Content uploaded by Stephanie Carlisle
Author content
All content in this area was uploaded by Stephanie Carlisle on Apr 11, 2019
Content may be subject to copyright.
Nonresearch Article
Development of a Novel
Emergency Department
Mapping Tool
Matthew Grzywinski, BS
1
, Stephanie Carlisle, MArch
2
,
James Coleman, BArch
2
, Connor Cook
2
, Geoffrey Hayden, MD
3
,
Robert Pugliese, PharmD
4
, Billie Faircloth, MArch
2
,
and Bon Ku, MD, MPP
3
Abstract
Objectives: Develop a built environment mapping workflow. Implement the workflow in the
emergency department (ED). Demonstrate the actionable representations of the data that can be
collected using this workflow. Background: The design of the healthcare built environment impacts
the delivery of patient care and operational efficiency. Studying this environment presents a series of
challenges due to the limitations associated with existing technology such as radio-frequency identi-
fication. The authors designed a customized mapping workflow to collect high-resolution spatial,
temporal, and activity data to improve healthcare environments, with emphasis on patient safety and
operational efficiency. Method: A large, urban, academic medical center ED collaborated with an
architecture firm to create a data collection, and mapping workflow using ArcGIS tools and data
collectors. The authors developed tools to collect data on the entire ED, as well as individual patients,
physicians, and nurses. Advanced visual representations were created from the master data set.
Results: In 48 consecutive hourly snapshots, 5,113 data points were collected on patients, physicians,
nurses, and other staff reflecting the operations of the ED. Separately, 84 patients, 10 attending
physicians, 10 resident physicians, and 17 nurses were tracked. Conclusions: The data obtained from
this pilot study were used to create advanced visual representations of the ED environment. This cost-
effective ED mapping workflow may be applied to other healthcare settings. Further investigation to
evaluate the benefits of this high-resolution data is required.
Keywords
mapping, workflow, hospital, health, built environment, emergency department, tool
1
Sidney Kimmel Medical College, Thomas Jefferson University, Philadelphia, PA, USA
2
KieranTimberlake, Philadelphia, PA, USA
3
Department of Emergency Medicine, Thomas Jefferson University, Philadelphia, PA, USA
4
College of Pharmacy, Thomas Jefferson University, Philadelphia, PA, USA
Corresponding Author:
Matthew Grzywinski, BS, Sidney Kimmel Medical College, Thomas Jefferson University, Health Design Lab, 925 Chestnut St.,
Philadelphia, PA 19107, USA.
Email: matthew.grzywinski@jefferson.edu
Health Environments Research
& Design Journal
1-13
ªThe Author(s) 2019
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/1937586719842349
journals.sagepub.com/home/her
Background
Since its introduction in the mid-20th century, the
emergency department (ED) has played a sub-
stantial role in the delivery of healthcare. The
modern ED specializes in treating patients with
complex and potentially life-threating conditions
and increasingly serves as the gateway to inpati-
ent care (Morganti et al., 2013). Based on 2014
data, there were more than 140 million visits to
the ED, with over 12 million visits resulting in
admission to the hospital (Centers for Disease
Control and Prevention, 2018). Moreover, the
ED serves as a social safety net for underserved
members of society (Morganti et al., 2013).
Reimbursement for emergency care is deter-
mined, to some extent, by healthcare quality mea-
sures, which tend to reflect clinical efficiency.
Quality measures for the ED include metrics such
as median time from ED arrival to ED departure
for discharged ED patients, door to diagnostic eva-
luation by a qualified medical professional, left
without being seen rate, and median time to fibro-
lytics. The U.S. Centers for Medicare and Medi-
caid Services (CMS, 2018) publishes ED quality
measurement data online to allow patients to make
informed decisions about their healthcare.
Because of the emphasis on efficiency, ED design
and workflow are constantly evolving. Current trends
in design include a unidirectional patient flow model,
physician or team triage, split workflow, flexible
capacity to meet demands based on patient census,
private patient rooms, largestaffworkstations,and
small nurse stations spread throughout the ED (Saba
& Bardwell, 2004; Welch, 2011, 2012). Quantitative
models to describe and predict how individuals inter-
act with their environment, including space syntax
analysis and discrete event simulations, have also
been studied (Morgareidge, Cai, & Jia, 2014).
