to reassess patients at a set time when they are busy sorting
patients on arrival to the ED. And one last FAQ: What do I do with ambulance patients? These patients
should receive an ESI acuity score using the same criteria we just discussed. What if I assign someone ESI Level 2 and I
can’t get them back right away? This is a great question. As the triage nurse, you are required to identify the triage level You are not accountable to place the patient. Some ESI Level 2 patients – psychiatric patients – may not be placed immediately according to your
protocol, but their triage score is the same regardless of your policy. However, it
is desirable that the rest of your level 2 patients be placed as quickly as possible. “As possible” is the key. You should never lower your triage category
because you know the patient must wait. You must be able to accurately represent the
acuity of each individual patient, as well as your
department case mix. [Music] So, you’re already a user of ESI and want
to know more about the changes from version 3 to version 4 of the algorithm. This segment of the DVD will provide a
succinct description of the changes. First, you are probably wondering, “why
the change?” Feedback from emergency departments using
ESI version 3 provided the impetus for further research. Triage nurses struggled with what they
perceived as two categories of ESI level 2 patients. Frequently, patients present to
triage in acute respiratory distress, may be pale, diaphorectic, hypotensive,
tachycardic or bradycardic, but still be awake, be breathing and have a pulse. Using version 3, these patients do not meet
ESI level 1 criteria. They are critically ill, unstable
and require immediate interventions. The other group of ESI level 2 patients are
those patients that are indeed high risk, in severe pain or distress, or have new onset mental changes. Examples of these patients include: a chest pain patient with normal vital signs
and no respiratory distress, patients with severe pain due to kidney stone or cancer, or perhaps and elderly patient that is weak and dizzy with a significant medical
history, but physiologically stable at triage. The triage nurse is often faced with
the dilemma of multiple level 2 patients and consequently, nurses have to reorganize
their level 2 patients accordingly. While this is certainly the right thing
to do, the research team discussed the situation at length. We began to feel strongly that the
physiologically unstable level 2 patients do require immediate care and probably deserve a higher triage prioritization. In reviewing the conceptual algorithm,
ESI level 1 criteria asks the question “Is the patient dying?” Clearly some of the patients we just described
above may meet this definition if left to wait. The actual definition of ESI Level 1 criteria on the algorithm lists the following criteria: intubated, apneaic, pulseless or nonresponsive. We began to feel strongly that these criteria
were too limiting. So, members of the research team went to work and designed a research project aimed at revising and expanding the ESI level 1 criteria. We conducted a research study at 5 different
emergency departments that had been using ESI for several years. We enrolled over 500 level 2 patients and asked the triage nurse to identify which patients she thought were going to need
immediate interventions. We also recorded vital signs, past medical
history and chief complaint. Patients were later divided into 2 groups:
those that actually received immediate intervention upon arrival
and those that did not. Interventions
were clearly defined. Actions such as starting an IV just to have
one or obtaining an ECG did NOT count. An ECG is a diagnostic intervention and has
never saved a life yet. The IV is nice access, but we don’t save a life with it unless we administer a medication,
fluid boluses or blood. Interventions that did count were any intervention
performed to secure an airway, breathing or circulation. Examples include the
following: prepare for intubation, starting an IV to administer vasoactive or drugs
to control a heart rate, cardioversion, and application of pacer pads. When we analyzed the data, we found that the
triage nurses’ prediction of need for an immediate intervention was the most
important factor that predicted whether or not patients actually RECEIVED
immediate intervention. In other words, triage nurses did not have a problem identifying which level 2 patients
were most sick. A few other factors also predicted immediate
interventions, including severe respiratory distress, and
SpO2 less than 90. We took these factors and re-worded them into
a clinically meaningful way and integrated them into decision point A
on the back of the algorithm. So how will this change practice? Actually, it should just make sense. Any patient that the triage nurse perceives will need immediate intervention, as defined on the back of the algorithm card,
will now meet ESI level 1 criteria. The definitions on the
back are again critical. In maintaining ESI as a reliable and valid
system, you can’t choose the interventions, we did. We have listed examples of interventions
