Available via license: CC BY 4.0
Content may be subject to copyright.
Understanding the Impact of Visualization Dashboards on Patient Safety: An Integrative
Literature Review
Aniesha Dukkipati, MSN, RN1, Heather Alico Lauria, BSN, RN1, Thomas Bice, MD2,
Debbie Travers PhD, RN1,3, Shannon Carson, MD2, Saif Khairat, PhD*1,3
1 School of Nursing, University of North Carolina- Chapel Hill, Chapel Hill, NC, USA
2 Pulmonary Diseases & Critical Care Medicine, School of Medicine, University of North
Carolina- Chapel Hill, Chapel Hill, NC, USA
3 Carolina Health Informatics Program, University of North Carolina- Chapel Hill, Chapel Hill,
NC, USA
*Corresponding Author:
Saif Khairat, PhD
Assistant Professor
Carolina Health Informatics Program (CHIP)
Division of Healthcare Systems - School of Nursing
University of North Carolina - Chapel Hill
Phone: 919-843-5413
Email: Saif@unc.edu
ICU VISUALIZATION DASHBOARD
2
Abstract
Background: Intensive Care Units (ICUs) help people with life-threatening illness or injuries
and involve close attention by a team of specially-trained health care providers. Delay between
condition onset and implementation of necessary interventions can dramatically impact the
prognosis of patients. A tool that improves display and retrieval of key clinical information has
great potential to benefit patient outcomes. The purpose of this review is to synthesize research
on the use of visualization dashboards in health care.
Methods: Articles were included if they addressed the development, testing, implementation, or
use of a visualization dashboard solution in a health care setting. An initial search was
conducted of literature on dashboards only in the intensive care unit setting, but there were not
many articles found that met the inclusion criteria. A secondary follow-up search was conducted
to broaden the results to any health care setting. The initial and follow-up searches returned a
total of 17 articles that were analyzed for this literature review.
Results: Visualization dashboard solutions decrease time spent on data gathering, difficulty of
data gathering process, cognitive load, time to task completion, errors, and improve situation
awareness, compliance with evidence-based safety guidelines, usability, and navigation.
Conclusions: Researchers can build on the findings, strengths, and limitations of the work
identified in this literature review to bolster development, testing, and implementation of novel
visualization dashboard solutions. Due to the relatively few studies conducted in this area, there
is plenty of room for researchers to test their solutions and add significantly to the field of
knowledge on this subject.
Keywords: Intensive care unit, Visualization, Dashboard, Cognitive load, Information overload,
Usability, User interface design, Health information technology, Electronic health record
ICU VISUALIZATION DASHBOARD
3
Background
State of the Problem
Critical Patient Population
Intensive Care Units (ICUs) in the United States admit more than 5.7 million people
each year [1]. The ICU level of care helps people with life-threatening illness or injuries and
involves close, constant attention by a team of specially-trained health care providers [2]. ICU
patients require frequent assessment and have a greater need for technological and clinical
support compared to non-ICU patients [1]. Important metrics in the ICU range from simple vital
sign monitoring and laboratory data to mechanical ventilator support, vasoactive medications,
and even complete circulatory support, depending on the unique needs of specific patients.
Although ICU patients receive care for a wide variety of disease states, the leading causes of
death in the ICU are multi-organ system failure, cardiovascular failure, and sepsis [2]. Delay
between condition onset and implementation of necessary interventions can dramatically impact
the prognosis of a patient with one of these life-threatening diagnoses,
Electronic Health Record Usability
EHR use has increased nationwide; however, the question remains whether EHRs are
actually being used in an effective and efficient way that improves clinical workflow and health
outcomes [3, 4]. A systematic review and meta-analysis intended to evaluate effects of health
information technology in the hospital and ICU on mortality, length of stay, and cost found
significant interstudy and intrastudy variability. The study demonstrated that more research is
needed with standardized interventions and endpoints to evaluate EHR use and implementation.
Currently, no conclusion can be made regarding the effect of health information technology on
inpatient and ICU outcomes such as mortality, length of stay, and cost [4].
Information Overload
ICU VISUALIZATION DASHBOARD
4
In 2013, Singh and et al. conducted a cross-sectional study of primary care providers to
evaluate predictors of missed test results in the setting of electronic health record (EHR) alerts.
Of the nearly 2,600 respondents, 87% perceived the quantity of alerts they received to be
excessive, 70% reported receiving more alerts than they could effectively manage, 56% reported
that the current EHR notification system made it possible for practitioners to miss test results,
and 30% reported having personally missed test results that led to care delays [5]. To address the
high volume of metrics used and the time-sensitive nature of responding to changes in a critically
ill patient's condition, a tool that improves ICU display and retrieval of key clinical information
has great potential to benefit patient outcomes.
Proposed Solution
Visualization is a field of study concerned with the transformation of data to visual
representations, where the goal is the effective and efficient cognitive processing of data [6].
Use of visualization techniques in the clinical setting have the potential to improve data display
and cognitive processing of data, reducing cognitive overload among clinicians [6]. Information
visualization involves the transformation from lower-level data to visual representations of
meanings extracted from the data [6]. Extraction is by either a computational process or a
human transcription process, the aim of which is to explore data and create new insights [6].