Despite advances in ED design and predictive
modeling, there is no cost-effective and intuitive
method that uses high-resolution data to evaluate the
configuration of spaces or occupant behavior in the
ED. Architects may observe the function of an ED,
facilitate focus groups, or perform user surveys
before designing a new ED; a postoccupancy evalua-
tion may be conducted after a new ED has opened.
When researchers and architects wish to study
a clinical facility, they often create their own
questionnaires, which is a labor-intensive process
(Kotzer, Zacharakis, Raynolds, & Buenning, 2011).
Digital tracking technology such as radio-
frequency identification (RFID) has also been
deployed in the ED to track the movements of indi-
viduals (Fry & Lenert, 2005). RFID has been shown
to provide benefits in healthcare, including improved
patient safety and reduction in medical errors, as well
as enhanced operational and patient workflow
(Yao, Chu, & Li, 2010). However, RFID also intro-
duces cost issues, privacy concerns, and technologi-
cal limitations such as questionable accuracy, lack
of industry standards, and possible interference with
medical devices. Moreover, it lacks the resolution
required to assess hospital workflow including the
precise activities superimposed on spatial data.
Lean techniques, such as value stream mapping
(VSM), have similarly been used to study the ED
built environment and workflow. A study from the
United Kingdom used VSM to evaluate patient wait
times in four separate EDs. Data collectors fol-
lowed patients during their ED visits and recorded
data on the patients’ ED workups. The authors
determined various wait times and the amount of
direct interaction the patients had with staff. The
study demonstrates that VSM can provide insight
on ED operations and patient experience (Swancutt
et al., 2017). However, such studies do not take into
account the effect of the healthcare built environ-
ment on workflow and patient experience.
ArcGIS is an established data collection, stor-
age, and analysis tool, used for decades to organize
and analyze spatial information, typically at a
landscape or municipal scale. Using a suite of new
programs and applications developed by Environ-
mental Systems Research Institute (2018) as part
of the ArcGIS data collection suite, researchers
can now build customized data collection tools
capable of collecting detailed positional data and
other data of interest. Data can be stored and ana-
lyzed using ArcGIS features or exported in data
tables and analyzed in other software. ArcGIS has
been used in the field of public health to study
disease and patient outcomes (Miranda, Ferranti,
Strauss, Neelon, & Califf, 2013; Wu, Jiang, & Di
Lonardo, 2018). Turrentine, Buckley, Sohn, and
Williams (2017) used ArcGIS to study readmis-
sion rates. The authors conducted chart reviews
and ArcGIS analyses which demonstrated that
2Health Environments Research & Design Journal XX(X)
patients who lived closer to a hospital had higher
rates of readmission. This research has signifi-
cance because CMS is currently implementing a
Hospital Readmissions Reduction Program. Read-
missions continue to have significant financial
implications for hospitals (U.S. CMS, 2019).
ArcGIS may facilitate a better understanding of
the healthcare environment by enabling the collec-
tion of data on the spatial heterogeneity of the hospital
and the behavior of hospital occupants. In this article,
the authors describe the development and implemen-
tation of a novel mapping workflow that includes
mapping tools, data collection methodology, a means
of analyzing the collected data, and intuitive repre-
sentations of the collected data. The workflow was
used to quantify how staff and patients inhabit the
ED. The objectives of the study were to design and
then deploy the mapping workflow in the ED, in
order to determine its feasibility in a dynamic and
complicated healthcare environment. Future applica-
tions and studies of the proposed workflow include
assessments of various hospital environments in
order to determine bottlenecks in efficiency, investi-
gate adverse events, or to evaluate the impact of spe-
cific changes to the built environment.
In this article, the authors describe the
development and implementation of a
novel mapping workflow that includes
mapping tools, data collection
methodology, a means of analyzing the
collected data, and intuitive
representations of the collected data. The
workflow was used to quantify how staff
and patients inhabit the ED. The
objectives of the study were to design and
then deploy the mapping workflow in the
ED, in order to determine its feasibility in
a dynamic and complicated healthcare
environment. Future applications and
studies of the proposed workflow include
assessments of various hospital
environments in order to determine
bottlenecks in efficiency, investigate
adverse events, or to evaluate the impact
of specific changes to the built
environment.
Method
Study Setting
Data were collected from June 2016 through
August 2016 at a large, urban, and academic med-
ical center ED in the United States. The ED has 54
beds and treats approximately 60,000 patients
annually. The ED operates on a split flow system.
