Open access Quality improvement report

Reducing time to admission in

emergency department patients: a cross-­
functional quality improvement project
Bryan Imhoff ‍ ‍,1 Kenneth Marshall,1 Niaman Nazir,2 Aroop Pal,3
Melissa Parkhurst3

To cite: Imhoff B, Marshall K, ABSTRACT

Nazir N, et al. Reducing time WHAT IS ALREADY KNOWN ON THIS TOPIC
Crowding and boarding are common issues facing
to admission in emergency ⇒ Studies have demonstrated that rapid admission
emergency departments (EDs) in the USA. These
department patients: a cross-­ protocols can reduce emergency department (ED)
issues have negative effects on efficiency, patient care,
functional quality improvement length of stay and boarding.
project. BMJ Open Quality satisfaction and healthcare team well-­being. Data from
an audit of the admissions process at a large, urban,
2022;11:e001987. doi:10.1136/ WHAT THIS STUDY ADDS
bmjoq-2022-001987 academic US ED demonstrated a lengthy process,
exceeding national benchmarks in both length of stay and ⇒ This study focused on the handoff and trans-
boarding of admitted patients. fer of care between the ED and internal medicine
Received 17 May 2022
We performed a pre–post study between July 2019 and hospitalists.
Accepted 25 August 2022
July 2021 focused on the first step of the admission HOW THIS STUDY MIGHT AFFECT RESEARCH,
process at our institution, the time to bed request. All PRACTICE OR POLICY
patients admitted to an internal medicine (IM) floor team
⇒ The study illustrates that process and systems
from the ED were included in the study. The primary
outcome was the time from decision to admit by the changes coupled with education and improved re-
emergency medicine physician to placement of the bed porting can result in sustained improvement.
request order by the IM physician. Quality improvement
(QI) occurred in three phases: an initial preintervention
process and electronic health record change to better benchmark by 72 min for similarly sized
capture admission times, a primary intervention focused EDs (institution 473 min, EDBA benchmark
on process change and provider education and a second
401 min).1 Furthermore, boarding time
intervention focused on improvements to provider
exceeded the EDBA 50% benchmark by
communication.
During the study period, 25 183 patients were admitted 44 min (institution 202 min, EDBA bench-
to IM floor teams and met inclusion criteria. Prior to the mark 158 min).
primary intervention, the mean time from ED decision Several definitions for ED boarding exist.
to admit to IM placement of the bed request order was For the purposes of this project, we have
75.1 min. Postintervention, the mean time decreased adopted the standard definition used in
to 39.7 min, a statistically significant improvement of the USA, that boarding starts at the time of
35.4 min (p value <0.0001). decision to admit and ends at departure. 2 3
© Author(s) (or their This QI project demonstrates the ability of interventions
employer(s)) 2022. Re-­use The admission process at our institu-
to reduce the time to admission bed request order, a key tion has three major steps (figure 1). The
permitted under CC BY-­NC. No
step in the overall admission process and a contributor to
commercial re-­use. See rights first step, and the subject of this quality
and permissions. Published by boarding at our institution. In making process changes,
improvement project, starts when an ED
BMJ. the team also reduced provider handoffs and improved
1 provider communication. provider decides to admit a patient and
Department of Emergency
Medicine, University of Kansas
requests admission from IM. This process
School of Medicine, Kansas City, step ends with IM placing admission and
Kansas, USA PROBLEM bed request orders.
2
Department of Population Admitting patients from the emergency Prior to the intervention described
Health, University of Kansas department (ED) to the internal medicine further, ED providers had requested admis-
Medical Center, Kansas City,
KS, USA
(IM) floor teams at our institution is a lengthy sion via a numeric page sent to a central IM
3
Department of Internal process composed of multiple steps with physician. The IM physician would return
Medicine, University of Kansas different stakeholders. This likely contributes the page with a phone call, discuss admis-
School of Medicine, Kansas City, to the institution underperforming bench- sion details with the ED provider, then
Kansas, USA marks in both length of stay and boarding accept or reject the admission. Accepted
Correspondence to times. Length of stay for admitted patients admissions would be assigned to members
Dr Bryan Imhoff; exceeded the 2020 Emergency Depart- of a rotating pool of IM physicians who
​bimhoff@​kumc.​edu ment Benchmarking Alliance (EDBA) 50% would evaluate the patient and ultimately

