Centralized Multipatient Dashboards' Impact on Intensive Care Unit Clinician Performance and Satisfaction: A Systematic Review

Abstract

Background Intensive care unit (ICU) clinicians encounter frequent challenges with managing vast amounts of fragmented data while caring for multiple critically ill patients simultaneously. This may lead to increased provider cognitive load that may jeopardize patient safety.

Objectives This systematic review assesses the impact of centralized multipatient dashboards on ICU clinician performance, perceptions regarding the use of these tools, and patient outcomes.

Methods A literature search was conducted on February 9, 2023, using the EBSCO CINAHL, Cochrane Central Register of Controlled Trials, Embase, IEEE Xplore, MEDLINE, Scopus, and Web of Science Core Collection databases. Eligible studies that included ICU clinicians as participants and tested the effect of dashboards designed for use by multiple users to manage multiple patients on user performance and/or satisfaction compared with the standard practice. We narratively synthesized eligible studies following the SWiM (Synthesis Without Meta-analysis) guidelines. Studies were grouped based on dashboard type and outcomes assessed.

Results The search yielded a total of 2,407 studies. Five studies met inclusion criteria and were included. Among these, three studies evaluated interactive displays in the ICU, one study assessed two dashboards in the pediatric ICU (PICU), and one study examined centralized monitor in the PICU. Most studies reported several positive outcomes, including reductions in data gathering time before rounds, a decrease in misrepresentations during multidisciplinary rounds, improved daily documentation compliance, faster decision-making, and user satisfaction. One study did not report any significant association.

Conclusion The multipatient dashboards were associated with improved ICU clinician performance and were positively perceived in most of the included studies. The risk of bias was high, and the certainty of evidence was very low, due to inconsistencies, imprecision, indirectness in the outcome measure, and methodological limitations. Designing and evaluating multipatient tools using robust research methodologies is an important focus for future research.

 

MissingLinkText

Keywords

centralized multipatient display - dashboard - user perception - intensive and critical care - performance improvement - testing and evaluation - translational research

Background and Significance

Intensive care unit (ICU) clinicians face frequent challenges with managing vast amounts of electronic medical record (EMR) data.[1] The fragmentation of EMR data and data synthesis from diverse sources may result in cognitive overload, leading to miscommunication, increased likelihood of medical errors, and threats to patient safety.[1] [2] [3]

The challenge of managing abundant data sources heightens when clinicians care for multiple critically ill patients simultaneously.[4] [5] [6] Increasing strain in the ICU leads to progressively worse patient outcomes.[6] To maintain situational awareness and provide guideline-based care more consistently, clinicians require a comprehensive view of the entire unit; however, conventional EMRs fall short in fulfilling this purpose.[2] [4] [7]

EMR-based multipatient dashboards have shown promise in supporting clinicians with their workload and improving situational awareness in the ICU.[4] [8] [9] Previous research in this field has demonstrated that technological solutions, such as widely visible EMR-based dashboards for monitoring multiple patients may effectively reduce the time required to prioritize patients in need of immediate care, improve the accuracy of assigning priorities, and alleviate the cognitive burden on clinical staff.[10] [11] These tools may facilitate team communication, overcoming the challenges of fragmented data,[4] and, by enhancing communication and situational awareness, also improve teamwork.[12]

Research to date has primarily evaluated interfaces designed for individual use rather than those intended for display on a large screen and accessible by multiple users within a shared space.[13] [14] Evidence suggests that presenting information from all patients in the unit in a format that can be viewed from a distance facilitates effective communication of data, promotes swift responses to changes in a patient's condition, and garners favorable feedback from ICU clinicians as they carry out their tasks.[7] [10] [15] Centralized multipatient displays are often positioned in prominent locations, such as walls near nursing stations or other areas easily accessible to medical professionals.[7] [10] [15] Nevertheless, only a limited number of these displays have undergone real-world testing in ICU environments and none of the previous reviews systematically evaluated the effects of these dashboards on clinician perceptions.[2] [10] [15] To fill this gap, we conducted a systematic review aimed to understand and evaluate the evidence about these novel digital tools and their ability to reduce task load and time for clinician decision-making in ICU practice.[16] [17]

 

MissingLinkText

Objectives

The objective of this systematic review was to assess the impact of widely accessible ICU dashboards on clinicians' performance, attitudes, perceptions, and satisfaction with these tools in the clinical context.

 

MissingLinkText

Methods

The study methodology adhered to the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines[18] and Synthesis Without Meta-analysis (SWiM) guidelines.[19] Data collection and management were performed using Covidence software.[20] No institutional review board approval was sought since this review involved secondary analysis of previously published work.

Data Sources and Research Strategy

On February 9, 2023, an experienced librarian conducted a comprehensive and organized search on seven databases, including EBSCO CINAHL, Ovid Cochrane Central Register of Controlled Trials, Ovid Embase, IEEE Xplore, Ovid MEDLINE, Scopus, and Web of Science Core Collection. The search was conducted without time, language, or study type constraints, using relevant keywords and standardized index terms. No additional articles were found through manual search strategies using gray literature and the reference lists of selected studies.

 

MissingLinkText

Study Selection

We included quasi-experimental and experimental studies that included ICU clinicians as study participants and both adult and pediatric patients. The intervention of choice was the use of multiuser centralized electronic dashboards in the ICU setting for monitoring multiple patients. The selected outcome was quantitative or qualitative data regarding the impact of the dashboards on the ICU clinician performance and satisfaction compared with the standard practice. The selected outcome metrics were prerounding data collection time, accuracy during multidisciplinary rounds (MDRs), daily documentation adherence, and decision-making time. Additionally, we investigated the attitudes of ICU clinicians toward the utilization of centralized multipatient EMR viewers.

We excluded studies solely examining dashboards on computer screens or personal devices as well as studies that did not assess clinician performance or satisfaction. Additionally, studies describing intervention protocols or dashboard development without testing in ICU settings as well as abstracts and conference reports were excluded.

Titles and abstracts were screened by one reviewer (I.S.), and full texts of the relevant papers were reviewed independently and in duplicate by reviewers (I.S., S.H., J.G., L.R.). Any disagreements were resolved by a third independent reviewer (V.H. or A.B.) or through consensus among all reviewers.