Some guidelines for the development of an information visualization solution include:
use realistic techniques to enhance mapping of data elements to visual objects
minimize user actions to accomplish a goal
provide flexibility in the ways to achieve the same goal
provide functionality to represent additional information
ICU VISUALIZATION DASHBOARD
5
spatially organize the visual layout
consistently apply design choices
place minimal cognitive load on the user
provide the user with information on alternatives when several actions are available
remove extraneous or distracting information
consider means to reduce the data set [6].
A dashboard is a data-driven clinical decision support tool capable of querying multiple
databases and providing a visual representation of key performance indicators in a single report
[7]. The utility of a dashboard comes from its ability to provide a concise overview of key
information [7]. Applied to the intensive care unit, a dashboard allows clinicians to quickly
identify changes in the patient's condition that require intervention. The clinician can choose
then either to dive deeper into the EHR data or refer back to the dashboard at a later point to
review changes. Depending on the design of the dashboard, features such as alerts and
documentation reminders can help clinicians improve compliance with best practice guidelines
and organizational standards [7].
Purpose of this Literature Review
The purpose of conducting this literature review is to present previous research on the use
of visualization dashboards to improve efficiency, clinician satisfaction, patient safety and
accuracy in the clinical setting. This evidence can be used to guide the development, testing and
implementation of new solutions to optimize the visualization of clinical information.
Methods
ICU VISUALIZATION DASHBOARD
6
Inclusion Criteria
Articles were included if they addressed the development, testing, implementation, or use
of a visualization dashboard solution in a health care setting. An initial search was conducted of
literature on dashboards only in the intensive care unit setting, but there were limited articles
found that met the inclusion criteria. A secondary follow-up search was conducted to broaden
the results to any health care setting. Ideally, the article would compare outcomes with the novel
solution to outcomes prior to or without the novel solution. However, articles were not excluded
simply due to lack of a specific comparison. Articles should contain quantitative or qualitative
outcomes related to clinician satisfaction, cognitive overload, or patient outcomes. Initially,
abstracts were scanned to identify if articles were relevant to the specified research questions.
Exclusion Criteria
Due to the specificity and novel nature of this topic, no filters were applied to the query.
This means that no articles were excluded solely based on type, publication date, or country of
origin. However, articles were excluded if there was not an English version of the article
available. Articles were excluded if review of the abstract and full text revealed the article did
not address at least one of the specified research questions and meet the inclusion criteria.
Databases
Databases selected for this search were PubMed, PMC, CINAHL, and EMBASE, all of
which are health sciences journal article databases.
Initial Search Terms - ICU Only
In order to capture alternative ways of denoting the terms of interest, the query of
("electronic medical record" OR "electronic health record" OR EMR OR EHR) AND
(visualization OR dashboard OR design OR interface) AND ("intensive care" OR ICU OR
"critical care" OR CCU) was used for the initial search. Abstracts were screened for relevance to
the intended investigation. Articles with relevant abstracts were then read in entirety to further
screen the relevance and quality of the data.
ICU VISUALIZATION DASHBOARD
7
Searching the above query into PubMed returned 151 results, 14 of which were analyzed,
and 8 of which were relevant, quality results for the final analysis. PMC returned 3,405 results,
13 of which were analyzed, 5 of which were excluded due to duplication, and 3 of which were
included in the final analysis. CINAHL returned 86 results, 7 of which were analyzed, 2 of
which were excluded due to duplication, and 0 included in the final analysis. EMBASE returned
646 results, 18 of which were analyzed, 6 of which were excluded due to duplication, and 1 of
which was included in the final analysis. Therefore, a total of 12 articles were obtained from this
primary search towards the final analysis.
Follow-up Search Terms - Any Health Care Setting
Due to the limited number of results obtained with the initial search, a secondary search
was completed using the query of visualization AND dashboard. The purpose of the secondary
search was to broaden the search to a visualization dashboard solution in any health care setting,
as opposed to only the intensive care unit setting. The same databases and process of screening
articles were maintained from the initial search process.
Searching the above query into PubMed returned 24 results, 7 of which were analyzed, 1
of which was excluded due to duplication, and 4 of which were relevant, quality results for the
final analysis. PMC returned 311 results, 10 of which were analyzed, 5 of which were excluded
due to duplication, and 1 of which was included in the final analysis. CINAHL returned 3
results, 2 of which were analyzed, 0 of which were excluded due to duplication, and 0 of which
were included in the final analysis. EMBASE returned 30 results, 4 of which were analyzed, 3
of which were excluded due to duplication, and 0 of which were included in the final analysis.
Therefore, a total of 5 articles were obtained from the secondary search towards the final
analysis, Figure 1.
Implications of Query Results
The initial and follow-up searches returned a total of 17 articles that were analyzed for
this literature review. The limited results reflect the novel status of this area of research.
!
"
#
$
ICU VISUALIZATION DASHBOARD
8
Supporting information from information and library science databases will be useful in the
analysis steps as much of this work involves development, testing, and implementation of a
novel software solution.
Figure 1a. Literature Review Process (Primary Search)
Figure 1b. Literature Review Process (Secondary Search)
%!
%!
!
"
&!