Low acuity patients are seen in fast track, while all
other patients are seen in the main ED. The main
ED is divided into an A side and a B side that
function in parallel and are staffed by their own
complement of physicians and nurses (Figure 1).
A convenience sample of patients, physicians,
and nurses was directly observed in the ED by
undergraduate and medical students serving as
data collectors. A total of eight data collectors
received approximately 10 hr of field observation
training and 2 hr of training in the use of the map-
ping software. Data were collected digitally using
tablets (iPad, Apple Inc., Cupertino, CA). An
architecture firm provided the data collection team
with technical support and general guidance
regarding effective field observation techniques.
The study was submitted for expedited review and
was approved by the institutional review board.
Mapping Workflow
In order to collect data on the location, behavior,
and movement of individuals (e.g., patients, phy-
sicians, and nurses), within the ED, the authors
developed a robust mapping workflow built
around the ArcGIS mapping and analysis plat-
form along with the digital plug-in collector for
ArcGIS. Prior to field data collection, the
researchers worked to develop and structure the
mapping tool, building a custom interface within
the collector app that could be used in the field in
lieu of paper maps or other means of data capture.
Data were uploaded directly from tablets to a
cloud server where data would be available to
researchers for spatial analysis in ArcGIS
(Figure 2).
The mapping tool included a georeferenced
floor plan of the ED, a characterization scheme
for individuals, and a list of behaviors to be
tracked. Careful placement of the floor plan in
the ArcGIS model allowed for all data to be
Grzywinski et al. 3
geolocated. Custom mapping tools were devel-
oped for each experiment performed, with a
description of mapping methods outlined indivi-
dually below.
Each mapping tool type was uploaded to a
shared, cloud-hosted database (ArcOnline), and
made available to all data collectors using the
collector app. To record all data, the data
Figure 1. A descriptive floor plan of the emergency department (ED). The main ED is comprised of the A and B
side of the ED.
Figure 2. Modeling workflow utilized a combination of geographic information system (GIS) tools paired with
Illustrator and Excel.
4Health Environments Research & Design Journal XX(X)
collector zoomed to a location on the ED floor
plan using the iPad screen and placed a point on
the map representing the individual’s location.
The data collector then entered the descriptive
information regarding the tracked individual
using drop-down menus which were customized
for that mapping tool by the researchers. See fol-
lowing sections for parameters explored in each
mapping tool. All field-collected data were auto-
matically time stamped and geolocated using the
collector tool in the ArcGIS suite and then syn-
chronized with a shared database in the cloud at
each data collector’s convenience. Data synchro-
nization occurred at the end of a data collection
shift or at various points during data collection
when there was no activity.
A high degree of spatial resolution was neces-
sary since the study was conducted in a complex
and equipment-rich indoor environment. The use
of geolocated floor plans allowed data collectors
to zoom into maps and manually input the loca-
tion of individuals without relying on Global
Positioning System. A single point represented
the location of a stationary individual; a multi-
point line represented the path of a moving indi-
vidual. Data collectors attached additional
metadata, notes, and photographs (taken using the
iPad camera) to individual data points (Figure 3).
The variables collected on research subjects are
presented in Tables 1 and 2. These variables
include demographics, activities, and posture.
Because this was a pilot study, the authors were
assessing a wide variety of variables that were
thought to be beneficial based on the input of
emergency medicine faculty physicians. Before
data collection, the principal investigator, an emer-
gency medicine physician with more than 10 years
of postgraduate clinical experience, generated a
list of possible variables and visual representations
of the data. After data collection, five emergency
medicine physicians with 6–10 years of postgrad-
uate clinical experience reviewed the visual repre-
sentations of the data and selected the most
intuitive and clinically significant figures. This
occurred during an open discussion after the pre-
sentation of the preliminary study results.
Full ED Mapping Methods
The full ED mapping tool was used to create an
hourly snapshot of all visible individuals in the
ED. The primary goal of the snapshot was to cap-
ture patterns of occupancy and activity over time.
The snapshotwas conducted every hour for 48 con-
secutive hours. In order for a data collector to cover
the entire ED, the floor plan was divided into a total
Figure 3. The ArcGIS collector interface for the emergency department mapping tool on an iPad.
Grzywinski et al. 5
of 10 zones. The 10 zones can be seen in Figure 4.
The creation of these zones was based on represen-
tative distributions, so that the floor plans can be
reasonably combined into an hourly snapshot.