 

MissingLinkText

Data Extraction

Data extraction from each of the five included articles was performed by pairs of independent reviewers (I.S., S.H., J.G., L.R.) using a standardized data abstraction form. Disagreements were resolved by additional reviewers (V.H., A.B.). The data collection form was designed following the Suggested Steps for Developing Rigorous Data Collection Forms and was pilot-tested before its use.[21] [22] We extracted data on participant demographic characteristics, study timeline, study design, study setting, country, intervention description, comparison, and primary and secondary outcomes ([Tables 1] and [2]). Specifically, the data extracted regarding the display intervention encompassed details about the intended users, display attributes, localization, manufacturer, quantity, interactivity, and the types of data showcased.

Table 1 Characteristics of included studies

Study	Country, multi or single center	Design	Duration	Setting	Participants (clinicians)	Participants (patients)	Intervention
Lai et al 2022[29]	Taiwan, single, academic	Cluster-randomized controlled trial	4 wk	2 SICUs with 10 and 8 beds	1. 90 physicians, NPs, nurses, RTs, pharmacists, and dieticians 2. Male 15 (16.7%), female 75 (83.3%) 3. ICU experience (years, %) <1 3 (3.3), 1–2 20 (22.3), 3–4 15 (16.7), 5–9 31 (34.4), >10 21 (23.3)	1. 25 surgical patients with LOS ≥ 7 d; 2. Male 14 (56.0%), female 11 (44.0%) 3. Age 70 (median = 58–73)	i-Dashboard (Advantech) for use during MDRs—wall-mounted 55-inch interactive touch screen
Lee et al 2017[30]	South Korea, single, academic	Cross-sectional study	30 mo	ICU, ER, OR delivery rooms	1. A total of 383 physicians and nurses 2. Male 55 (14.36%), female 328 (85.64%) 3. Length of service (years, %) <5 268 (69.97), 5–9 74 (19.32), >10 41 (10.7)	NA	Bundang Excellent & Smart Touch Board (BEST Board) for retrieving all types of health information from EMR—55-inch touch screen
Pageler et al 2014[26]	California, United States, single, academic	Pre–post	39 mo	PICU with 24 beds	64 staff (pre) 65 staff (post)	860 patients (medical, surgical, transplant, trauma)	Multipatient dashboard of CLABSI prevention to increase compliance with catheter care bundle and decrease CLABSIs—a large monitor at a central location in the PICU
Shaw et al 2015[27]	District of Columbia, United States, single, academic	Pre–post	14 mo	A 44-bed PICU	NA	1. 450 patients, LOS (days) = 1–2 2. Male 1–84 (56%), 2–83 (55%), 87 (58%) 3. Median age (years) = 5	Unit-wide dashboard of admission compliance to improve timeliness of compliance with quality and safety measures—two large flat-screen dashboards located within the PICU
Boudreault et al 2023[28]	Canada, single, academic	Crossover study	During COVID-19 pandemic	A 32-bed PICU	4 intensive care unit pediatricians, 1 RT, and 1 head nurse	NA	Situation awareness-oriented dashboard—55-inch HDTV (1920 × 1080 pixels) screen

Abbreviations: CLABSI, central line-associated bloodstream infection; EMR, electronic medical record; ER, emergency room; ICU, intensive care unit; LOS, length of stay; MDRs, multidisciplinary rounds; NA, not applicable; NPs, nurse practitioners; OR, operating room; PICU, pediatric ICU; RTs, respiratory therapists; SICU, surgical intensive care unit.

Table 2 Key display features and study outcomes according to information display category