Imhoff B, et al. BMJ Open Quality 2022;11:e001987. doi:10.1136/bmjoq-2022-001987 1

Open access

Figure 1 ED to floor admission process. ED, emergency department; QI, quality improvement.

place admission orders. An initial process audit iden- hallway beds for boarding patients and providing consis-
tified the lack of reliable data as a primary obstacle for tent ancillary services during weekends and off-­hours.
quality improvement. Specifically, the manual paging Less resource intensive changes have also been studied,
process did not generate a systematically captured including active bed management and prioritising early
timestamp for the ED decision to admit (time of ED discharges.5 7 17 The persistence of crowding despite
request to admit). The first improvement cycle aimed these well-­understood causes and proposed solutions
to modify the process and systems to collect this demonstrates a need for continued strategies to relieve
important data point. access block, particularly using strategies with low-­
Aided by the newly captured data outlined previously, resource costs. To our knowledge, few studies have been
the team discovered that, on average, 43% of the time undertaken to speed the process for transfer of care from
IM-­admitted patients spent in the ED could be considered ED to inpatient teams. Though those studies have shown
boarding (occurring after the ED provider’s decision to potential improvements in ED length of stay (LOS),
admit). Over one-­third of that time (77 min or 16% of time to admission orders and ED boarding time.20 21 The
total time in the ED) occurred after an ED decision to efforts described here were an attempt to expedite these
admit but before IM admission and bed request orders processes, which represent a limited component of the
had been placed. The primary intervention aimed to access block problem over which clinicians can exert
decrease overall IM admission process time by reducing direct control.
the mean time from ED admission request to IM bed
request order from 77 to 30 min.
MEASUREMENT
BACKGROUND The first improvement cycle of this project focused on
Our organisation’s struggle with ED crowding and implementing a new admission request order through
boarding are not unique. While the national ED census Epic (Epic Systems Corporation), our EHR, with asso-
dipped during 2020, the initial year of the COVID-­19 ciated process changes to enable systematic capture of
pandemic, that trend has subsequently reversed, exac- the time of ED decision to admit. As such, no reliable
erbating factors associated with crowding, including baseline data existed prior to this improvement.
increasing acuity, percentage of patients being admitted, Baseline data for a detailed process audit were collected
length of stay and boarding times.1 4 These findings via retrospective data extraction from the health system
are consistent with the reported experience of emer- EHR. All patients seen in the ED during a 6-­month period
gency physicians that ED boarding and crowding has (1 January 2019–30 June 2019) and ultimately admitted to
continued to intensify over the years of the pandemic.5 6 IM were included in the baseline. The baseline included
This crowding is problematic due to its negative effects 6451 patients who met the inclusion criteria. Of these
on patient care, patient satisfaction and well-­being of the patients, less than 2% (126) were excluded due to errors
healthcare team.7–16 in data captured in the EHR. These included missing time
Over the last three decades, multiple studies have both stamps and/or negative durations. The major outcome
investigated the causes of ED crowding and suggested variable for this quality improvement study was time to
potential solutions; these focus on one of the three bed request order (IM bed request order timestamp – ED
major determinants of crowding, including input, request to admit order timestamp). Additional patient
throughput and output of patients. Of these, sluggish and encounter data were collected, specifically age,
output of patients to inpatient beds (also known as ‘inpa- gender, ED acuity (Emergency Severity Index) and hour
tient access block’) is widely recognised to be the largest of presentation to the ED.
contributing factor to ED crowding.7 16–19 A variety A detailed analysis of this dataset demonstrated 16%
of tactics to improve ED output have been studied, of total time in the ED (77 min) occurred after the ED
including structural changes such as smoothing surgical decision to admit but before admission and bed request
operating schedules, mandating the use of inpatient orders had been placed by IM.