Within each zone, a location existed in which the
data collector could stand and observe all individ-
uals in that zone over a short period of time (approx-
imately 3 min). The data collector spent an equal
amount of time in each zone to avoid overcounting
and to capture movement and distribution of indi-
viduals. Every hour, the data collector circulated
through the ED along an established path, placing
a data point for each individual observed in the ED.
The snapshot captured both stationary and moving
individuals. It also noted physical and social char-
acteristics. The snapshot consisted of a map which
included all the individuals encountered by the data
collector. An ED snapshot was completed in
approximately 30 min. The entire ED was studied.
ED Patient Mapping Methods
Convenience sampling was used to select
patients who arrived at the ED between 08:00
and 15:00 on 8 weekdays. The ED was divided
into four zones based on the existing ED work-
flow (intake/triage, fast track/waiting rooms, A
side, and B side). A data collector was stationed
in each zone. Data collectors were stationed at
physician or nurse work stations and were
responsible for recording data on patients in
their zone of the ED. A data collector was
responsible for collecting data on multiple
patients (generally two to three patients at any
given time). Only one data collector entered data
on a patient at a given time. Data collectors
handed off responsibility for recording data
when a patient left their zone. Multiple patients
could be tracked simultaneously using this
method. Data collectors did not enter patient
rooms. With the exception of the trauma bay, all
areas of the ED were observed.
ED Physician and Nurse Mapping Methods
A data collector directly observed and recorded
data on a physician during the hours of 06:00–
23:59, or a nurse during the hours of 07:00–19:59.
The data collector tracked the physician or nurse
for three consecutive hours. During data collec-
tion, the data collector followed the physician or
nurse as they moved about the ED. With the
exception of patient rooms, all areas of the ED
were observed. Data on physicians were collected
over 3 weekdays, and data on nurses on physi-
cians were collected over 4 weekdays.
Table 1. Full ED Mapping Variables.
Category Entry
Patient
Gender Male/female/unclear
Age Baby (0–3)/child (4–17)/adult
(18–64)/elderly (65þ)
Posture Sitting/standing/lying down
Wheelchair Yes/no
Activity (multiple
choices allowed)
Phone/reading/talking/eating or
drinking/watching TV
Family/friend
Gender Male/female/unclear
Age Baby (0–3)/child (4–17)/adult
(18–64)/elderly (65þ)
Posture Sitting/standing/lying down
Wheelchair Yes/no
Activity (multiple
choices allowed)
Phone/reading/talking/eating or
drinking/watching TV
Physician (including advanced practice providers)
Type Attending/resident/nurse
practitioner/other/I don’t
know
Gender Male/female/unclear
Posture Sitting/standing
Activity Charting/talking to team/talking
to patient/procedure/other
Nurse
Gender Male/female/unclear
Posture Sitting/standing
Activity Charting/talking to team/talking
to patient/procedure/other
Other staff
Type ED tech/transport/EMT/
registration/environmental/X-
ray/other/I don’t know
Gender Male/female/unclear
Posture Sitting/standing
Activity Procedure/labs/moving X-ray/
moving cart/paperwork/
cleaning/other
Note.ED¼emergency department; EMT = emergency
medicine technician.
6Health Environments Research & Design Journal XX(X)
Data Analysis and Visualization
Data were imported into ArcMap and analyzed using
built-in spatial analytics tools. Segmentation was
used to isolate paths of tracked individuals (patients
and nurses) or to create maps that show distribution
of individuals based on identity, activities, or attri-
butes. Kernel density was used to map hot spots of
behavior across the full ED for the entire population
or by role. Kernel density calculates a magnitude-
per-unit area from both point and polyline features,
Table 2. Individual Patient, Physician, and Nurse Variables.
Patient Gender Male/Female/Unknown
Age Baby (0–3)/child (4–17)/adult (18–64)/elderly (65þ)
Activity Interacting with physician/interacting with nurse/interacting with
nonstaff/phone call/other/phone use/watching TV/reading/sleeping/
eating/bathroom/I cannot see
Procedure X-ray/ultrasound/IV/blood draw/history and physical/ortho/neuro
psych/stitches/EKG/other
Characteristics Calm/distressed/in pain/sleeping/cell phone in hand/under blanket
carrying belongings/staring absently/I can’t tell/other
Lights Exam light on–lights on (100%)–lights on (50%)–lights off
Attending Physician/
Resident Physician/
Nurse
Gender Male/female/unknown
Posture Sitting/standing/moving
Activity Charting/interacting with patient/family/interacting with care team/
interacting with other staff/checking phone/phone call/procedure
teaching/eating/drinking/paperwork/initial history and physical/other
Emotion Calm/happy/frustrated/tired/I can’t tell/other
Note IV ¼intravenous catheter placement; EKG ¼electrocardiogram.