#
&
'(
ICU VISUALIZATION DASHBOARD
9
Results
The dashboard solutions that we identified in the 17 articles are presented with
organization by study findings related to efficiency, quality/safety, accuracy, and user
satisfaction. Some solutions were discussed in multiple articles, whereas others were unique to a
single article. Table 1 presents the sample size, metrics of interests, results and findings for each
study.
Table 1. Study characteristics and results
Study Metric of Interest Sample (n) Result Findings
Ahmed et. al
(2011) Accuracy, Efficiency 160
##)*+,%
-.*!%#+,++/-0)
*,%-.*%
'1+,+++&%2()*&%
&!%&!$%&&!%-.*$%+%%
%'1+,+++&%(&++
,&+!
3##%
(0(#4
#%(#05)
23678
Farri et al. Accuracy, Efficiency 8
*98!,:&,!
%,:&,!
%#
+,+%4&,;+,
,!,;+,%'+,+$
30(8
#4(
,9
4,58",
2(
,
Koch et al. Accuracy, Efficiency 12
2()*!,&%<*
!,+%'1,++&%)
&,=%<*
,=%'1+,++&
520(#(
48
#(#
#,
>0!+&
%?%
4
&(0
# 4
<4@0A
B2
(,(
?4C2(48
#(#4##
4
Faiola et al (2015) Accuracy, Efficiency,
Satisfaction 12
2()#(8#4
8D*EFG
&+,&&%#,+&&%,/+HEFG
&+,%#,++%,/
)#(9,%<
,+%8#9,%<
,I!&%&!,+%#,+
>)(08
88
,J)(08
!KD,F
4##(#49LM
!,+
'08
!+& ?
'+%'
2(#)8
'*&!(%'*$(
L(#?44(
(8(#H
((
ICU VISUALIZATION DASHBOARD
10
Dolan et. al Efficiency,
Quality/Safety 25 9(8
,(%5
(#8#,
L
#4484(
#(08#
,
Pageler et al. Efficiency,
Quality/Safety 64
L(#
884(/=$+=
',++%#8/=$=
#1+,++&%#8
$=$='1+,++&H
(#
(#67% to 62% (P = .
001), 2.6 CLABSIs per 1000 line-
days before intervention to 0.7
CLABSIs per 1000 line-days
L(#(#
)%#)#"
,
C8
!+&+ FK4 5K 5D"
'(##4%
((
0N
Shaw et al. Quality/Safety 450
2()'*$(%'*!+!
(%'+,+%FK4)
<&
&&%'+,+&
2((4('L>B
(8(
$=,'
#O$
4(&&&,
Dziadzko et al
(2016) Satisfaction 361 '*&(%'*&!(%
'+,+
3(#88
8
@0!+&!
4%
FK4 5K 5D"
L8
4%"4P#
%(#
C
!+& 2<#(
#&%+
#! 5D"
888#
08)4
4#,
@8
!+& 2#( 5K 5D"
5#%4
#(
!+& 2#( ! 5D"
38(4
(0H
#48
!+&! 2#( 5K 5D"
2#(4(
D44(
8
!+& 2#( 5K 5D"
,2
(#(,@Q8
44(
#(
Efficiency
The Ambient Warning and Response Evaluation (AWARE) system was tested in two articles
included in this literature review [8,9]. AWARE is an ICU-specific patient viewer and
monitoring system that was developed at Mayo Clinic [8]. AWARE is a superstructure for
existing EHR. The development of this tool was guided by clinicians based their information
ICU VISUALIZATION DASHBOARD
11
needs [8]. Pickering et al. (2015) used a step wedge cluster randomization trial to demonstrate a
decrease in time spent on pre-round data gathering using the AWARE system [9]. Compared to
the existing EHR, AWARE was reported to improve information management (data presentation
format and efficiency of data access) and make the task of gathering data for rounds significantly
less difficult and mentally demanding [9].
Scripps Clinic and Green Hospital used a rapid-cycle evaluation process to develop the
algorithms, alert systems, and interfaces intended to facilitate patient-provider interactions and
determination of treatment plans [10]. Brooke's Standardized Usability Tool was used to
evaluate usability and two independent appraisers reviewed the think aloud sessions for usability
themes [10]. Results pointed to positive results regarding usability and efficiency to identify
pertinent components in the patient's plan of care with use of the prototype [10].
Ahmed et al. (2011) evaluated a novel .NET based application by conducting a randomized
crossover study [11]. This study demonstrated improved workload (using NASA-task load
index), decreased time to task completion, and decreased number of errors of cognition, [11].
Additionally, the standard EHR contained a much larger data volume compared with the novel
user interface[11].
Koch et al. (2013) evaluated nurses’ situation awareness and task completion time using
an integrated information display compared to traditional displays [12]. Task completion time
(response time from seeing the question to submitting the answer) was measured using paper
prototypes of both displays [12]. Task completion times were nearly half with integrated
displays compared to traditional displays[12].
Farri et al. (2012) carried out three iterations of planning, risk analysis, design, and
evaluation of an EHR prototype. This user interface contained specific functionalities for
clinical documents [13]. They used a spiral model for software development and the EHR
ICU VISUALIZATION DASHBOARD
12
system user interface framework of the Veterans Affairs computerized patient record system
(VistA CPRS) [13].