Study	Intervention	Number of intervention sessions	Information displayed	Clinician performance	User perception and satisfaction outcomes	Patient outcomes
Lai et al 2022[29]	i-Dashboard (Advantech)	91 MDRs	1. Static mode—lists of patients, information to aid emergency evacuation, on-duty physicians, nurses, and NPs 2. Patient-level data—overview page, RT–pharmacist–dietitian page, menu; colored signaling for values outside the reference ranges 3. Supports built-in automated calculation of severity scores (APACHE II) 4. Time-series data (vital signs, laboratory data, or severity scores) transformed into line charts (the hyperlinks located at the left upper corner of element blocks on the overview page; both fixed (last 24 h/last 3 d) and relative custom time frames are available)	1. Prerounding data gathering time: intervention—4.6 (3.5–5.8) min, control—10.4 (9.1–11.8) p < 0.001 2. Misrepresentations during MDRs: i-Dashboard 0 (0-0) EMR (4 (3–5; p < 0.001, laboratory data 2 (1–3) 0 (0–0) p < 0.001, nonlaboratory data 2 (1–3) 0 (0–0) p < 0.001 3. Effective recommendations, n (%) 0–15 (16.5), 1–34 (37.4), 2–25 (27.5), 3–17 (18.7)	1. Customer satisfaction with dashboard 16.68 (2.02) vs. 15.62 (2.27), p < 0.002 2. Task productivity using i-Dashboard 15.91 (2.28) vs. 14.14 (2.35) using EMR, p < 0.001 3. Management control 16.75 (2.21) vs. 15.03 (2.30), p < 0.001	–
Lee et al 2017[30]	Bundang Excellent & Smart Touch Board (BEST Board)	NA	1.Main screen—a map of beds, vital signs, infections, allergies, equipment, APACHE II score, severity score, ventilator mode, number of admissions/discharges 2. EMR review screen—center area is arranged for data retrieval, both side areas feature menus that provide easy touch access to frequently used items (allergies and infections are marked with alert symbols; test results, including laboratory, radiology, pathology, other specific tests) 3. PACS (allows users to check images and interpretations simultaneously on the same screen 4. Users can retrieve all orders by period and review lists of current medications, administration times and schedules		1. Satisfaction with dashboard 3.30 (0.87) p > 0.05, performance expectancy 3.15 (0.92), effort expectancy 3.43 (0.81), attitude 3.69 (0.89), social influence (0.90), facilitating conditions 3.29 (0.87), intention to use (0.94) 2. Useful for ward rounds 3.57 (0.92), multidisciplinary teamwork 3.41 (0.92), when clinicians explain patient states to patients or caregivers 3.54 (0.98), for educating trainees 3.59 (0.87), for medical research 3.27 (0.91), for understanding the occupation status of hospital rooms 3.61 (0.89), for determining patient status at a glance 3.57 (0.88) 3. Reflects the actual disposition of patients 3.44 (0.88) 4. Different colors (green, yellow, red) are helpful for understanding patients' referrals to other departments, progress of laboratory and imaging tests 3.46 (0.84)	–
Pageler et al 2014[26]	Multipatient dashboard of CLABSI prevention	1942 line-days	1. A real-time overview of critical patient data, specifically highlighting whether there was compliance with the CLABSI prevention guidelines (if there was compliance with all the items, a green light was visible for that patient; if any of the items were out of compliance or required attention, then a yellow or red light displayed	1. Increased compliance with daily documentation of line necessity (from 30% to 73%; p < 0.001 2. Decreased compliance with insertion bundle documentation (from 67 to 62%; p = 0.001	1. Perceptions of team communication—did not reveal a perceived overall improved team communication or knowledge of the CLABSI compliance bundle; p = 0.39	1. Decreased CLABSI rates from 2.6 to 0.7 per 1000-line days; p = 0.03
Shaw et al 2015[27]	Unit-wide dashboard of admission compliance	NA	1. For each component of the PICU Safety Bundle, patients in noncompliance, identified by their room and bed space. 2. The monitor displayed which of the PICU patients were at any given point during the day not in compliance with one or more of the six measures tracked 3. The tool audited a specific location in the chart for each measure, scanned the nursing flow sheets for most of the information but also queried provider orders in the case of DVT prophylaxis and restraint orders (if an order was not found, the patient's room number was displayed on the dashboard as noncompliant) 5. For medical reconciliation, the tool looked for the completion check mark in the medical reconciliation section of the EMR (if items were not charted, the tool would read that information as “not done,” which would lead to either a flagging of the item by presenting the patient in question on the dashboard or the “absence” of the patient	1. Median time from ICU admission to treatment consent - no significant difference at pre implementation (393 min), 1 month post implementation (304 min), and 4-mo postimplementation (202 min); p = 0.13	–	There was no significant difference in the development or worsening of pressure ulcers
Boudreault et al 2023[28]	Situation awareness-oriented dashboard	NA	22 key indicators: one on admission capacity (open to admission [green], selective [yellow], very selective [red], reorientation [black]), 15 at the bedside (COVID [red, yellow] or non-COVID [blue], room number, isolation type, given name or mother's name, bed number, care types, ventilation type, treatments [extracorporeal membrane oxygenation], nurse activities score [Quantis], the pediatric logistic organ dysfunction score (Pelod-2), the medical team identification [Ped A/B/C], the resident/fellow full name, the schedule of appointments [time, place], room and telephone numbers), and six displayed as statistics in the central area (bed occupancy, COVID and non-COVID counts, care types, nurse-to-patient, and doctor-to-patient ratios, waiting to admission and treatment types)	1. Time required to decide if a new patient could be admitted - less than 30 seconds with the situation awareness dashboard and 30 to 60 seconds with the resource management dashboard. 2. Improved situation awareness for all three scenarios.	–	–

Abbreviations: APACHE II - Acute Physiology and Chronic Health Evaluation II score; CLABSI, central line-associated bloodstream infection; DVT, deep venous thrombosis; ECMO, extracorporeal membrane oxygenation; EMR, electronic medical record; ICU, intensive care unit; MDRs, multidisciplinary rounds; NA, not applicable; NPs, nurse practitioners; PACS, Picture Archiving and Communication System; PICU, pediatric ICU; RT, respiratory therapist.

 

MissingLinkText

Risk of Bias/Quality Assessment

Two independent reviewers (I.S., L.R.) assessed risk of bias (ROB) in the included studies using the Cochrane RoB2 tool for randomized studies[23] and Risk of Bias in Non-randomized Studies—of Interventions (ROBINS-I) assessment tool.[24]

 

MissingLinkText

Outcome Measures

We analyzed the difference in performance of ICU clinicians when utilizing centralized multipatient dashboards versus standard practice. The included studies evaluated a variety of metrics, including prerounding data collection time, accuracy during MDRs, daily documentation adherence, and decision-making time. Additionally, we investigated the attitudes of ICU clinicians toward the utilization of centralized multipatient EMR viewers.

 

MissingLinkText

Data Synthesis and Analysis

We were unable to conduct a meta-analysis due to high heterogeneity of study designs, populations, dashboard types, and assessment methods in the included studies. Instead, we narratively synthesized eligible studies following the SWiM guidelines.[19] Studies were grouped based on the most likely sources of heterogeneity, including clinical dashboard type and outcomes assessed in the study.

 

MissingLinkText

 

MissingLinkText

Results

Study Selection

Our search strategy yielded 2,407 studies, with no additional studies identified through manual searches. After removing duplicates, we screened 1,541 abstracts ultimately eliminating 1,497. Relevant review papers and systematic reviews were used for reference mining and excluded, resulting in 44 articles for full-text review.[25] Based on our inclusion and exclusion criteria, five studies were included. The study selection process and reasons for exclusion are presented in the PRISMA diagram ([Fig. 1]).

Fig. 1 Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram.

 

MissingLinkText

Description of Eligible Studies and Participants' Characteristics

All five eligible studies were conducted at single centers. Two of the five studies were conducted in the United States,[26] [27] one in Canada,[28] one in Taiwan,[29] and one in South Korea.[30] One study was a cluster-randomized clinical trial,[29] and four were nonrandomized, including two pre–post studies,[26] [27] one cross-sectional study,[30] and one mixed-methods study using interviews and comparative usability evaluation of the intervention.[28]

Three studies were conducted in the PICU,[26] [27] [28] one in the surgical ICU,[29] and one was conducted in multiple units, including general ICUs, emergency rooms, operating rooms, and delivery rooms.[30] The study duration varied, ranging from 4 weeks[29] to 39 months.[26] One study was conducted during the coronavirus disease 2019 (COVID-19) pandemic and did not report the timeline.[28]

Eligible studies included 542 clinician participants and analyzed clinical data from 1,325 patients. The number of participants in each study ranged from 4 to 383 clinicians and from 25 to 860 patients. Two studies did not specify the number of included patients.[28] [30] Study participants were clinicians with varying levels of experience in the ICU. One study included physicians, nurse practitioners (NP), respiratory therapists (RT), pharmacists, and dietitians,[29] whereas another included ICU physicians, RTs, and nurses.[28] One study included only physicians and nurses.[30] Shaw et al did not provide the number of participants,[27] and Pageler et al did not specify the roles.[26] Study and participant characteristics are shown in [Table 1].