Figure 4. The 10 data collection zones.
Grzywinski et al. 7
thereby capturing instances of both moving and still
individuals in space. It uses a kernel feature to create
a field of activity thereby, allowing activity hot spots
to be visualized over time for the entire ED or for
“neighborhoods,” which for the purpose of this study
were groupings of rooms or distinct spaces.
Quantitative analysis of the data was done with
Microsoft Excel. Adobe Illustrator was used to cre-
ate additional diagrams and infographics
that accompany spatialmaps. Qualitative and quan-
titative results were combined into graphics that
provide detailed insights into the ED environment.
Results
Full ED Mapping
A total of 5,113 data points was collected in the
48 hourly snapshots. The hourly snapshots were
aggregated to create a master data set which was
analyzed using a process of visual data discovery
and mapping. Figure 5 displays all the data points
on a single map with the hourly ED census.
Location densities were presented in heat
maps, which are visual representations of the
density of individuals where darker colors or war-
mer colors correlate to higher densities and
lighter colors or cooler colors correlate to lower
densities. The location and activities of physi-
cians and nurses were visually displayed on the
floor plans. Occupancy rates of each patient room
were calculated from the master data set and dis-
played over the ED floor plan.
ED Patient Mapping
Data collectors directly observed and collected
data on 83 patients. A map of each patient path
in the ED was created and paired with quantita-
tive graphs on the location of the patient. A qua-
litative comparison of patient paths was then
conducted. Figure 6 displays the path of an ED
patient. The patient path is paired with a pie graph
that provides a quantitative representation of the
patient’s location during their ED visit.
ED Physician Mapping
Data were collected from 10 attending (super-
vising) physicians and 10 resident (trainee)
Figure 5. All data points collected during the 48 hr of data collection.
8Health Environments Research & Design Journal XX(X)
physicians. Maps depicting the location and
activity of ED physicians were created and
paired with time lines of the physicians’ shifts.
The location and activity of resident physicians
were compared to attending physicians. As an
illustration of the utility of our workflow, the
qualitative analysis revealed that attending phy-
sicians spent more time at workstations and resi-
dent physicians spent more time in patient rooms
and moving around the ED. Figure 7 presents the
path of an attending physician. The path is
paired with a time line of the physician’s loca-
tion and activity.
ED Nurse Mapping
The movements and activities of 17 nurses were
observed and analyzed. Data on all nurses were
combined. Figure 8 displays nurse density. As
noted in the figure, nurses spent most of their shift
at nurse workstations.
Discussion
In the current study, the authors describe the devel-
opment of ED mapping tools, mapping methodol-
ogy, and mean to analyze the collected data. The
proposed workflow may then be used to create
intuitive representations of the data. The mapping
workflow can either complement or replace exist-
ing methodologies, while also providing research-
ers with a customizable instrument to study a
hospital orother healthcare environment ofinterest.
Perhaps, due to cost, technical complexity, dif-
ficulty of use, and coarseness of data resolution,
most EDs do not have access to existing tools to
assess the built environment. While automated
tracking procedures, such as RFID technology, can
produce useful information on the location of
selected individuals in the built environment, it is
costly, requires significant infrastructure, has tech-
nical limitations, and poses privacy issues (Yao
et al., 2010). Additionally, automated tracking
Figure 6. Map of an individual patient’s journey through the emergency department from intake to discharge.
Data collection began once the patient entered the intake waiting room. Dashed lines represent the presumed
path the patient took as he or she entered and exited the building. Pie chart below shows percentage of total visit
that the patient spent in each space type.
Grzywinski et al. 9
technologies do not provide information on the
nuanced behaviors, interactions, or environmental
conditions that drive patient care. Lean techniques,
such as VSM, provide little insight on spatial
trends and the healthcare built environment since
positional information is not collected.
The proposed mapping workflow has several
uses. First, the mapping workflow creates a com-
prehensive journey map ofa patient to identify bot-
tlenecks and other inefficiencies in the ED clinical
course. Winger and Hector (2015) previously stud-
ied the effect of the ED built environment on ED
Figure 8. Nurse density map.