The researchers used a mixed methods approach to evaluate a sample of eight medical
interns as they synthesized EHR clinical documents in four pre-formed clinical scenarios [13].
Despite the non-significant difference in total times to task completion the researchers observed
shorter times for two scenarios with the visualization tool. This may suggest that the timesaving
benefits may be more evident with certain clinical processes [13]
Dolan et al. (2013) used a mixed quantitative and qualitative evaluation process to
evaluate their dashboard prototype [14]. The researchers observed the time participants spent
using the dashboard before choosing a preferred drug, ease of use, acceptability, decisional
conflict, and an open-ended qualitative analysis [14]. Qualitative findings were positive,
suggesting potential for informed decision making and patient centered care [14].
Medical Information Visualization Assistant, v.2 (MIVA 2.0) is an EHR dashboard
technology that uses a visualization engine to deliver multivariate biometric data by transforming
it into temporal resolutions [15]. ICU clinicians are able to use selection menus to control the
viewability of data in various time periods to assist with diagnosis and treatment [15]. The
usability speed test identified no significant difference in time-on-task between the control group
and the experimental group [15]. However, a significant difference was noted in speed with use
of MIVA 2.0 [15].
Clinician Satisfaction
Dziadzko et al. (2016) studied the before- and-after- implementation experience and
satisfaction of ICU providers at two hospitals using the AWARE system [8]. Providers agreed
that data gathering using the existing EHR system was difficult and time-intensive [8]. In a
survey analysis, researchers found that prescribers were significantly more satisfied with the
ICU VISUALIZATION DASHBOARD
13
delivery of content and information output with AWARE due to the improvement of the
presentation of information [8]. Bakos et al. (2012) showed an increased use of the dashboard
tool at Virginia Commonwealth University Health System throughout the first year of
implementation, demonstrating clinician satisfaction with usage. Interviews further confirmed
the benefit and helpfulness of using the tool as staff confirmed its usefulness in their workflow
[16].
Quality and Safety
Pageler et al. (2013) discuss use of a checklist enhanced by the EHR and a unit-wide
dashboard to improve compliance with an evidence-based, pediatric-specific catheter care bundle
[17]. The researchers performed a cohort study with historical controls that included all patients
with a central venous catheter at a 24-bed Pediatric ICU (PICU) in an academic children's
hospital [17].
Central line associated bloodstream infection (CLABSI) rates decreased after the
checklist intervention [17]. Analysis of specific bundle elements demonstrated decreased
compliance with insertion bundle documentation. However, there was an increase in compliance
with daily documentation of line necessity, dressing changes, cap changes, and port needle
changes.
Shaw et al. (2015) evaluated a real-time visual display that showed data on presence of
consent for treatment, restraint orders, presence of urinary catheters, deep venous thrombosis
(DVT) prophylaxis, Braden Q score, and medication reconciliation [18]. An automated EHR
querying tool was created to assess compliance with a PICU safety bundle, and querying of the
EHR for compliance and updating of the dashboard automatically occurred every five minutes
[18].
Baseline compliance and duration of noncompliance was established during three time
periods: before activation of the dashboard, at one month following activation of the dashboard,
ICU VISUALIZATION DASHBOARD
14
and at three months after activation [18]. There was no difference between the three periods in
presence of restraint orders, DVT prophylaxis, or development or worsening of pressure ulcers
[18]. Between the first and third time periods, the median time from PICU admission to
obtaining treatment consent decreased [18]. The number of patients with urinary catheters in
place > 96 hours decreased significantly after the intervention [18]. The researchers concluded
that a unit-wide dashboard could increase awareness for potential interventions, thereby affecting
patient safety in a dynamic manner [18].
Although Bakos et al. (2012) speculate that their visualization dashboard will contribute
to having zero events of preventable harm to patients, employees and visitors; there is no
quantifiable data to support this at this time [16]. Similarly, Hagland et al. (2010) discuss the
potential to improve patient safety, communication and clinician workflow using a new clinical
dashboard without quantifiable results [19].
Accuracy
Koch et al. (2013) used the paper prototypes of their displays to measure situation
awareness (accuracy of the participants’ answer). Nurses had a higher situation awareness and
accuracy when using the integrated display versus the traditional display [12].
To evaluate the accuracy of, MIVA 2.0, Faiola et al. (2015) used quantitative clinical
decision-making task questions. The clinical decision-making accuracy test identified an overall
significant improvement in accuracy of the eight-question test between the experimental versus
control groups. Qualitative results were obtained from seven open-ended interview questions,
wherein participants acknowledged the potential impact of MIVA 2.0 for reducing cognitive load
and enabling more accurate decision-making [15]. Overall, a significant difference was noted in
accuracy with use of MIVA 2.0 [15].
Dziadzko et al. (2016) surveyed healthcare providers who reported an improvement in
the accuracy of decision-making using AWARE but no quantifiable data is available [8]. Using
ICU VISUALIZATION DASHBOARD
15
the .NET based application, Ahmed et al. (2011) found that the median number of errors per
provider decreased significantly for the novel user interface compared to the standard electronic
medical record interface [13].
Farri et al. (2012) evaluated the accuracy of using their spiral model software. The resulting
differences in unretrieved patient information and accurate inferences were not statistically
significant, but suggested some improvement with the new information visualization tool [13].