 

MissingLinkText

Risk of Bias/Quality Assessment

The ROB for the randomized study was high due to biases arising from the randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, and selection of the reported results.[29] The ROB among the nonrandomized studies was moderate to high. This was mainly due to confounding biases,[26] [27] [28] [30] biases in selecting participants for the study,[26] [27] [28] biases in classifying interventions,[26] [27] [28] [30] deviations from intended interventions,[28] [29] [30] missing data,[26] [27] [28] and outcome measurement[26] [27] [28] [30] ([Tables 3] and [4]).

Table 3 Revised Cochrane risk-of-bias tool for randomized trials (RoB-2)

Study	Bias arising from the randomization process	Bias arising from the timing of identification or recruitment of participants in a cluster-randomized trial	Bias due to deviations from the intended interventions	Bias due to missing outcome data	Bias in measurement of the outcome	Bias in selection of the reported result	Overall risk of bias
Lai et al 2022[29]	Low	Low	Moderate	Moderate	High	Moderate	High

Table 4 The Risk of Bias in Nonrandomized Studies—of interventions (ROBINS-I) assessment tool

Study	Bias due to confounding	Bias in selection participants into the study	Bias in classification of interventions	Bias due to deviations from intendent interventions	Bias due to missing data	Bias in measurement of outcomes	Bias in selection of the reported results	Overall risk of bias
Lee et al 2017[30]	Moderate	Moderate	Moderate	Moderate	Low	Moderate	Moderate	Moderate
Pageler et al 2014[26]	Moderate	Moderate	High	Moderate	Moderate	High	Moderate	High
Shaw et al 2015[27]	Moderate	Moderate	High	Moderate	Moderate	High	Moderate	High
Boudreault et al 2023[28]	High	Moderate	High	High	Moderate	High	High	High

The certainty of the evidence was very low, primarily due to serious inconsistency across studies,[26] [27] [28] [29] [30] imprecision, and indirectness in the outcome measure. As observed in [Table 5], all included studies had methodological limitations.[26] [27] [28] [29] [30]

Table 5 Certainty of evidence of findings

Outcome	Effect	Number of studies (participants)	Certainty in the evidence (factors affecting certainty)
Clinician performance[26] [27] [28] [29]	Centralized multipatient displays in the ICU improved clinician performance	4 (161)	Very low (inconsistency, imprecision, indirectness, methodological limitations)
User perception and satisfaction[26] [29] [30]	Centralized multipatient displays were associated with increased user perception and satisfaction	3 (538)	Very low (indirectness, imprecision, inconsistency)
Patient outcomes[26] [27]	Centralized multipatient displays were associated with improved patient outcomes	2 (1,310)	Very low (inconsistency, imprecision, indirectness, methodological limitations)

Abbreviation: ICU, intensive care unit.

 

MissingLinkText

Dashboard and Information Display Characteristics

Three of the included studies assessed large screen public ICU displays.[28] [29] [30] Among these, two studies evaluated interactive displays with a touch screen mode intended to improve ICU team performance and foster an interdisciplinary approach.[29] [30]

The i-Dashboard described by Lai et al[29] in a static mode was designed to be a substitute for a station whiteboard with lists of patients, information to aid emergency evacuation, on duty physicians, nurses, and NPs. Patient-level data included an overview page with hyperlink access to time-series data (vital signs, laboratory data), a RT–pharmacist–dietitian page and a link to Picture Archiving and Communication System (PACS). It featured colored signaling for values outside the reference ranges, automated calculation of severity scores. Data were retrieved from different sources in the EMR. The main ICU screen in BESTBoard[30] was a map of patient beds, displaying important information such as vital signs, infections, allergies. Users could also view overall ICU data such as the Acute Physiology and Chronic Health Evaluation (APACHE II) score, equipment usage, ventilator mode, and the number of admissions and discharges. The BESTBoard server could access the EMR database through the application server using the Windows Communication Foundation service. The system was set to automatically log out and return to the main screen, which did not contain personal health information (PHI), if no input signals were received for 5 minutes. It also was installed in a secure space inside the nurse stations so that only authorized health care professionals could see and use it.

Pageler et al and Shaw et al primarily focused on compliance with prevention bundles incorporated into ICU practice.[26] [27] Both studies used existing documentation in the EMR with updates occurring automatically every 5 minutes. Pageler et al introduced a monitor providing a unit-wide overview of critical patient data, highlighting compliance with the maintenance bundle for central line-associated bloodstream infection (CLABSI) prevention.[26] Shaw et al presented the implementation of two large flat-screen dashboards, where one dashboard served as a querying tool, auditing a specific location in the chart for each measure (it scanned nursing flow sheets for most information but also queried provider orders for deep venous thrombosis [DVT] prophylaxis, and restraint orders).[27] Boudreault et al mainly focused on managing the hospital load during the COVID pandemic with a situational awareness-oriented dashboard with 22 key performance indicators, including one on admission capacity (open to admission [green], selective [yellow], very selective [red], reorientation [black]), 15 at the bedside (COVID [red, yellow] or non-COVID [blue], room number, isolation type, given name or mother's name, bed number, care types, ventilation type, treatments [extracorporeal membrane oxygenation], nurse activities score [Quantis], the pediatric logistic organ dysfunction score [Pelod-2], the medical team identification [Ped A/B/C], the resident/fellow full name, the schedule of appointments [time, place], room and telephone numbers), and six displayed as statistics in the central area (bed occupancy, COVID and non-COVID counts, care types, nurse-to-patient, and doctor-to-patient ratios, waiting to admission, and treatment types).[28] This dashboard was not connected to the hospital EMR but used local database and administrative staff had to replicate data entry to keep them up to date.[28]

 

