DO NO HARM:AMULTIFACTORIAL APPROACH
TO PREVENTING EMERGENCY DEPARTMENT
FALLS—A QUALITY IMPROVEMENT PROJECT
Authors: Nicole S. Cook, BSN, CEN, CCRN, TCRN, Brittany J. Komansky, MHA, BSN, RN, CEN, and
Michael S. Urton, DNP, APRN, AGCNS-BC, Raleigh and Greenville NC
Earn Up to 8.0 Hours. See page 722.
Contribution to Emergency Nursing Practice
The current literature indicates that emergency department falls may be difficult to prevent and predict, however there is a significant gap in available research on
this topic.
This article contributes to the existing literature by
describing a unique, multifactorial approach to emergency department fall prevention to include data review,
fall risk assessment, remote video monitoring, exit alarm
strategy, fall prevention culture, and communication.
Key implication for emergency nursing practice found in
this article: a multifactorial fall prevention program is
necessary, as no single intervention can address all potential causes of patient falls in the emergency department.
Abstract
Introduction: Patient falls in the emergency department are
a unique patient safety issue because of the often challenging
nature of the environment. As there are a variety of potential
causative factors for patient falls in the emergency department,
this project employed a multifactorial approach to prevent patient falls in a Level 1 trauma center emergency department
(adult only) in an urban tertiary care teaching hospital.
Methods: This project was a single-unit quality improvement
intervention that compared postintervention monthly unit-level
data to historic monthly rates on the same unit. The intervention
was multifaceted with patient-level, nurse-level, and unit-level
interventions employed. A task force was convened to review
and identify specific departmental gaps related to fall prevention, complete a retrospective review of departmental patient
falls to determine causative factors, and implement interventions to reduce ED falls. A comprehensive program consisting
of an ED-specific fall risk assessment tool, remote video monitoring (RVM), stretcher alarms, and a robust patient safety culture, among other interventions, was implemented. Patient falls
and falls with injuries were tracked as an outcome measure.
Results: After data driven analysis of causation, selection of
key interventions, staff education, and sustained focus for 2
years, the department experienced a 27% decrease in falls
and a 66% decrease in falls with injuries.
Discussion: A multifactorial approach was an effective strategy to decrease patient falls in the emergency department.
Key words: Emergency; Fall; Multifactorial; Remote video
monitoring; Safety
Introduction
Patient falls are a safety concern for emergency departments
across the country. These events contribute to hospital admissions, increased patient morbidity and mortality, and
in addition, to increased health care costs.1,2 High patient
volumes and the wide variety of patient acuities in the emergency department make it difficult to predict and prevent
patient falls. Additional factors that have been implicated
in ED falls include long distances to restrooms, acute illness
Nicole S. Cook, Member, Cardinal Chapter-446, is Trauma Performance
Improvement Coordinator, WakeMed Health & Hospitals, Raleigh, NC.
ORCID identifier: http://orcid.org/0000-0001-6672-1546.
Brittany J. Komansky, Member, Cardinal Chapter-446, is Director, Emergency
Services, WakeMed Health and Hospitals, Raleigh, NC.
Michael S. Urton is CNS Concentration Director and Clinical Assistant
Professor, East Carolina University, Greenville, NC.
For correspondence, write: Nicole S. Cook, BSN, CEN, CCRN, TCRN,
WakeMed Health and Hospitals, Trauma Services, 3000 New Bern Ave,
Raleigh, NC 27610; E-mail: [email protected].
J Emerg Nurs 2020;46:666-74.
Available online 4 June 2020
0099-1767
Copyright 2020 Emergency Nurses Association. Published by Elsevier Inc.
All rights reserved.
https://doi.org/10.1016/j.jen.2020.03.007
666 JOURNAL OF EMERGENCY NURSING VOLUME 46 ISSUE 5 September 2020
PRACTICE IMPROVEMENT
states, intoxication, and departmental crowding.1,2 As there
are a wide variety of factors potentially leading to ED falls, it
stands to reason that there could be a wide variety of interventions that may aid in preventing ED falls. Unfortunately,
most fall prevention literature is related to inpatient falls,
with limited ED-related literature and available screening
tools.