Other observed effects of the tool included more intuitive navigation between patient details and
increased effort towards methodical synthesis of clinical documents [13].
Scripps Clinic and Green Hospital demonstrated an improved accuracy of the healthcare provider
“Heart Team” in clinical decision-making using 15 mock patients. However, a complete data
analysis was not performed [10].
Tool Development
Five articles did not focus on efficiency, quality/safety, accuracy and satisfaction
outcomes but discussed their process of tool development. Their findings during visualization
tool development are included in the discussion section below [20, 21,22, 23, 24].
Discussion
Table 1.
Overview
The 17 articles included in this literature review demonstrate how efficiency,
quality/safety, clinician satisfaction and accuracy can be improved using a visualization
ICU VISUALIZATION DASHBOARD
16
dashboard. These 17 articles share many themes regarding how the each dashboard was
designed and what user-friendly features are available when using the dashboard. These themes
are discussed below. With each idea outlined, a discussion of its application to prior
visualization dashboard solutions and its implications for future studies follows. Application of
these approaches, methods, and features may serve useful in future efforts related to this subject
matter. A summary of findings from the aforementioned articles is depicted in Figure 1.
Figure 2. Summary of findings from the literature review.
Human- Centered
Design
A
The IView dashboard was developed for use on three ICU's at the Children's Hospital at
Pittsburgh, and resulted from intensive clinician-IT team-based work and a collaborative
relationship with the hospital's clinical IT vendor [15]. Qualitative measures regarding perceived
ICU VISUALIZATION DASHBOARD
17
patient safety, clinician workflow, and physician-nurse communications pointed to positive
outcomes in all three categories [15].
Swartz et al. (2014) discuss the creation of iNYP, a Java-based service-oriented web
application, to meet the specific information needs of emergency medicine clinicians [21]. A
combination of survey and structured interview were used to inform the development of this
specialty-specific clinical dashboard [20].
Hartzler et al. (2015) discuss use of human-centered design methods to create visual displays of
patient reported outcomes [21]. Targeted, iterative design activities were used to inform
development of a dashboard that visually displays patient-reported pain and disability outcomes
following spine surgery [21]. The Multi-signal Visualization of Physiology (MVP) was
developed at the Neuroscience ICU of the National Neuroscience Institute in Singapore to
provide a more visual, straightforward, and intuitive diagnosis process [23]. The MVP makes
use of a polygram that incorporates live readings of physiological signs and colors to highlight
different patient statuses [23].
Overview
With each idea outlined, a discussion of its application to prior visualization dashboard solutions
and its implications for future studies follows. Application of these approaches, methods, and
features may serve useful in future efforts related to this subject matter. Interdisciplinary
Approach
Nine articles mentioned use of an interdisciplinary approach in developing, testing, and
implementing their visualization solution. The benefit of an interdisciplinary approach is that the
varied professional perspectives and skills that come with different disciplines are integrated into
each step of the process [8, 10, 11, 15, 16, 19-21, 24]
Use of an Interactive Prototype
Prototyping is a useful process as it allows developers to strategize product design and
obtain feedback from end users without the expansive investment of resources required to make
changes in the EHR format [10, 12- 15, 21, 22]. While there are viable electronic prototyping
ICU VISUALIZATION DASHBOARD
18
options available, paper-based prototyping can be a useful, cost-effective solution in early stages
of product design [25]. A mixed evaluation process of quantitative and qualitative measures can
be used to direct feedback from end user interaction with the prototype and improve design on
subsequent revisions [10, 12-15, 21, 22].
Using Open-Source Technology vs. Adapting a Third-Party Vendor's EHR System
A team with limited resources may not be able to invest financial, temporal, and staff
resources into developing a suitable product [26]. Those teams with limited resources may have
to wait for a solution to stem from others using open-source technology or for the third-party
vendor to provide an option that will be suitable [22, 27]. Use of an open-source technology can
allow more freedom for the user to develop and share their tool with others than when making
adjustments to an EHR developed by a third-party vendor [22]. Ultimately, each team at a
specific organization will decide which route aligns better with their own resources and goals,
but the distinct opportunities and risks inherent with each option are important to consider.
Adapting a vendor's EHR system will require continual consultation with the vendor and
there may be significant limitations imposed by the contract between the organization and the
vendor [27].
Application of Evidence-Based, Clinical Practice Guidelines to the EHR
Clinical practice guidelines (CPGs) are intended to improve the quality, consistency, and
effectiveness of care by applying evidence-based medicine [28]. A review of physician
adherence to clinical practice guidelines suggested that as many as 38% of physicians consider
clinical practice guidelines as inconvenient or too difficult to use [28]. Incorporation of clinical
practice guidelines into the structure and display of the EHR may help improve convenience of
access to practice guidelines and increase use in clinical decision making [16-18].
Clinician Controlled Selection Menus
Allowing the clinician to adjust the data displayed in alignment with the preference and
needs of that individual may further improve clinician satisfaction with the system [15, 24]. This
capability can also help meet the goal of reducing cognitive overload [15, 24]. If clinicians are
ICU VISUALIZATION DASHBOARD
19
able to filter out information that is not pertinent to them, the remaining information will have
improved visibility without obstruction from extraneous information [15, 24]. The capability to
filter information by location, service lines, and specific diagnoses may also serve useful to
improve efficiency, accuracy and user satisfaction of clinicians managing many patients [15].