MissingLinkText

Effect of Information Displays on Clinical Care Team Performance, User Perception and Satisfaction

Included studies evaluated the effectiveness of centralized multipatient dashboards in improving various aspects of ICU care such as efficiency,[30] accuracy of data representation during MDRs,[29] awareness of admission processes,[28] monitoring the care of inserted devices,[26] [27] and daily documentation compliance.[27]

We grouped the studies based on the outcomes assessed: studies evaluating ICU clinician performance using i-dashboards,[26] [27] [28] [29] studies evaluating user perception and satisfaction outcomes,[26] [29] [30] and those assessing patient outcomes.[26] [27] Pageler et al assessed user perception and satisfaction with the dashboard, performance, and patient outcomes.[26] Lai et al assessed both user perception and clinician performance outcomes.[29] Pageler et al and Shaw et al assessed both clinician performance and patient outcomes.[26] [27] Lee et al assessed only user satisfaction[30] and Boudreault et al assessed only clinician performance.[28]

Among the studies evaluating ICU clinician performance using multipatient displays, three studies reported time measurements before and after the intervention.[26] [27] [28] One study reported a significant improvement in the efficiency of prerounding data gathering, reducing the median time from 10.4 to 4.6 minutes compared with the established EMR (p < 0.001).[29] Shaw et al assessed the median time from PICU admission to obtaining treatment consent and reported a decrease from 393 minutes (interquartile range [IQR], 70–1,110) to 304 (IQR, 48–1,020) to 202 (IQR, 10–890) in preimplementation, early implementation, and late implementation study period accordingly.[27] However, this decrease was not statistically significant (p = 0.13).[27] The study with situational awareness-focused dashboard did not show a statistically significant difference in decision times on admission process.[28]

Other ICU clinician performance outcomes assessed in the included studies included an increase in effective recommendations regarding the weaning from mechanical ventilation, sedation, dosage/rate adjustment, drug–drug interaction, total parenteral nutrition issues, and route of feeding initiated by RTs, pharmacists, and dietitians, accordingly (all p < 0.001).[29] Lai et al showed a reduction in data misrepresentations during MDRs (median = 0 [IQR, 0–0] vs. 4 [IQR, 3–5]; p < 0.001) when comparing with established EMR system (Philips IntelliSpace Critical Care & Anesthesia Information System and the Hospital Information System, including the Laboratory Information System and the PACS).[29] Shaw et al reported improved completion of medication reconciliation, from 80% in the preintervention to 93 and 92%, respectively, in the subsequent two periods of early and late implementation (p = 0.002).[27] However, there was no significant difference in use of restraint orders and DVT prophylaxis.[27] This study also reported an improvement in the management of long-term indwelling urinary catheters; the number of patients with urinary catheters in place > 96 hours decreased from 16 (32%) to 11 (19%) (p = 0.01).[27] Pageler et al reported that compliance with specific bundle elements increased in daily documentation of line necessity but decreased in insertion bundle documentation.[26]

Pageler et al and Shaw et al reported patient outcomes related to multipatient dashboards implementation. Pageler et al showed that CLABSI rates decreased significantly from 2.6 per 1000-line days before the intervention to 0.7 per 1000-line days after the intervention (p = 0.03).[26] Shaw et al reported that there was no significant difference in the development or worsening of pressure ulcers.[27]

Among the studies evaluating user perception and satisfaction outcomes, survey results in the Lai et al study indicated that participants preferred using i-Dashboard to enhance care plan development and team participation during MDRs and showed higher customer satisfaction with the clinical dashboard 16.68 (standard deviation [SD] = 2.02), compared with 15.62 (SD = 2.27) with standard practice) (p = 0.002).[29] Lee at al presented results indicating high satisfaction rates with BESTBoard among dashboard users, with a mean score of 3.3 (SD = 0.87).[30] Users found BESTBoard to be a useful tool for team round visits, interdisciplinary team approach, and collecting the status of hospital rooms.[30] However, among older users with length of service >10 years, 41 (10.7% of all participants) were less likely to consider it useful for an interdisciplinary team approach and collecting hospital room status (p < 0.05).[30]

Pageler et al assessed the effect of this intervention on perceptions of team communication. However, the survey data did not reveal a perceived overall improved team communication or knowledge of the CLABSI compliance bundle (p = 0.39).[26]

 

MissingLinkText

 

MissingLinkText

Discussion

The purpose of this systematic review was to assess the impact of centralized dashboards for monitoring multiple patients on ICU clinician performance, attitudes, perceptions, and satisfaction with these tools in the clinical context.[26] [27] [28] [29] [30] There were significant methodological limitations, differences in study designs, and measurement approaches.[26] [27] [28] [29] [30] Overall, included studies indicated that electronic multipatient dashboards were positively perceived.[26] [27] [28] [29] However, it is important to note that one study did not show any effect.[28]

To date several reviews and meta-analyses have examined the use of display systems in the ICU, aiming to understand their impact on performance, satisfaction, and the attitude of ICU clinicians toward the integration of these displays into clinical practice.[13] [14] [17] Nevertheless, our study differs from these reviews in several aspects. The systematic review by Waller et al stated that multipatient dashboards may effectively propagate the representational state of all patients in a unit regarding prespecified activities and goals.[14] However, user satisfaction was not assessed.[14] Moreover, not all the included studies in that systematic review subjected monitors to evaluation within a real ICU.[14] An essential component of translational research involves closing the divide between laboratory-based studies and clinical practice settings.[15] Although certain innovations have demonstrated potential in simulated settings, they still require evaluation in an environment that more closely resembles the ICU setting.[7] [10] [15] In their study, Herasevich et al did not find an overall significant association between the utilization of health information technologies and improved mortality or lengths of stay (LOS) in the randomized controlled trial meta-analyses. However, in the pre–post study meta-analyses of that work, the use of such tools was linked to reduced mortality and increased LOS.[17] In contrast to our systematic review the before-mentioned studies also investigated single-user EMR-based interfaces and far-view monitors designed for individual patient monitoring.[13] [14] [17]