that do not count. These definitions should be adhered to in
order to avoid every level 2 patient becoming an ESI level 1 patient. We really anticipate this change will
affect very few patients, but these patients should be easily identified. In our study,
20% of level 2 patients actually received immediate intervention, and using
version 4 criteria would now become ESI level 1. So, moving forward, the triage nurse should
now ask the following question to help guide the decision to assign ESI level 1: Do I need to bring a physician to the bedside now? If the answer is yes, the patient is probably
an ESI level 1. Patients in physiological distress who require
advanced airway management, NOT just O2, immediate cardioversion, potential
pacing or the administration of vasopressors or other cardiac drugs are good
examples. The administration of large volumes
of fluid or blood are also another example. These patients require
immediate resuscitation efforts. It is important to remember that most level 2 patients should remain ESI level 2, but they
should still remain a high priority. For example, most patients with chest pain will remain an ESI level 2. A high risk trauma patient by mechanism, that walks in and is physiologically stable will remain ESI Level 2. They may be
walked to your trauma room and assigned a different trauma category, but if
stable, they would remain ESI level 2. So, how will this change help you? Again, the language will be beneficial to
your team of nurses and physicians. When you walk back or call with an ESI level 1 patient, and communicate this, it should be
clear to everyone on the team that a physician and multiple nurses will be needed
at the bedside immediately. A stable patient with chest pain is easily managed by the emergency nurse for at
least the first 10 minutes without a physician. The nurse can facilitate an
ECG, obtain IV access, and administer oxygen without a physician present. However, if the patient was hypotensive, the
nurse will need orders for medications and more specific direction. We believe that ESI version 4 will help
facilitate care of the most critically ill patients in your department and also
more accurately describe your acuity mix. This was the major change in the
algorithm from version 3 to version 4 and details of the research project can be obtained
in the literature. You may ask,
“Was this the only change?” Actually, the research team took this opportunity to update one more portion of the algorithm
– pediatric fever criteria. The pediatric fever criteria were strict and outdated. The team reviewed the literature
and updated the criteria to reflect the American College of Emergency Physicians’
Pediatric fever guidelines published in the October 2003 issue of the Annals of Emergency Medicine. There is clear direction that any child less than 28 days with a fever of greater than 38.0
C or 100.4 F or higher should be considered high risk. There is debate
in the literature over what to do with a child between 28 days and 3 months old.
ESI version 4 suggests the triage nurse follow their institutional policy for these
children and may triage these children as ESI level 2 or 3. For children 3 months
to 3 years of age, the triage nurse should consider assigning ESI level 3 if they have a temperature of 39.0 C, 102.2 F, or their immunizations are incomplete or they have no obvious source of fever. In summary, we believe the pediatric fever
criteria will help clinicians more safely assign triage scores to children with fever. So that’s it. Version 4 expands Level 1
criteria to include any patient that requires an immediate life saving intervention and these interventions are research-based and defined on the back of the
algorithm. The second change updates the pediatric fever
criteria to reflect current practice guidelines. Again, we stress, do not change the algorithm:
conduct the research and publish. The ESI research team will continue to evaluate
the research and will update the algorithm as needed. [Music] You now have a very good understanding of
the nuts and bolts of the ESI triage system. It’s time to get serious about implementing
it in your own setting. First off, you need to remember the beauty of ESI is that is it a research based triage
system with established reliability and validity. Having said that, it is important
not to “mess with it”. It is not perfect. However, research is ongoing. If there is something you don’t like about it, you should not change the algorithm. We encourage our peers to
conduct further research and publish. The ESI Triage Research Group constantly reviews the literature and has made changes
in the algorithm based on important research and clinical guidelines. Again, do not change the algorithm, conduct the research. Changes by individuals and
hospitals will compromise the reliability and validity of the system. So, it’s time to
implement and you’re in charge. We are going to discuss several key concepts
to consider based our experiences and those of others who have been using ESI
for a period of time. We will discuss timing of the change, commitment,