The purpose of this article was to provide an
example of how a comprehensive, ED-based fall prevention initiative was created and implemented, including
the following components: triage-based fall risk assessment, application of new monitoring technologies,
improved post event analysis, and awareness and recognition activities.
Background
The National Database of Nursing Quality Indicators defines a patient fall as a sudden, unplanned descent to the
floor (or other unintended surface), with or without injury.3
The organization tracks and reports both falls and fallrelated injuries to National Database of Nursing Quality Indicators as part of its quality improvement efforts. In the
inpatient setting, both falls and falls with injury (FWI) rates
are reported as the number of events per 1,000 patient days
(number of events/patient days x 1,000).3 In the emergency
department, fall and FWI rates are reported as the number
of events per 1,000 patient visits.
3
This quality improvement project was conducted in an
adult emergency department, Level 1 trauma center in an
urban tertiary care teaching hospital that sees approximately
89,000 adults (aged 18 years and older) per year. It is part of
a 3-hospital system with 3 additional stand-alone emergency
departments and a separate children’s emergency department. As the largest emergency department within the system, this emergency department’s policies and clinical
practices serve as the basis for much of the entire organization’s ED operations.
Although systemwide efforts to eliminate preventable
hospital-acquired adverse events had resulted in some success in reducing patient falls and FWI, a unit-level analysis
identified emergency departments as a significant contributor of fall events for the organization. From October
2014 through September 2016 (Figure 1) the adult emergency department averaged 62 falls per year for a rate of
0.75 falls per 1,000 visits with approximately 15% of these
falls resulting in injury (FWI rate 0.09 injury falls/1,000
visits). During this time, falls from patients in emergency
departments across the organization contributed 8.4% of
the total system falls.
Reducing the number of ED falls was identified as a critical focus for reducing overall system falls and improving patient safety outcomes. Given the relative size and patient
volume of the adult emergency department in comparison
to the other emergency departments within the system,
unit-level leadership decided that initiating a quality improvement project within the adult emergency department would
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
Q1 FY15 Q2 FY15 Q3 FY15 Q4 FY15 Q1 FY16 Q2 FY16 Q3 FY16 Q4 FY16
Fall Rate (events/1000 visits)
Fall Rate
FWI Rate
FIGURE 1
Baseline adult emergency department falls and falls with injuries, October 2014 to September 2016. Rate ¼ events/(Adult ED Visits/1000). FWI, falls with injuries; Q1, quarter
1; Q2; quarter 2; Q3; quarter 3; Q4, quarter 4; FY, fiscal year.
September 2020 VOLUME 46 ISSUE 5 WWW.JENONLINE.ORG 667
Cook et al/PRACTICE IMPROVEMENT
provide the best opportunity for program implementation
and evaluation of outcomes.
Review of Key Evidence
In emergency departments, falls may be difficult to predict
and prevent with the acute nature of patient visits and the
brief, episodic encounters.4-6 A review of the current
literature surrounding ED fall incidence and ED fall
prevention revealed a significant gap in the research
surrounding this topic. Most of the research done on the
topic of fall prevention has been focused on inpatient
hospital settings, including the design of fall risk
assessment tools and recognition of fall prevention
interventions.2,5,7 In many hospitals, these inpatient tools
have been applied to the ED setting despite lack of validation in these populations owing to lack of an ED-specific
alternative.8 It has only been in the last few years that
small-scale and retrospective studies have been conducted
in ED populations resulting in new insight and potential
screening tools specifically for the ED setting.8,9
Recommendations for effective fall prevention programs generally include key elements such as (1) population
identification, (2) risk factor screening, and (3) an individualized, comprehensive plan of care.8,10,11 In addition, programs are encouraged to incorporate a multifactorial
approach, working to address individual risk, environmental
considerations, and staff knowledge and engagement along
with policies and equipment that support fall prevention activities.10,12
Recent ED-based studies and improvement projects
seem to indicate that modified versions of inpatient fall prevention tools and policies can be effective in the ED setting
but there is a strong need for high-quality research on (1) atrisk populations in the emergency department, (2) EDspecific fall risk prevention activities, and (3) application
of new technologies for screening, monitoring and preventing falls in the emergency department.4,7
Methods
This project was a single-unit quality improvement intervention that compared postintervention monthly unitlevel data to historic monthly rates on the same unit. The
intervention was multifaceted with patient-level, nurselevel, and unit-level interventions employed. Statistical analysis of the pre and postintervention fall rates was conducted
to establish if the changes were statistically significant
(a ¼ 0.05). As this work was consistent with a quality
improvement project and was not considered research, the
work was exempt from Institutional Board Review.