Improved Display of Trends in Physiological Signs
In a setting such as an intensive care unit, the stability of a patient's condition can quickly
deteriorate [2]. While clinicians have primary responsibility to assess their patient's condition
and intervene appropriately, adding features to the EHR that can assist with this process can
expedite these steps; improving efficiency [23]. With the vast array of physiological parameters
under continuous monitoring in the ICU setting, improved display of data trends may improve
the clinician's responsiveness in adding or weaning interventions based on the patient's changing
condition [23].
Classification of Data by Body System
Classification of data using a body system approach was a common decision for the EHR
designs in this literature review [8, 11]. By using a body system approach, clinicians are able to
follow a systematic approach to optimizing the patient's holistic health. The design choice of
matching the body system approach used by intensive care unit clinicians allows for congruency
between the EHR display and cognitive organization of clinical information [8, 11].
Applicability of a Visualization Dashboard to Non-ICU Clinical Settings
While this literature review focused primarily on the application of a visualization
dashboard to the intensive care unit setting, the same intervention could have benefit in other
clinical settings as well [10, 13, 14, 16 20-22]. The emergency department could be well-suited
for this intervention as the EHR could then assist in alerting clinicians to new results and a
change in the patient's clinical status that modifies the plan of care [20]. Step-down units and
inpatient floors may not have the same extent of clinical data as the intensive care unit setting but
clinicians may still find benefit from features related to improved display of clinical information.
ICU VISUALIZATION DASHBOARD
20
Once a visualization dashboard is successfully implemented in the ICU setting, the dashboard
can be modified, tested, and implemented in non-ICU clinical settings; working towards similar
goals [22].
Conclusions
. The findings, strengths, and limitations discussed in this section can drive future research
efforts on visualization dashboard solutions.
Information needs varied based on patient population and clinical role. Key findings regarding
clinician needs for the solution included the following:
application of evidence-based, clinical practice guidelines
clinician controlled selection menus
use of color coded visual indicators
classification of data by body system
match of EHR design to process of interdisciplinary rounds
Strengths and Limitations of Solutions
This literature review includes information on several visualization dashboards that have
been tested with positive results from quantitative and qualitative analysis. These positive results
support the potential benefits of a visualization dashboard solution to clinical practice
environments.
Limitations were noted in the following areas:
The interpretation of what a visualization dashboard solution entails varied widely among
the researchers of the different studies included.
ICU VISUALIZATION DASHBOARD
21
Many of the visualization dashboard solutions were evaluated with a solely qualitative
approach, rather than with a quantitative or mixed methods approach.
Some articles included details about the design, implementation, and evaluation
processes, but did not include full detail on the data obtained.
Some studies used a simulated setting in lieu of a live clinical setting, which means that
results may differ when the solution is applied to a live clinical setting.
Most studies tested a single solution in a single implementation setting, which limits the
generalizability of the findings to other solutions and other implementation settings.
Future Direction
Researchers can build on the findings, strengths, and limitations of the work identified in
this literature review to bolster development, testing, and implementation of a novel visualization
dashboard solution. Due to the relatively few studies conducted in this area, there is plenty of
room for researchers to test their solutions and add significant information to the field of
knowledge on this subject. An effective solution in this area can drive process improvement and
improved patient outcomes for not only the initial setting of implementation, but also to any
further clinical units and organizations that adopt the intervention.
Conclusions
Overall, successful visualization dashboards utilized an interdisciplinary approach to
develop a human-centered design. Dashboards were flexible and could be adjusted to the users’
preferences as well as organized based on body system, color-coded and adapted for clinician
team rounding. These features are important due to the variety in patient population and the
diverse way that clinicians interpret information. Utilizing these common themes to develop
ICU VISUALIZATION DASHBOARD
22
visualization tools for patient care has shown to improve efficiency, quality/safety, clinician
satisfaction and accuracy in a variety of patient settings.
This section synthesizes the major findings of the 17 articles. As discussed, visualization
tools have the potential to impact accuracy, efficiency, user satisfaction and quality/safety of care
in the ICU and other settings. Numerous factors such as clinician controlled displays,
organization by body system, an interdisciplinary design team and using open-source technology
can result in successful implementation of a visualization dashboard. The findings, strengths,
and limitations discussed in this section can drive future research efforts on visualization
dashboard solutions.
Design Recommendation based on Clinician Needs
Information needs varied based on patient population and clinical role. Key findings regarding
clinician needs for the solution included the following:
application of evidence-based, clinical practice guidelines
clinician controlled selection menus
use of color coded visual indicators
classification of data by body system
match of EHR design to process of interdisciplinary rounds
ICU VISUALIZATION DASHBOARD
23
As demonstrated in the results section of this paper, the combination of the above components
can allow for user-friendly dashboard designs that have the potential to impact accuracy,
efficiency, user satisfaction and quality/safety of care.
Declarations
Ethics Approval and Consent to Participate
Not applicable.
Consent for Publication
Not applicable.
Availability of Data and Material
Not applicable.
Competing Interests
The authors declare that they have no competing interests.