ICU health care professionals operate within a demanding and fast-paced environment that increases the risk of errors, compromising patients' safety.[31] With the advancement of technology, ICUs generate an ever-expanding volume of data that medical professionals must monitor and interpret.[32] Implemented within the ICU, a dashboard allows clinicians to quickly identify changes in a patient's condition that requires intervention. Clinicians can choose to dive deeper into the EMR data or refer to the dashboard at a later point to review changes.[13] Centralized multipatient dashboards may offer a solution to fragmented provider-centered care, a common cause of medical errors within the ICU.[33] Just as planning the layout of the unit to optimize the visibility of patients is essential, locating the dashboards displaying relevant information in easily visible areas may support patient care.[2] These interactive dashboards serve as an efficient visualization mechanisms for presenting data, enabling relevant stakeholders to engage in rapid decision-making. Multiuser displays allow health care providers to monitor more than one patient at a time by presenting data from multiple individual bedside monitors.[15] [26] [27] [28] [29] [30] Such tools are also helpful in organizing the admission process, especially in unusual contexts, such as the simultaneous arrival of patients with severe COVID-19.[28] [34] In particular, multiuser solutions may assist in maintaining situational awareness across the unit. The integration of multipatient dashboards that are visible to the entire health care team may also enhance clinician understanding of their daily responsibilities.[14] This approach can cultivate a collaborative and well-informed working environment that may facilitate cohesive and thorough management of critically ill patients.[26] [27]

As technology evolves rapidly, the potential impact of emerging technologies (e.g., artificial intelligence, machine learning) on the development and functionality of multipatient dashboards may help ICU staff to effectively allocate resources, such as staff and equipment, based on patient needs and workload demands when utilizing statistical or epidemiological models to forecast patient admissions, transfers, and discharges, pending procedures, facilitating efficient patient flow throughout the ICU.[35] By aggregating and synthesizing diverse data, artificial intelligence algorithms may help to identify patterns and trends analyzing historical patient data, which may support the development of predictive models that can forecast patient deterioration or outcomes and complete score calculations.[36]

The issue that needs consideration when discussing centralized multiuser solutions is patient privacy. The implementation of these technological solutions can elicit concerns regarding the data displayed, which may inadvertently reveal information to nonclinicians in the ICU, such as visiting family members and other stakeholders who are not directly involved in patient care. Positioning screens in more “off stage” areas visible to health care workers and less visible to families may address security and privacy issues same as implementing strict access controls to ensure that only authorized personnel have access to interact with board. Access permissions should be granted based on roles, ensuring varying levels of access aligned with users' responsibilities and required information.

Most dashboards typically employ lists of indicators presented in tabular format with color-coded abnormal values, whereas certain tools incorporate alternative visual representations, such as bar graphs and pie charts, to convey information.[37] It is crucial that the key indicators featured on dashboards meet the information needs of clinicians. Therefore, future research should focus on gathering direct feedback that involves actively engaging health care professionals in the design process, soliciting their input on the types of data and visualizations that would best support their decision-making processes and maintain patient privacy.

The use of multipatient viewers in clinical practice has been shown to lead to improved compliance with care protocols, resulting in positive outcomes.[26] [27] Particularly in the ICU they may serve to significantly reduce the time and effort required to review previous records, simplifying the process.[38] Makic et al and Zaydfudim et al suggested that dashboards have the potential to help clinicians address overlapping risks and potential patient harm in the unit by managing multiple values for preidentified health care-associated infection risk.[39] [40] This approach contrasts with information being fragmented among multiple pages in the EMR.[39] [40] Additionally, such tools can be particularly powerful when representing dynamic time frame data. For instance, an e-antibiogram, generated and regularly updated from both laboratory and pharmacy data, presents a viable alternative to the static, manually compiled antibiogram typically produced annually.[41]

The key to clinical dashboard usability and usage is data recency.[42] In the studies included in our systematic review, the majority of dashboards were connected to hospital EMR systems, with only one requiring manual updates. As the EMR systems varied among the included studies, it is challenging to determine if there is a superior EMR system for facilitating dashboards. However, exploring how each ICU's EMR may influence the usefulness and usability of dashboards is an interesting subject for future research.

User-centered design is also an important component of digital solutions, especially when anticipating its use during MDRs. This includes considerations about electronic ink displays that make information easy to read from various viewing angles and lighting conditions ensuring health care providers can quickly and accurately interpret information displayed.[43]

This systematic review presents fresh insights to the field focusing its efforts on the systematic assessment of centralized multipatient dashboards, representing a distinct category of display systems.[26] [27] [28] [29] [30] The synthesis of available evidence from these studies has shown a trend suggesting improvements in ICU clinical practices associated with the adoption of these solutions.[26] [27] [29] [30] Furthermore, this study highlights the positive views of clinicians toward the integration of these displays into their daily clinical routines.[26] [29] [30] Such findings underscore the potential value of centralized multipatient dashboards in enhancing both professional performance and clinician satisfaction with new tools within the challenging and dynamic context of the ICU.[26] [27] [28] [29] [30]

When considering the implementation of multipatient dashboards in the ICU, it is crucial to carefully evaluate potential hazards and limitations.[1] [8] [15] There may be concerns related to data security and privacy, user acceptance, costs, training requirements, so that getting clinicians to accept them can be very challenging. Furthermore, specific medical situations or patient needs may demand modifications to the technology.[15] [34] Considering the implementation and training process, from our perspective, integrating these dashboards into rounding could aid in further investigating their effectiveness. However, implementation may vary greatly in each ICU depending on workflow.[44] This can be achieved by providing the necessary information and directly or through survey soliciting input from health care professionals regarding the specific data and pertinent information they need from a multiuser multipatient dashboard.[8] Such insights enable clinicians to make well-informed decisions about integrating this technology into their clinical practice.[34] [37] [42]

Strengths

We believe the included studies greatest strengths lies in the assessment of the multiuser type multipatient dashboards within an actual ICU environment. This systematic review was conducted in collaboration with a professional, experienced medical librarian with serial input from the authors. To our knowledge, this is the first study of its kind and contributes a unique perspective to the literature. We specifically chose to limit our review to studies of multiuser type multipatient dashboards in which user perception, performance, and patient outcomes were assessed. We excluded studies that assessed multiuser dashboards in non-ICU settings, and those assessing multipatient dashboard mockups instead of the unit-wide displays.