Figure 7. Map and time line of an attending physician’s shift activity.
10 Health Environments Research & Design Journal XX(X)
staff. Through surveys, they found that the layout of
the care space affected staff productivity. The cur-
rent mapping tool can be used in conjunction with
questionnaires to determine specifically what ele-
ments of the built environment contribute to ED
efficiency.
Second, the mapping workflow can be used to
determine how the built environment contributes to
medical errors. Huisman, Morales, van Hoof, and
Kort (2012) outline several examples of how the
healthcare built environment contributes to medical
errors. Their review found that higher lighting levels
and standardized patient rooms decrease errors. It is
possible that by studying the detailed activity of phy-
sicians and nurses, additional causes of medical
errors can be identified. The proposed mapping
workflow may also serve to evaluate medical errors
after they have occurred. A “root cause analysis” is
typically performed after serious adverse events
occur in the healthcare setting. This structured
method uses a systems approach that seeks to iden-
tify both active and latent errors by closely analyzing
the time line of events that led to the error. A high-
resolution ED mapping tool would be especially
helpful in this process, as it could serve to provide
a simulation of workflow surrounding an adverse
event. For example, if a patient with a life-
threatening infection (e.g., sepsis) experiences a long
delay in receiving appropriate antibiotics and conse-
quently deteriorates clinically, it would be important
to identify the particular breakdown in the clinical
workflow that contributed to the adverse event (e.g.,
delay to antibiotics). Mapping the patient, physician,
and nurse journey from ED presentation to initial
evaluation/examination, computerized physician
order entry, and eventual administration of antibio-
tics could theoretically uncover the underlying cause
leading to a delay in antibiotics. With the ED map-
ping workflow, this would be visualized in a broader
context of physician/nurse communication, compet-
ing clinical activities that might have led to delayed
medication administration, miscommunication with
the pharmacy, and so on.
Third, changes to clinical workflow or altera-
tions to the built environment may undergo pre-/
postimplementation review using the mapping
workflow. For example, the mapping workflow
may be used to evaluate the current or proposed
location of a supply closet or automated medication
dispensing system in the clinical environment. An
examination of physician or nursing paths would
quickly uncover an inefficient location to these
essential storage areas in the ED. Changing their
location to better accommodate the observed paths
of the ED personnel would theoretically improve
the time required to retrieve needed supplies while
maximizing the core duties of patient care.
In summary, the proposed ED mapping work-
flow was found to be both feasible in terms of
implementation by the student research team
while also producing actionable representations
of the collected data. This mapping workflow
may also help researchers create robust data
visualizations of other healthcare environments
such as operating rooms, inpatient units, and out-
patient clinics. With further refinement of the
workflow and input variables, the application of
this comprehensive, high-resolution mapping tool
has significant potential to provide a clearer
understanding of how individuals inhabit the
physical spaces of healthcare environments.
Limitations
The ED mapping workflow has several limita-
tions. Because the study was conducted at a single
ED, the workflow may not be applicable to other
settings. Second, subjective data, such as the
mood of a patient, physician, or nurse, may not
be valid because the authors did not perform
interrater reliability on these observations. Third,
some behaviors and actions may not have been
recorded because researchers did not enter patient
rooms. Fourth, data were collected using a con-
venience sample of subjects. Finally, the pro-
posed workflow requires data collectors to be
present in the healthcare space at all times, which
is by nature labor intensive. Future study would
benefit from prospective analysis of consecu-
tively tracked subjects in a larger sample size,
with specific research questions that might be
answered through analysis of the collected data.
Implications for Practice
This novel workflow can effectively and
efficiently map patient, physician, and nurse
Grzywinski et al. 11
“journeys” through the ED built environ-
ment using cost-effective tools made from
commercially available applications.
High-resolution spatial, temporal, and activ-
ity data may be used to uncover operational
inefficiency or to identify patient safety
issues.
The design of the mapping workflow
can be customized to address specific
research questions pertaining to the built
environment.
The mapping workflow may be used to per-
form pre-/postassessments of changes to the
clinical built environment without requiring
users to create bespoke or custom-scripted
tools.
Acknowledgment
Nora Herrero, Colleen O’rourke, Justin Turpin,
Micaela Collins, Molly Allanoff, Terry Gao,
Yashmi Mahat, Hila Ghersin, and Jessica Sheng.