Funding
Not applicable.
Authors' Contributions
AD and carried out the literature review and drafted the manuscript under the direct
supervision of SK and DT. AD, HL, SK, DT, TB, and SC analyzed and interpreted the results,
revised the manuscript, and support this research. All authors contributed to, read, and approved
the final manuscript.
ICU VISUALIZATION DASHBOARD
24
Abbreviations
EHR: Electronic Health Record
ICU: Intensive Care Unit
AWARE: Ambient Warning and Response Evaluation
CPGs: Clinical Practice Guidelines
MVP: Multi-signal Visualization of Physiology
DVT: Deep Vein Thrombosis
ICU VISUALIZATION DASHBOARD
25
References
1. Society of Critical Care Medicine. (2016). Critical care statistics. Available from: URL:
http://www.sccm.org/Communications/Pages/CriticalCareStats.aspx. Accessed 8
February 2017.
2. U.S. National Library of Medicine. (2016). Critical care. Available from: URL:
https://www.nlm.nih.gov/medlineplus/criticalcare.html. Accessed 8 February 2017.
3. Charles D, Gabriel M, Searcy. Adoption of electronic health record systems among U.S.
non-federal acute care hospitals: 2008-2014. ONC Data Brief. 2015;23:1-10.
4. Thompson G, O'Horo JC, Pickering BW, Herasevich V. Impact of the electronic medical
record on mortality, length of stay, and cost in the hospital and ICU: A systematic review
and metaanalysis. Critical Care Medicine. 2015;43(6):1276-82. doi:
10.1097/CCM.0000000000000948.
5. Singh H, Spitzmueller C, Petersen N, Sawhney M, Sittig D. Information overload and
missed test results in EHR-based settings. Jama Intern Med. 2013;173(8):1-4.
doi:10.1001/2013.jamainternmed.61.
6. Engelbrecht L, Botha A, Alberts R. (2015). Designing the visualization of information.
International Journal Of Image & Graphics. 2015;15(2):1-19.
doi:10.1142/S0219467815400057.
7. Wilbanks BA, Langford PA. A review of dashboards for data analytics in nursing. CIN:
Computers, Informatics, Nursing. 2014;32(11): 545-9. doi:
10.1097/CIN.0000000000000106.
8. Dziadzko MA, Herasevich V, Sen A, Pickering BW, Knight AM, Moreno Franco P. User
perception and experience of the introduction of a novel critical care patient viewer in the
ICU setting. International Journal of Medical Informatics. 2016;88:86-91.
doi:10.1016/j.ijmedinf.2016.01.011.
ICU VISUALIZATION DASHBOARD
26
9. Pickering BW, Dong Y, Ahmed A, Giri J, Kilickaya O, Gupta A, . . . Herasevich V. The
implementation of clinician designed, human-centered electronic medical record viewer
in the intensive care unit: A pilot step-wedge cluster randomized trial. International
Journal of Medical Informatics. 2015;84(5): 299-307.
doi:10.1016/j.ijmedinf.2015.01.017.
10. Clarke S, Wilson ML, Terhaar M. Using clinical decision support
and dashboard technology to improve heart team efficiency and accuracy in a
transcatheter aortic valve implantation (TAVI) program. Stud Health Technol Inform.
2016;225:98-102. PMID: 27332170
11 Ellsworth MA, Lang TR, Pickering BW, Herasevich V. Clinical data needs in the neonatal
intensive care unit electronic medical record. BMC Medical Informatics and Decision
Making. 2014;14(92): 1-6. doi:10.1186/1472-6947-14-92.
. Ahmed A, Chandra S, Herasevich V, Gajic O, Pickering BW. The effect of two different
electronic health record user interfaces on intensive care provider task load, errors of
cognition, and performance. Critical Care Medicine. 2011;39(7): 1626-34.
doi:10.1097/CCM.0b013e31821858a0.
12. Koch SH, Weir C, Westenskow D, Gondan M, Agutter J, Haar M,... Staggers N.
Evaluation of the effect of information integration in displays for ICU nurses on situation
awareness and task completion time: A prospective randomized controlled study.
International Journal of Medical Informatics. 2013;82:665-75. doi:
10.1016/j.ijmedinf.2012.10.002
13. Farri O, Rahman A, Monsen KA, Zhang R, Pakhomov SV, Pieczkiewicz DS, ... Melton
GB. Impact of a prototype visualization tool for new information in EHR clinical
documents. Applied Clinical Informatics. 2012;3:404-18. doi:10.4338/ACI-2012- 05-
RA-0017.
ICU VISUALIZATION DASHBOARD
27
14. Dolan JG, Veazie PJ, Russ AJ. Development and initial evaluation of a treatment decision
dashboard. BMC Medical Informatics and Decision Making. 2013;13(51).
doi:10.86/1472-6947-13-51.
15. Faiola A, Srinivas P, Duke J. Supporting clinical cognition: A human-centered approach
to a novel ICU information visualization dashboard. AMIA...Annual Symposium
Proceedings / AMIA Symposium. 2015;560-9. PMID: 26958190
16. Bakos K.K, Zimmermann D, Moriconi D. Implementing the clinical dashboard at
VCUHS. NI 2012: Proceedings of the 11th International Congress on Nursing Informatic.