2 Imhoff B, et al. BMJ Open Quality 2022;11:e001987. doi:10.1136/bmjoq-2022-001987

 Open access

DESIGN Admitting providers shared feedback that they were

The design of this QI project relied on a three-­ part not prioritising placing the admit and bed request orders
approach: early, preferring instead to interview and examine patient
1. Improve process-­related data collection. before completing admit and bed request orders in a
2. Undertake detailed process analysis and identify op- batch along with all the other orders needed to provide
portunities for improvement. care for the patient. Prior to our intervention, there was
3. Implement resultant process, systems and people no clear standard or goal for placement of these orders,
changes. and no monitoring, sharing or incentivising performance.
The QI team consisted primarily of physician stake- As a result, for the first intervention, our team focused
holders from the EM and IM departments and included on provider education and performance tracking. IM
experts in systems/EMR and operations. This project leadership reviewed the process, obtained feedback from
enjoyed support from departmental leadership at the the admitting teams and then established a target goal.
highest level. The two departments partnered on the Through mandatory meetings and educational mate-
initial EHR-­related step. The ED team led the subsequent rials, all hospitalists trained on the changes. The Plan-­
data collection, process audit and analysis. Finally, IM led Do-­Study-­Act (PDSA) cycle identified opportunities to
the implementation of process, education and reporting simplify the process. The team made process changes
changes. to reduce information handoff among IM providers.
The implementation team met on an ad hoc basis After this intervention, the central IM physician would
throughout the project to monitor results and plan the quickly chart review the proposed admission to ensure
next steps. no contraindication for admission to IM, and then assign
to a second IM physician (without discussion with the
STRATEGY ED provider) who would examine and admit the patient
Our goal with this QI project was to reduce the time to in addition to first calling the ED provider to receive
bed request order. To do so, we undertook three inter- patient handoff/checkout. Examples of contraindica-
vention cycles. tions included patients owned by a different service and
patients unstable for floor admission.
Preintervention improvement
Our initial process change was primarily designed to Intervention 2
capture the time of ED admission in our EHR. This times- Based on provider feedback, for our final intervention,
tamp allowed detailed process analysis to drive subse- we focused on better closing the communication loop
quent interventions. This change was realised via a new between EM and IM providers. Specifically, after the
EHR order with an associated system-­logged timestamp. central IM physician received an admission page, he or
This new order automatically triggered a page to the she would now message both the requesting EM provider
central IM physician that included standardised informa- and second (admitting) IM physician via a system that
tion including patient information, ordering EM provider enabled Health Insurance Portability and Accountability
name and provider phone number. The inclusion of this Act- (HIPAA-) compliant encrypted group messaging
standardised information represents the second process-­ (Voalte Messenger, Hill-­ Rom Holdings). This message
related improvement within this intervention. Its intent would include the names of each provider and facilitate
was to improve challenges ED and IM providers shared in easier communication/follow-­up in the event of a delay or
connecting to one another to discuss a requested admis- missed phone call. By including the IM triage physician,
sion, as well as to facilitate more efficient handoffs by admitting hospitalist and EM provider in a group text,
giving IM providers basic information about the patient this facilitated closed-­loop communication among them
in advance of the handoff conversation with the EM team. that the admit request had been received and assigned,
as well as opening an additional channel for communica-
Intervention 1
tion about the patient. These benefits were widely viewed
Analysis of data made possible by the change described
as being worth the extra effort for the triage provider.
previously demonstrated a lengthy admission process, in
particular, the time to IM bed request order. Interestingly,
there was wide variability among IM physicians in their
efficiency in placing bed request orders. The fastest quar- RESULTS
tile was 77 min faster in placing the bed request order The primary outcome variable for this QI project was
than the slowest quartile. This implied the possibility of time to bed request order (ie, the time elapsed between
sharing best practices between quartiles and other poten- the initial ED request to admit order timestamp and the
tial training-­related efficiency gains. There was also varia- subsequent IM bed request order timestamp). This vari-
bility between our academic IM services (those with resi- able was measured at the patient encounter level and
dents) and private IM services (those without). Academic averaged for each period of interest.
services took on average 13 min longer to place the bed Preintervention process changes enabled systematic
request order. capture of time of ED request for admission in the EMR.

Imhoff B, et al. BMJ Open Quality 2022;11:e001987. doi:10.1136/bmjoq-2022-001987 3

Open access

Figure 2 Control chart, time from ED admission request to IM bed request order, July 2019–July 2021. ED, emergency
department; IM, internal medicine.