involvement, planning, education, the go-live phase and on-going monitoring. First, you want to clearly identify and think
about why you are making the change. Have their been sentinel events with bad outcomes? Or, are you just being
pro-active? What do you expect by implementing ESI? The purpose should
really only be to improve the safety of triage and generate accurate acuity data. ESI will not fix your length of stay or necessarily
ease the burden of long waits at triage or improve your customer service. But it should ensure that the patients
who are waiting are safe to wait. It will also accurately describe your acuity and may generate some very “bad” data. For example, it may demonstrate that your
average wait to physician evaluation for ESI Level 2 patients is 30 minutes. This is clearly not a good thing and has patient safety implications. But, this data can be used to help make
improvements in your department. Another advantage of implementing ESI,
a standardized triage system, is the ability to use a
common language to describe the acuity of patients in your waiting room. The day that you implement ESI, all physicians and nurses in your department will be
speaking the same language regarding patient acuity. You will no longer need to
describe individual patient presentations to describe your waiting room. Next, you need to consider,
how much time do we need to get ready to implement ESI? We are frequently asked this question, and the answer is between 3 and 6
months, depending on the complexity of your department and size of your staff. Typically, institutions don’t leave enough
time to plan for implementation and we know that poorly planned change is the number
one reason for failure. Changing a triage system is a HUGE endeavor, don’t take it lightly. Many times, another change drives
the desire to change the triage system. For example, an ED decides to implement a computer tracking system and they already have a go-live date. At the same time, they realize they should be moving to a 5-level system so they want to implement both on the
same day. Bad idea. One change at a time. Go ahead and implement your tracking system
with 5-level capability. Let everyone get used to that system, but
continue to use 3 levels or whatever system you’re currently used to. When things have settled down,
when everyone is used to the tracking system, then begin the triage acuity rating
system change. Once you begin using your tracking system
AND you have implemented ESI, you will then be using all
5 levels of your tracking system. It is not possible to expect staff to learn a new tracking system and triage system at
the same time. At least, not if you expect lasting change
with either one. In addition, be mindful of other big changes that may be happening in your department. You should also think about key players and make sure they are available for
training and for go-live. Vacations are very important, plan around them. The next thing to consider when implementing ESI is commitment. This will tie in to your
reasons for making the change. Everyone needs to understand the reason for
change. Commitment means commitment to training and
education, and this
means a commitment of financial resources. ESI cannot be implemented
successfully without adequate training and this costs money. You should form
an implementation task force consisting of emergency nurses, triage nurses,
physicians, educators, clinical specialists and administrators. All members’
perspective are critically important. For example, physicians must be included as
they are end users of the system. Triage and staff nurses are the front line users. Educators and administrators are critical
as they will be accountable for planning training and organizing other aspects
of implementation. All members
must be actively involved in the planning process and must understand that the
triage system provides a common language to describe real-time patient acuity. Each member of the task force represents their
discipline and will provide unique perspectives. Nursing will need to identify strong triage
nurses to use as triage preceptors. All triage nurses should have competency
assessed with real patients in addition to paper cases in a classroom
setting. The implementation
team should plan to meet regularly. No one likes going to meetings. So, set goals, stick to
them, have agendas, accomplish tasks at your meetings andmake them very
productive. And fun is also a very good thing. One of your first tasks for your
group is to pick the go-live date. Pick a realistic date and stick to it. Make it a
big deal. Everyone should know about it. There should be very clear
communication in advance. During the task force meetings, you will also
need to plan your educational strategy, not only for
classroom, but also for competency assessment of real patients. It is important for each triage nurse to be
assessed by an expert or preceptor when triaging real
patients. This will require some
serious commitment and significant planning, but it can be done. Your educator or
CNS can be a critical person in this phase of validation. Also, strong triage