At the start of this project, a fall prevention task force
was convened, composed of emergency management, emergency nurses, and other emergency staff. The objectives of
the task force were to (1) conduct a comprehensive review
of adult ED falls to identify trends, (2) identify the EDspecific gaps/needs related to fall prevention, and (3) implement interventions to reduce ED falls and FWI. The team
worked with a clinical nurse specialist (CNS) with expertise
in patient safety and fall prevention to provide consultation
and to assist with the improvement initiative.
The ED fall prevention task force met biweekly in addition to several small group meetings with ED clinical staff to
better understand the unique challenges and current gaps in
the ED fall prevention practices. An in-depth, retrospective
review of fall data helped the task force to better identify the
specific characteristics of the ED falls (eg, age, sex, diagnosis,
timing, fall risk). A tour of the adult emergency department
was conducted with the task force and CNS consultant to
review concerns and challenges related to the environment
and workflow. These findings were summarized and
presented to ED leadership, the system falls committee,
and adult ED nursing staff as an initial step to generate interest in the topic and encourage engagement from clinical
staff.
The key fall prevention barriers identified by the task
force were (1) departmental volume and acuity, (2) unit
layout and flow (eg, line of sight, shared bathrooms), and
(3) the lack of an ED-specific fall risk assessment tool. In
addition to these challenges, the task force recognized the
need for a strong fall prevention culture on the unit, backed
by an engaged and supportive leadership team. Given the
varied factors identified and recommendations found in
fall prevention literature, a multifactorial approach to
preventing falls, including risk assessment, unit culture,
electronic medical record (EMR) updates, and novel patient
monitoring interventions was initiated.
Interventions
DATA REVIEW
Although chart review and data analysis may not be thought
of as an intervention, the success of this fall prevention
initiative depended heavily on the initial analysis of ED falls
data. Though many patients are identified as being “high
risk” for falling, the reality is that most do not fall. Having
additional information may aid staff to better predict falls.
668 JOURNAL OF EMERGENCY NURSING VOLUME 46 ISSUE 5 September 2020
PRACTICE IMPROVEMENT/Cook et al
In addition to establishing baseline fall rates for the adult
emergency department, chart reviews were completed to
identify common factors and trends related to the falls.
Where appropriate, findings were used to guide changes
to care and operations based on identified factors. For
example, although no significant time-of-day trends were
noted, it was found that a disproportionate number of
intoxicated patients had fallen, which highlighted the need
for changes to the monitoring and bed assignment for these
patients.
FALL RISK ASSESSMENT
During the retrospective chart review it was calculated that
the Morse Fall Scale (MFS)13 had a specificity of 91% (low
rate of false negatives) but only a 23% sensitivity (high fall
risk score and went on to fall). Given the poor sensitivity of
the MFS for patients in this emergency department, a literature search was conducted by the CNS to identify available
ED-specific fall risk assessment tools. At the time, only 2 validated fall screening tools had been created for the ED environment—the KINDER 18 (Figure 2) and the Memorial
ED Fall Screening Tool.9 There were no published studies
reporting the sensitivity or specificity of either tool, but adult
ED leadership and key clinical staff reviewed both tools for
potential use in the organization and the KINDER 1 was
selected as a potential replacement for the MFS for adult patients in emergency departments across the system.