 

MissingLinkText

Limitations

The number of eligible studies evaluating the impact of centralized dashboards for monitoring multiple patients on ICU clinicians' performance and satisfaction was small. We could not perform a meta-analysis due to the high heterogeneity among the included studies regarding design, populations, dashboard types, and outcome measures. While the narrative synthesis approach is appropriate given this heterogeneity, future reviews could benefit from efforts to categorize studies more precisely or identify common metrics that allow for meta-analysis, enhancing the comparability of results. In addition, there were high ROB and confounding among before–after studies and the certainty of evidence was very low, due to inconsistencies and methodological limitations. Future studies should aim for higher-quality research designs, such as pragmatic clinical trials or well-designed observational studies, to strengthen the evidence base.

 

MissingLinkText

 

MissingLinkText

Conclusion

This systematic review evaluated the impact of centralized multipatient dashboards on ICU clinician performance, perceptions, satisfaction with these tools in the clinical setting and patient outcomes. Most of the included studies indicated that the implementation of centrally visible multipatient dashboards may improve ICU clinician performance or efficiency. In the studies that evaluated staff satisfaction with the new dashboard overall there was a positive perception of these tools. However, given the small study sample and the very low certainty of evidence with inconsistency across studies, inferences should be interpreted with caution. As there is a wide range of dashboard interfaces noted in eligible studies in this systematic review, input from end users at a very early stage of the design process regarding the data and information they expect to observe in a new tool is needed. Future work seeking clinician input using mixed methods approaches to study the potential impact of the displays on improving situational awareness, driving patient outcomes, improving health care worker satisfaction, and reducing information overload would be valuable.

 

MissingLinkText

Clinically Relevant Statement

This systematic review provides detailed insights into the use of centralized multiuser dashboards in real-world ICU settings for monitoring multiple patients. This analysis is foundational in informing the design of both existing and future dashboards for ICU clinicians.

 

MissingLinkText

Multiple Choice Questions

1. By using centralized multipatient dashboards clinicians may be able to:

   • Observe real time patient information from numerous digital sources across the unit.

   • Observe patient protected health information displayed.

   • Only review EMR data.

   • Review patients' referrals to other departments and the progress of laboratory and imaging.

   Correct answer: The correct answer is option a. The tools described in the included studies were designed to cope with fragmented electronic medical data in the ICU and offer various patient information monitoring, including laboratory and imaging data. The included studies evaluated dashboards with no patient-protected data displayed.

2. Were the centralized multipatient dashboards used for patient monitoring during the COVID-19 pandemic?

   • An overall lower usage during the pandemic compared to the prepandemic period was observed.

   • Those tools were not in use during the pandemic period.

   • One study reported that the COVID-19 pandemic boosted the use of this digital technology.

   • All included studies observed no difference in usage when comparing the pandemic and prepandemic period.

   Correct answer: The correct answer is option c. One of the included studies[28] assessed the impact of centralized dashboards on ICU workflow during the COVID-19 pandemic.

 

MissingLinkText

 

MissingLinkText

Conflict of Interest

None declared.

Acknowledgments

The authors would like to thank the librarian, Gerberi, Dana J, MLIS, AHIP, for the comprehensive search provided.

Protection of Human and Animal Subjects

Not applicable.