nurses can also be used to help assess staff competency. This live assessment is very important. Even though you will use paper cases to assess
competency prior to implementation, patient scenarios are always more unclear. Triaging real patients is always a challenge
and patients do not present with black and white scenarios, but always very
grey. Educational planning in
general is critical. The didactic component can be accomplished
either by using video as a group setting, having clinicians
view segments individually, or by enhancing the video with other cases. The intent of this video is to provide each
triage nurse with an opportunity to understand the system at their own pace. This video will also allow maximum flexibility in how an organization decides to plan
its’ educational component. If you use this video in a group setting,
you can always stop the DVD and lead a group discussion. Group settings can also be
used to discuss aspects of your triage policy. For example, standing orders at triage, how to handle patients brought by the police, etc. Finally, you must always consider how you
would like to complete physician education. Most physicians will appreciate copies
of the ESI research publications. At a minimum, it is
suggested physicians review the section of the DVD that reviews the algorithm
and provides definitions of each triage level. So, classroom training is now
complete. It’s go-live time! And, there should be lots of support, 24/7
on the day-of and surrounding the go-live date. Don’t forget the nightand weekend shifts,
they triage patients too!!! We strongly suggest a team member is available
24/7 and weekends to be physically in the department
to answer questions and assess triage level accuracy. Ideally, a triage preceptor will work with
each triage nurse for a minimum of 4 hours to validate
competency in assigning the ESI triage acuity ratings. If this is not possible, you should plan for
a super-user available on the unit to spot check triage
scores and address problems. This is a large commitment,
but a critically important step in ensuring proper use and
reliability of the system in your institution. Don’t expect smooth sailing. With any change, there will be unexpected situations to deal with, no matter how you
planned. So, provide a mechanism for everyone to contribute
feedback, both positive and negative. You will not be successful if you stifle negative
feedback. And, most importantly, provide lots of positive
feedback to everyone. Discuss cases in which there is disagreement. Do not be negative or punative. Provide a tremendous amount of positive feedback. And finally, evaluation of your use of
ESI is never finished. You should plan for how you are going to conduct
on-going monitoring. There are many ways to accomplish this. You should, at a minimum,
evaluate the accuracy of the triage ratings. This can be accomplished by chart
audits. Ideally, all triage nurses can participate
in the review at some point. Each chart should also be reviewed by a triage expert – your educator, CNS or
designee. Feedback from these reviews should be provided
to the staff. When reviewing charts, it is important to review only the triage note. It is very easy to
look at the discharge diagnosis and predict a triage score. However, the triage
nurse has limited information. So, when reviewing notes, only review triage
information and make your decision regarding accuracy of the triage acuity score
using that information. One successful example was, as a CNS, I reviewed 15 charts per week with staff nurses. Any “mis-triages” were typed up with a
summary of why the triage score was inaccurate. We would accumulate 5 or 6
cases a month and distribute the summaries to each nurse. The nurses enjoyed reading them and
we were able to correct any misunderstandings or knowledge deficits. They were used as a learning tool, not a punitive
mechanism. Another useful strategy may be to establish a mechanism in which anyone can have a case reviewed because they thought the triage score was inaccurate. Most importantly; nurses and physicians should be encouraged to discuss their
concern or rationale with each other. We all learn best by a collaborative
approach. And one final note on something that should
NEVER be audited as a measure of ESI. We are often asked if facilities should actually
count the resources that were used to determine if the
triage nurse was “right”. This should never be done. The triage nurse estimates resources for
ESI Level 3, 4, and 5 patients on arrival only to help differentiate
between the large numbers of patients that do not require immediate evaluation. Triage nurses actually do a very good
job with this. But, patient stories change, new clinical
information is provided, and the number of resources is not a
good indicator to monitor. So, in summary,
there is a lot of planning to be done, so get to it, and best of luck! You and your colleagues will be successful. And you will feel more comfortable that those
patients in your waiting room are safe to wait!

One Comment

Add a Comment

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