Therefore, prior to these changes, we could not measure control limits and a ‘steady state’ period with a mean of
our variable of interest. 36.3 min and relatively narrower control limits.
Intervention 1 represented our largest intervention and Subanalysis of IM admitting physicians demonstrated
was conducted during July 2020. One year of data both that all quartiles demonstrated statistically significant
preintervention and postintervention 1 was collected and improvement (p value <0.0001) and that the slowest
analysed for this manuscript. A total of 13 263 encoun- admitters improved by larger absolute amount. Interest-
ters occurred preintervention and 11 920 encounters ingly, all admitters improved by approximately the same
postintervention. The primary outcome variable (mean) relative amount (45%–54% improvement) (figure 3).
decreased from 75.1 min preintervention to 39.7 min Likewise, subanalysis of IM admitting physicians by
postintervention, a statistically significant improvement service type demonstrated that both service types demon-
of 35.4 min (p value <0.0001) (figure 2). strated statistically significant improvement (p value
Based on the control chart, the postintervention 1 <0.0001) with near-­ identical mean absolute improve-
period appeared to have two distinct phases: an ‘improve- ments of 35.7 min. Interestingly, the mean difference in
ment’ phase with a mean of 47.2 min and relatively wider admitting performance between academic and private

Figure 3 Time from ED admission request to IM bed request order by quartile, July 2019–July 2021. ED, emergency
department; IM, internal medicine.

4 Imhoff B, et al. BMJ Open Quality 2022;11:e001987. doi:10.1136/bmjoq-2022-001987

 Open access

services was 13.2 min both before and after intervention providers are hired, these detailed are covered both
1. This number appears to represent the efficiency cost during their orientation and prior to their first shift. IM
of residents performing admissions at our academic leadership monitors individual and group performance.
institution. These factors have all helped to sustain this improvement
Intervention 2 represented a small intervention over time.
and occurred on 1 April 2021. While this intervention Having separate bed request and care orders enabled
improved communication between EM and IM physi- much of the improvement realised in this project. At our
cians, the control chart demonstrated no obvious change organisation, IM hospitalists place both the bed request
in primary outcome variable. and patient care orders. Postintervention, in many cases,
hospitalists placed the bed request order immediately
Lessons and limitations after speaking with their emergency medicine colleague,
The project aim was to improve the time to admission allowing the hospital bed placement team to work in
for ED patients at the handoff point between EM and parallel with the hospitalist who then interviews and
IM. As such, partnership was a necessary key to success. examines the patient before placing care orders. This
Role definition and delineation within this multidiscipli- parallel processing and quicker bed placement contrib-
nary team was equally important. Boarding of admitted uted to the throughput gains described in this paper but
patients affects the ED more directly than IM. As a result, in rare situations resulted in the need to resubmit a bed
the EM team played the larger role in the initial data request when patients needed a higher (or lower) level of
collection, process audit and analysis necessary to develop care at the time the patient was examined by the hospi-
the problem statement. In addition, the EM team devel- talist. Though not measured directly in this study, this
oped an initial set of high-­level recommendations. As the circumstance was rare.
organisation responsible for the hospitalists who place the The project demonstrates that interventions initially
admission and bed request orders, the IM team played the conceived to improve easily measured process metrics
larger role in development of detailed recommendations also provide opportunities for unmeasured positive
as well as the implementation of these process, education impacts on patient safety. This project focused on the
and reporting changes. The collaboration between EM time to bed request order. In making process changes to
and IM partners on this project provided valuable insights deliver improvement in this metric, the team also made
into efficiency and how communication and care quality changes to reduce the number of provider handoffs
could be enhanced by careful design and implemen- by changing the process to have admitting hospitalists
tation of our interventions. Our work, including timely contact the admitting ED physician directly, rather than
support from our informatics team, was supported at the through a centralised attending on duty. In addition,
highest level by senior health system and departmental the team improved provider communication by imple-
leadership committed to driving change. menting a system-­generated page to IM that included
This project reaffirmed the management adage, ‘if you more consistent, richer information than the previous
can't measure it, you can't improve it’. Prior to the initial free-­form and typically numeric-­ only pages. Further-
process and EMR changes, the organisation did not accu- more, after positive feedback and constructive sugges-
rately capture the time of ED decision to admit. This lack tions from ED and hospitalist providers, communication
of data prevented the ED from measuring the time to via a closed-­loop group message was introduced as our
bed request order. The success of this project required an third intervention. These changes to reduce handoffs
initial systems change to track and measure this data. This and improve provider communication no doubt contrib-
was foundational, providing justification for the subse- uted to increased patient safety and were well received by
quent interventions that delivered measurable process our providers.
improvement. In terms of limitations, this project resulted in a signifi-
Furthermore, the success of this QI project rested cant reduction in the time to bed request order. However,
largely in convincing individual providers to change their while IM is the largest admitter from our ED, they only
behaviour, which can be exceedingly difficult. This team account for approximately 60% of our ED admission
felt that our success in doing so rested first on explaining volume. Additional interventions will need to be designed
the reason for change. This required providing hard and implemented to apply to other services to maximise
data to illustrate an opportunity, but also concretely potential benefits from this type of intervention, though
demonstrating why change matters to the patient and the this project does provide a roadmap for success.
organisation. Second, the team felt that transparency in Due to competing organisational priorities, this project
performance was necessary to motivate individual change. suffered a lengthy delay prior to intervention 1. In addi-
Specifically, the team provided data to individual hospital- tion, while the initial process audit and data analysis was
ists showing them how they compare with their peers and completed prior to the COVID-­19 pandemic, the major
continues to share individual and group performance as interventions (interventions 1 and 2) occurred during the
part of the monthly scorecard. This scorecard includes changing phases of the pandemic with unknown effects
their personal performance, average department perfor- on project outcomes. The institution experienced signif-
mance and the departmental goal of 30 min. As new icant swings in ED patient arrivals and hospital census, a