REMOTE VIDEO MONITORING
Although still a relatively new technology, video monitoring
has been effective in reducing falls in the inpatient
setting.14-17 Remote video monitoring (RVM) was
FIGURE 2
KINDER 1 fall risk assessment tool, reprinted with permission from Alexander et al.22
September 2020 VOLUME 46 ISSUE 5 WWW.JENONLINE.ORG 669
Cook et al/PRACTICE IMPROVEMENT
identified as a potential means to address the department’s
challenging layout and to assist with the high volume and
acuity. The organization had recently introduced 36
RVM devices to the inpatient setting, and following
discussions with nursing leadership, 2 were assigned to the
adult emergency department. These devices (Avasys
TeleSitter, Avasure, Belmont, MI) were nonrecording,
mobile video cameras that were monitored remotely by
hospital staff (max 12:1 monitor-to-staff ratio). The devices
had the capability to communicate directly with the patient
or staff via a speaker in addition to the ability to sound a
local alarm if needed. RVM could be initiated without
physician order on any patient that was identified at high
risk for falling. The 2 RVM devices allowed for monitoring
400 to 450 patients per month (20%-25% of high fall-risk
patients). Patient selection was based on the KINDER 1
score along with evaluation of the clinical team and nursing
leadership and consideration of patient risk factors (eg, age,
diagnosis, bed placement), staffing, and department acuity.
EXIT ALARM TECHNOLOGY
Bed (or stretcher) alarms are an intervention with mixed fall
prevention effectiveness18 and can contribute to alarm fatigue.19 However, with the bay-based nature of the ED
layout and challenges with patient monitoring, bed alarms
were identified as a potential benefit for a comprehensive
fall prevention program. At the time of this improvement
project, the organization was in the process of purchasing
new stretchers (Stryker Prime Series, Stryker, Kalamazoo,
MI) and integrated bed exit alarm technology was a feature
that was considered and ultimately purchased. Following
equipment introduction and staff education, bed alarms
were then incorporated into the plan of care for those patients identified as a fall risk during triage.
FALL PREVENTION CULTURE
Although a successful fall prevention program requires policies
and resources, it is the engagement, knowledge, and commitment of the staff and leadership that will sustain improvement.
There were a variety of actions that were used to engage staff
and generate buy-in and excitement regarding fall prevention.
One activity was the creation of the “Catch a Falling Star”
recognition program to recognize staff who had played a vital
role in preventing a patient fall or injury. Staff were nominated
by their peers or management team and were recognized on a
unit display board, presented with a certificate, and given a
small token of appreciation. In addition, milestone celebrations were held on the unit and signage that indicated the
date of the last fall event was placed in the department and
visible to staff, patients, and families.
COMMUNICATION
Poor communication is one of the most common contributors to adverse patient safety events.20 Documentation of
the KINDER 1 began the communication about fall risk
between the triage nurse and the primary nurse and triggered a fall prevention plan of care. In addition, changes
were made to the existing EMR to better communicate
fall risk and fall history. Flowsheets were created for documenting the specifics of a patient fall, which then triggered
an alert within the EMR in the form of a red banner across
the top of the chart alerting staff that the patient had fallen
during their current encounter. On subsequent encounters,
the alert would populate to notify staff that the patient had
fallen during a previous encounter.
Results
With regard to fall risk assessment, screening of patients was
consistent at more than 95% compliance using both the
MFS (preimprovement) and the KINDER 1 (postimprovement). These assessments were all completed during the
initial triage or by the primary registered nurse in the case
of a direct ED admission. During the year following the
change to KINDER 1, more than 80,000 patients were
assessed with KINDER 1 with 31% reported as positive
(ie, high fall risk). This was compared to the 10% of MFS
screens that resulted in a high-risk assessment. Using these
results and the patient fall data, the organization’s KINDER
1 sensitivity was calculated at 68% (pre-MFS ¼ 23%) and
the specificity 68% (pre-MFS ¼ 91%).