Declaration of Conflicting Interests
The authors declared no potential conflicts of
interest with respect to the research, authorship,
and/or publication of this article.
Funding
The authors received no financial support for
the research, authorship, and/or publication of
this article.
References
Centers for Disease Control and Prevention. (2018,
January 27). Emergency department visits.
Retrieved from https://www.cdc.gov/nchs/fastats/
emergency-department.htm
Environmental Systems Research Institute. (2018,
November 2). ArcGIS Geostatistical Analyst.
Retrieved from https://www.esri.com/en-us/arcgis/
products/geostatistical-analyst/overview
Fry, E. A., & Lenert, L. A. (2005). AMIA Annual Sym-
posium Proceedings, 261–265. Washington DC.
Huisman, E. R. C. M., Morales, E., van Hoof, J., &
Kort, H. S. M. (2012). Healing environment: A
review of the impact of physical environmental
factors on users. Building and Environment,58,
70–80.
Kotzer, A. M., Zacharakis, S. K., Raynolds, M., &
Buenning, F. (2011). Evaluation of the built envi-
ronment: Staff and family satisfaction pre- and
post-occupancy of the children’s hospital. Health
Environments Research & Design Journal,4(4),
60–78.
Miranda, M. L., Ferranti, J., Strauss, B., Neelon, B., &
Califf, R. M. (2013). Geographic health information
systems: A platform to support the ‘triple aim.’
Health Affairs,32(9), 1608–1615.
Morganti, K. G., Bauhoff, S., Blanchard, J. C., Abir,
M.,Iyer,N.,Smith,A.,... Kellermann, A. L.
(2013). The evolving role of emergency depart-
ments in the United States. Rand Health Quarterly,
3(2), 3.
Morgareidge, D., Cai, H., & Jia, J. (2014).
Performance-driven design with the support of digi-
tal tools: Applying discrete event simulation and
space syntax on the design of the emergency depart-
ment. Frontiers of Architectural Research,3(3),
250–264.
Saba, J. L., & Bardwell, P. L. (2004). Universal design
concepts in the emergency department. The Jour-
nal of Ambulatory Care Management,27(3),
224–236.
Swancutt, D., Joel-Edgar, S., Allen, M., Thomas, D.,
Brant, H., Benger, J., ... Pinkney, J. (2017). Not all
waits are equal: An exploratory investigation of
emergency care patient pathways. BMC Heath
Services Research,17(1), 436–445.
Turrentine, F. E., Buckley, P. J., Sohn, M., & Williams,
M. D. (2017). Travel time influences readmission
risk: Geospatial mapping of surgical readmissions.
The American Surgeon,83(6), 573–582.
U.S. Centers for Medicare & Medicaid Services. (2018,
November 1). Hospital Outpatient Quality Report-
ing Program. Retrieved from https://www.cms.gov/
medicare/quality-initiatives-patient-assessment-
instruments/hospitalqualityinits/hospitaloutpatient
qualityreportingprogram.html
U.S. Centers for Medicare & Medicaid Services. (2019,
January 16). Hospital Readmissions Reduction Pro-
gram (HRRP). Retrieved from https://www.cms.
gov/medicare/medicare-fee-for-service-payment/
acuteinpatientpps/readmissions-reduction-program.
html
Welch, S. J. (2011). Quality matters: The emergency
department, by design. Emergency Medicine News,
33(12), 21–22.
12 Health Environments Research & Design Journal XX(X)
Welch, S. J. (2012). Using data to drive emergency
department design: A metasynthesis. Health
Environments Research & Design Journal,5(3),
26–45.
Winger, D., & Hector, R. (2015). Demonstrating the
effect of the built environment on staff health-
related quality of life in ambulatory care environ-
ments. Health Environments Research & Design
Journal,8(4), 25–40.
Wu, W. Y., Jiang, Q., & Di Lonardo, S. S. (2018). Poorly
controlled diabetes in New York City: Mapping
high-density neighborhoods. Journal of Public
Health Management and Practice,24(1), 68–74.
Yao, W., Chu, C. H., & Li, Z. (2010). The use of
RFID in healthcare: Benefits and barriers. In 2010
IEEE International Conference on RFID-Technology
and Applications (RFID-TA). (pp. 128–134). doi:
10.1109/RFID-TA.2010.5529874
Grzywinski et al. 13