Imhoff B, et al. BMJ Open Quality 2022;11:e001987. doi:10.1136/bmjoq-2022-001987 5

 Open access

significant confounder in measuring this project’s effect and license their derivative works on different terms, provided the original work is
on LOS and boarding times. properly cited, appropriate credit is given, any changes made indicated, and the use
is non-­commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
This project focused primarily on reducing the time
to bed request and admission orders. We cannot know ORCID iD
whether this pressure to admit quickly changes the quality Bryan Imhoff http://orcid.org/0000-0002-2515-6226
of care provided by admitting hospitalists or whether
this pressure changes the quality of resident–faculty and
student–faculty education on the admitting teams. REFERENCES
1 Directors TEBAB. 2020 Annual report. Madison, WI ED
This QI project focused on improving the time from Benchmarking Alliance; 2021. www.edbenchmarking.org
decision to admit to placement of a bed request. At our 2 Wiler JL, Welch S, Pines J, et al. Emergency department
institution, these steps are completed by different physi- performance measures updates: proceedings of the 2014 emergency
department benchmarking alliance consensus summit. Acad Emerg
cians. EM drives the initial decision to admit while IM ulti- Med 2015;22:542–53.
mately places the bed request. At other institutions, EM 3 American College of Emergency Physicians. Definition of Boarded
patient, 2018.
both initiates the decision to admit and enters the bed 4 Lucero A, Sokol K, Hyun J, et al. Worsening of emergency
request order, limiting the generalisability to all hospitals. department length of stay during the COVID-­19 pandemic. J Am Coll
Finally, this was a single academic centre, before–after Emerg Physicians Open 2021;2:e12489.
5 Kelen GD, Wolfe R, D’Onofrio G, et al. Emergency department
study with specific improvements enabled by our EHR crowding: the Canary in the health care system. 26, 2021.
and secure group messaging platform which may also 6 Davis J. Emergency Department “Boarding” At Seemingly All-­Time
High Levels Regs and Eggs: ACEP Regulatory Affairs Blog; 2021.
limit generalisability. https://www.acep.org/federal-advocacy/federal-advocacy-overview/​
regs-eggs/regs-eggs-articles/regs-eggs-august-26-2021/ [Accessed
10 Mar 2022].
CONCLUSION 7 Morley C, Unwin M, Peterson GM, et al. Emergency department
crowding: a systematic review of causes, consequences and
Through three improvement cycles, the multidisciplinary solutions. PLoS One 2018;13:e0203316.
project team was able to reduce the time to admission bed 8 Moskop JC, Geiderman JM, Marshall KD, et al. Another look at the
persistent moral problem of emergency department crowding. Ann
request order, a key step in the overall admission process. Emerg Med 2019;74:357–64.
In making process changes, the team also reduced 9 Epstein SK, Huckins DS, Liu SW, et al. Emergency department
provider handoffs and improved provider communica- crowding and risk of preventable medical errors. Intern Emerg Med
2012;7:173–80.
tion. To sustain this improvement, the team added admis- 10 Pines JM, Pollack CV, Diercks DB, et al. The association between
sion process cycle time to individual monthly perfor- emergency department crowding and adverse cardiovascular
outcomes in patients with chest pain. Acad Emerg Med
mance scorecards, with tracking of trends and compar- 2009;16:617–25.
ison with the entire group. As a result, this improvement 11 Mills AM, Shofer FS, Chen EH, et al. The association between