• References

• 1 Lindroth HL, Pinevich Y, Barwise AK. et al. Information and data visualization needs among direct care nurses in the intensive care unit. Appl Clin Inform 2022; 13 (05) 1207-1213
• 2 Egan M. Clinical dashboards: impact on workflow, care quality, and patient safety. Crit Care Nurs Q 2006; 29 (04) 354-361
• 3 Nijor S, Rallis G, Lad N, Gokcen E. Patient safety issues from information overload in electronic medical records. J Patient Saf 2022; 18 (06) e999-e1003
• 4 Herasevich S, Pinevich Y, Lindroth HL, Herasevich V, Pickering BW, Barwise AK. Who needs clinician attention first? A qualitative study of critical care clinicians' needs that enable the prioritization of care for populations of acutely ill patients. Int J Med Inform 2023; 177: 105118
• 5 Bauer DT, Guerlain S, Brown PJ. The design and evaluation of a graphical display for laboratory data. J Am Med Inform Assoc 2010; 17 (04) 416-424
• 6 Pilcher DV, Hensman T, Bihari S. et al. Measuring the impact of ICU strain on mortality, after-hours discharge, discharge delay, interhospital transfer, and readmission in Australia with the activity index. Crit Care Med 2023; 51 (12) 1623-1637
• 7 Görges M, Westenskow DR, Markewitz BA. Evaluation of an integrated intensive care unit monitoring display by critical care fellow physicians. J Clin Monit Comput 2012; 26 (06) 429-436
• 8 Barwise A, Caples S, Jensen J, Pickering B, Herasevich V. Information needs for the rapid response team electronic clinical tool. BMC Med Inform Decis Mak 2017; 17 (01) 142
• 9 Reese TJ, Kawamoto K, Del Fiol G. et al. Approaching the design of an information display to support critical care. 2017 IEEE International Conference on Healthcare Informatics (ICHI) 2017: 439-443
• 10 Görges M, Kück K, Koch SH, Agutter J, Westenskow DR. A far-view intensive care unit monitoring display enables faster triage. Dimens Crit Care Nurs 2011; 30 (04) 206-217
• 11 Alhmoud B, Melley D, Khan N, Bonicci T, Patel R, Banerjee A. Evaluating a novel, integrative dashboard for health professionals' performance in managing deteriorating patients: a quality improvement project. BMJ Open Qual 2022; 11 (04) 9
• 12 Miller ME, Scholl G, Corby S, Mohan V, Gold JA. The impact of electronic health record-based simulation during intern boot camp: interventional study. JMIR Med Educ 2021; 7 (01) e25828
• 13 Khairat SS, Dukkipati A, Lauria HA, Bice T, Travers D, Carson SS. The impact of visualization dashboards on quality of care and clinician satisfaction: integrative literature review. JMIR Human Factors 2018; 5 (02) e22
• 14 Waller RG, Wright MC, Segall N. et al. Novel displays of patient information in critical care settings: a systematic review. J Am Med Inform Assoc 2019; 26 (05) 479-489
• 15 Thomas MM, Kannampallil T, Abraham J, Marai GI. Echo: a large display interactive visualization of ICU data for effective care handoffs. 2017 IEEE Workshop Vis Anal Healthc VAHC (2017) 2017 October: 47-54
• 16 Herasevich S, Pinevich Y, Lipatov K. et al. Evaluation of digital health strategy to support clinician-led critically ill patient population management: a randomized crossover study. Crit Care Explor 2023; 5 (05) e0909
• 17 Herasevich S, Lipatov K, Pinevich Y. et al. The impact of health information technology for early detection of patient deterioration on mortality and length of stay in the hospital acute care setting: systematic review and meta-analysis. Crit Care Med 2022; 50 (08) 1198-1209
• 18 Page MJ, McKenzie JE, Bossuyt PM. et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021; 372: n71
• 19 Campbell M, McKenzie JE, Sowden A. et al. Synthesis without meta-analysis (SWiM) in systematic reviews: reporting guideline. BMJ 2020; 368: l6890
• 20 Covidence systematic review software. Veritas Health Innovation, Melbourne, Australia. Accessed May 6, 2024 at www.covidence.org
• 21 Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ. et al., eds. Cochrane Handbook for Systematic Reviews of Interventions version 6.4 (updated August 2023). Cochrane 2023. Accessed at: www.training.cochrane.org/handbook
• 22 Moher D, Liberati A, Tetzlaff J, Altman DG. PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009; 6 (07) e1000097
• 23 Sterne JAC, Savović J, Page MJ. et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ 2019; 366: l4898
• 24 Sterne JA, Hernán MA, Reeves BC. et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ 2016; 355: i4919
• 25 Horsley T, Dingwall O, Sampson M. Checking reference lists to find additional studies for systematic reviews. Cochrane Database Syst Rev 2011; 2011 (08) MR000026
• 26 Pageler NM, Longhurst CA, Wood M. et al. Use of electronic medical record-enhanced checklist and electronic dashboard to decrease CLABSIs. Pediatrics 2014; 133 (03) e738-e746
• 27 Shaw SJ, Jacobs B, Stockwell DC, Futterman C, Spaeder MC. Effect of a real-time pediatric ICU safety bundle dashboard on quality improvement measures. Jt Comm J Qual Patient Saf 2015; 41 (09) 414-420
• 28 Boudreault L, Hebert-Lavoie M, Ung K. et al. Situation awareness-oriented dashboard in ICUs in support of resource management in time of pandemics. IEEE J Transl Eng Health Med 2023; 11: 151-160
• 29 Lai CH, Li KW, Hu FW. et al. Integration of an intensive care unit visualization dashboard (i-Dashboard) as a platform to facilitate multidisciplinary rounds: cluster-randomized controlled trial. J Med Internet Res 2022; 24 (05) e35981
• 30 Lee K, Jung SY, Hwang H. et al. A novel concept for integrating and delivering health information using a comprehensive digital dashboard: an analysis of healthcare professionals' intention to adopt a new system and the trend of its real usage. Int J Med Inform 2017; 97: 98-108
• 31 Pickering BW, Herasevich V, Ahmed A, Gajic O. Novel representation of clinical information in the ICU: developing user interfaces which reduce information overload. Appl Clin Inform 2010; 1 (02) 116-131
• 32 Manor-Shulman O, Beyene J, Frndova H, Parshuram CS. Quantifying the volume of documented clinical information in critical illness. J Crit Care 2008; 23 (02) 245-250
• 33 Pickering BW, Litell JM, Herasevich V, Gajic O. Clinical review: the hospital of the future - building intelligent environments to facilitate safe and effective acute care delivery. Crit Care 2012; 16 (02) 220
• 34 Fareed N, Swoboda CM, Chen S, Potter E, Wu DTY, Sieck CJ. U.S. COVID-19 state government public dashboards: an expert review. Appl Clin Inform 2021; 12 (02) 208-221
• 35 Mao Z, Liu C, Li Q, Cui Y, Zhou F. Intelligent intensive care unit: current and future trends. Intensive Care Res 2023; 3: 1-7
• 36 Yoon JH, Pinsky MR, Clermont G. Artificial intelligence in critical care medicine. Crit Care 2022; 26 (01) 75
• 37 Dowding D, Randell R, Gardner P. et al. Dashboards for improving patient care: review of the literature. Int J Med Inform 2015; 84 (02) 87-100
• 38 Nelson O, Sturgis B, Gilbert K. et al. A visual analytics dashboard to summarize serial anesthesia records in pediatric radiation treatment. Appl Clin Inform 2019; 10 (04) 563-569
• 39 Makic MBF, Stevens KR, Gritz RM. et al. Dashboard design to identify and balance competing risk of multiple hospital-acquired conditions. Appl Clin Inform 2022; 13 (03) 621-631
• 40 Zaydfudim V, Dossett LA, Starmer JM. et al. Implementation of a real-time compliance dashboard to help reduce SICU ventilator-associated pneumonia with the ventilator bundle. Arch Surg 2009; 144 (07) 656-662
• 41 Simpao AF, Ahumada LM, Larru Martinez B. et al. Design and implementation of a visual analytics electronic antibiogram within an electronic health record system at a tertiary pediatric hospital. Appl Clin Inform 2018; 9 (01) 37-45
• 42 Lim HC, Austin JA, van der Vegt AH. et al. Toward a learning health care system: a systematic review and evidence-based conceptual framework for implementation of clinical analytics in a digital hospital. Appl Clin Inform 2022; 13 (02) 339-354
• 43 Jambaulikar GD, Marshall A, Hasdianda MA. et al. Electronic paper displays in hospital operations: proposal for deployment and implementation. JMIR Form Res 2021; 5 (08) e30862
• 44 Zoellner JM, Porter KJ. Chapter 6 - Translational research: concepts and methods in dissemination and implementation research. In: Coulston AM. et al, eds. Nutrition in the Prevention and Treatment of Disease, 4th ed. Academic Press; 2017: 125-143

Publication History

Received: 22 January 2024

Accepted: 03 April 2024

Accepted Manuscript online:
04 April 2024

Article published online:
29 May 2024