2012; 11-5. PMID: 24199038
17. Pageler NM, Longhurst CA, Wood M, Cornfield DN, Suermondt J, Sharek PJ, Franzon
D. (2014). Use of electronic medical record-enhanced checklist and electronic dashboard
to decrease CLABSIs. Pediatrics. 2014;133(3):e738-46. doi:10.1542/peds.2013-2249.
18. Shaw SJ, Jacobs B, Stockwell DC, Futterman C, Spaeder MC. Effect of a real-time
pediatric ICU safety bundle dashboard on quality improvement measures. The Joint
Commission Journal on Quality and Patient Safety. 2015;41(9):414-20. PMID:
26289236
Farri O, Rahman A, Monsen KA, Zhang R, Pakhomov SV, Pieczkiewicz DS, ... Melton
GB. Impact of a prototype visualization tool for new information in EHR clinical
documents. Applied Clinical Informatics. 2012;3:404-18. doi:10.4338/ACI-2012- Dolan
JG, Veazie PJ, Russ AJ. Development and initial evaluation of a treatment decision
dashboard. BMC Medical Informatics and Decision Making. 2013;13(51).
19.Proceedings / AMIA Symposium. 2015;560-9. Hagland, M. Improving patient care
ICU VISUALIZATION DASHBOARD
28
through data availability in the ICU. Healthcare Informatics: The Business Magazine
for Information and Communication Systems. 2010;27(3):20-8. PMID:20344925
20. Swartz, JL, Cimino JJ, Fred MR, Green RA, Vawdrey DK. Designing a clinical
dashboard to fill information gaps in the emergency department. AMIA Annual
Symposium Proceedings. 2014;1098-1104. PMID: 25954420
21. Hartzler AL, Chaudhuri S, Fey BC, Flum DR, Lavallee D. Integrating patient-reported
outcomes into spine surgical care through visual dashboards: Lessons learned from
human-centered design. 2015;3(2):1-17. doi:10.13063/2327-9214.1133.
22. Badgeley MA, Shameer K, Glicksberg BS, Tomlinson MS, Levin MA, McCormick PJ,
Kasarskis A, Reich DL, Dudley JT. EHDViz: Clinical dashboard development using
open-source technologies. BMJ Open. 2016;6(3): 1-11. doi:10.1136/bmjopen-2015-
010579.
23 NI 2012: Proceedings of the 11th International Congress on Nursing Informatic. .
Sebastian K, Sari V, Loy LY, Zhang F, Zhang Z, Feng M. Multi-signal
Visualization of Physiology (MVP): A novel visualization dashboard for physiological
monitoring of traumatic brain injury patients. Engineering in Medicine and Biology
Society (EMBC), Annual International Conference of the IEEE. 2012;2000-3. doi:
10.1109/EMBC.2012.6346349.
24. Ellsworth MA, Lang TR, Pickering BW, Herasevich V. Clinical data needs in the neonatal
intensive care unit electronic medical record. BMC Medical Informatics and Decision
Making. 2014;14(92): 1-6. doi:10.1186/1472-6947-14-92.
ICU VISUALIZATION DASHBOARD
29
25 Clarke S, Wilson ML, Terhaar M. Using clinical decision support
and dashboard technology to improve heart team efficiency and accuracy in a
transcatheter aortic valve implantation (TAVI) program. Stud Health Technol Inform.
2016;225:98-102. . Sutcliffe A, Bruijn O, Thew S, Buchan I, Jarvis P, McNaught J,
Procter R. Developing visualization-based decision support tools for epidemiology.
Information Visualization. 2014;13(1):3-17. doi: 10.1177/1473871612445832.
26. Marsolo K. Informatics and operations -- Let's get integrated. Journal of the American
Medical Informatics Association: JAMIA. 2013;20(1):122-4. doi: 10.1136/amiajnl-2012-
001194.
Ghazisaeidi M, Safdari R, Torabi M, Mirzaee M, Farzi J, Goodini A. Development of
performance dashboards in healthcare sector: Key practical issues. Acta Inform Med.
2015;23(5):317–21. doi: 10.5455/aim.2015.23.317-321.
27. Marsolo K. Informatics and operations -- Let's get integrated. Journal of the American
Medical Informatics Association: JAMIA. 2013;20(1):122-4. doi: 10.1136/amiajnl-2012-
001194.
28. Armstrong JJ, Goldfarb AM, Instrum RS, MacDermid JC. Improvement evident but still
necessary in clinical practice guideline quality: A systematic review. Journal of clinical
epidemiology. 2016;1-9. doi: 10.1016/j.jclinepi.2016.08.005
UNDERSTANDING THE IMPACT
UNDERSTANDING THE IMPACT 31
UNDERSTANDING THE IMPACT 32
UNDERSTANDING THE IMPACT 33
UNDERSTANDING THE IMPACT 34
UNDERSTANDING THE IMPACT 35
UNDERSTANDING THE IMPACT 36
UNDERSTANDING THE IMPACT 37
UNDERSTANDING THE IMPACT 38
UNDERSTANDING THE IMPACT 39
UNDERSTANDING THE IMPACT 40
UNDERSTANDING THE IMPACT 41
UNDERSTANDING THE IMPACT 42