has already been sustained for 12 months. While IM emergency department crowding and analgesia administration in
acute abdominal pain patients. Acad Emerg Med 2009;16:603–8.
admissions comprise most admissions from the ED at our 12 Ward MJ, Baker O, Schuur JD. Association of emergency
institution, other services account for a material number department length of stay and crowding for patients with ST-­
elevation myocardial infarction. West J Emerg Med 2015;16:1067–72.
of admissions. This project can serve as a framework 13 Bernstein SL, Aronsky D, Duseja R, et al. The effect of emergency
for similar improvements within our institution (and department crowding on clinically oriented outcomes. Acad Emerg
Med 2009;16:1–10.
beyond). At our institution, further work is being done to 14 Wang H, Kline JA, Jackson BE, et al. The role of patient perception
extend these results to the next most frequent ED admis- of crowding in the determination of real-­time patient satisfaction at
sion service, family medicine. emergency department. Int J Qual Health Care 2017;29:722–7.
15 Fernandez-­Parsons R, Rodriguez L, Goyal D. Moral distress in
emergency nurses. J Emerg Nurs 2013;39:547–52.
Contributors BI performed the initial data gathering and preliminary analysis. 16 Hospital Emergency Departments. Crowding continues to occur,
AP and MP designed and oversaw the interventions. NN provided comprehensive and some patients wait longer than recommended time frames
statistical analysis. BI drafted the manuscript, and all authors contributed Government Accountability Office; 2009. https://www.gao.gov/​
substantially to its revision. BI takes responsibility for the manuscript as a whole. products/gao-09-347 [Accessed 11 May 2022].
17 Farley HL, Kwun R. Emergency department crowding: high impact
Funding The authors have not declared a specific grant for this research from any solutions. emergency medicine practice Committee of the American
funding agency in the public, commercial or not-­for-­profit sectors. College of emergency medicine; 2016. https://www.acep.org/​
globalassets/sites/acep/media/crowding/empc_crowding-ip_092016.​
Competing interests None declared. pdf [Accessed 07 Oct 2019].
Patient and public involvement Patients and/or the public were not involved in 18 Forster AJ, Stiell I, Wells G, et al. The effect of hospital occupancy on
the design, or conduct, or reporting, or dissemination plans of this research. emergency department length of stay and patient disposition. Acad
Emerg Med 2003;10:127–33.
Patient consent for publication Not applicable. 19 Dunn R. Reduced access block causes shorter emergency
department waiting times: an historical control observational study.
Provenance and peer review Not commissioned; externally peer reviewed. Emerg Med 2003;15:232–8.
Data availability statement No data are available. 20 Quinn JV, Mahadevan SV, Eggers G, et al. Effects of implementing a
rapid admission policy in the ED. Am J Emerg Med 2007;25:559–63.
Open access This is an open access article distributed in accordance with the 21 Amarasingham R, Swanson TS, Treichler DB, et al. A rapid admission
Creative Commons Attribution Non Commercial (CC BY-­NC 4.0) license, which protocol to reduce emergency department boarding times. Qual Saf
permits others to distribute, remix, adapt, build upon this work non-­commercially, Health Care 2010;19:200–4.

6 Imhoff B, et al. BMJ Open Quality 2022;11:e001987. doi:10.1136/bmjoq-2022-001987