Outcome measures included both the adult ED fall
rate (adult ED falls per 1,000 patient visits) and the adult
ED FWI rate (adult ED injury falls per 1,000 patient
visits). Baseline data were reported for the 8 quarters
(January 2015 to September 2016) before the project’s
implementation. The implementation timeframe was
identified as January 2017 to June 2017. The postimplementation time period was identified as July 2017 to June
2019. A t test (equal variance, 2-tailed) was conducted on
the pre- and postintervention data (fall rate and FWI rate)
to determine the statistical significance (a ¼ 0.05) of the
changes in fall rates following the improvement program.
Figure 3 displays the results of the adult ED fall rate
over these time frames. Figure 4 displays the adult ED injury
fall rate for the same time. Following implementation of the
670 JOURNAL OF EMERGENCY NURSING VOLUME 46 ISSUE 5 September 2020
PRACTICE IMPROVEMENT/Cook et al
improvement project, the fall rate decreased from 0.73 falls
per 1,000 visits (pre) to 0.55 falls per 1,000 visits (post),
representing a 25% decrease (t ¼ 1.41, P ¼ 0.18). The
injury rate decreased from 0.09 FWI per 1,000 visits (pre)
to 0.03 FWI per 1,000 visits (post), which was a 66%
decrease in injuries (t ¼ 2.29, P < 0.05). These decreases
represented 27 fewer falls and 10 fewer injuries over the
24-month postimplementation period despite a 3% increase
in adult ED volume over this time frame.
Discussion
As previously noted there has been a distinct lack of literature related to ED-specific fall prevention programs and interventions. Available fall risk assessment tools have either
been inpatient-specific (not been validated for use in the
emergency department) or ED-specific tools that have not
been fully validated. As a result of this lack of evidence,
this patient safety initiative incorporated interventions
that were previously studied for use in the inpatient setting.
However, by incorporating these interventions into a multifactorial fall prevention program to address the unique
nuances of the ED setting we hope to fill this need for
ED-specific fall prevention literature.
Understanding the underlying trends and patient safety
gaps found in the fall events was a crucial initial step in this
initiative. This knowledge laid the foundation for an effective fall prevention program. Once the key implicating factors were identified, interventions were chosen from the
existing literature and implemented.
The use of RVM, although not new to the organization,
was a new approach in the emergency setting. The
utilization of RVM provided nurses with an additional
means of ensuring that high-risk falls patients were still
visibly monitored and communicated with, and that the
nurse was alerted when needed. Use of RVM was encouraged for patients determined to be at risk for falling, such
as intoxicated or altered patients or any patient who might
benefit from continuous monitoring. If all RVMs were in
use, exit alarms were available on all ED stretchers as were
constant-observer staff, providing nurses with multiple
monitoring strategies.
Implementation of the KINDER 1 not only enabled staff
to assess patients with an ED-specific tool, but also contributed to the overall fall prevention culture. Staff appreciated
the fact that the KINDER 1 allowed for nursing judgment,
and it reinforced the belief that staff were empowered to assess
for and prevent patient falls. By selecting “nursing judgment”
on the KINDER 1, triage staff were able to identify patients
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
Q2
FY15
Q3 Q4 Q1
FY16
Q2 Q3 Q4 Q1
FY17
Q2 Q3 Q4 Q1
FY18
Q2 Q3 Q4 Q1
FY19
Q2 Q3
Fall Rate (events/1000 visits)
Fall Rate Mean Falls Rate
Pre-implementaon Intervenon Post-implementaon
FIGURE 3
Adult ED fall rates pre and post improvement. Rate ¼ falls/(adult ED visits/1000). Improvement activities included: ED task force (Q2-17), data analysis (Q2-17), staff recognition program (Q2-17), KINDER 1 tool implementation (Q2-17), remote video in emergency department (Q3-17), and new ED stretcher with bed alarms (Q3-17). Q1,
quarter 1; Q2; quarter 2; Q3; quarter 3; Q4, quarter 4; FY, fiscal year.
September 2020 VOLUME 46 ISSUE 5 WWW.JENONLINE.ORG 671
Cook et al/PRACTICE IMPROVEMENT
