Improving Providers' Decision-Making and Reducing Information Overload Using Information Visualization in EHRs

This study aims to test the efficacy of an information visualization dashboard on decision-making using a randomized controlled trial with crossover.

This study aims to test the efficacy of using an information visualization dashboard on ICU providers' decision-making, efficiency, and performance compared to their institutional EHR through a randomized control trial with crossover.

Study Overview

• Status: Completed
• Conditions: Critical Care; Usability; Electronic Health Records; Information Seeking Behavior
• Intervention / Treatment: Device: AWARE; Device: Electronic Health Record (EHR)

Detailed Description

Electronic Health Records (EHRs) are a major source of data for ICU clinicians. Synthesizing complex, electronic patient data is key to effective care delivery. EHRs contain both a record of past medical data and present continuous flows of new clinical data from various sources such as physiological inputs, laboratory results, imaging studies, and clinician notes. This complex, continuous stream of patient data can contribute to information overload, which can create barriers to the key cognitive tasks of data identification, extraction, and interpretation. Intensive Care Unit (ICU) providers must quickly synthesize data from more than 200 variables during critical care rounds, with critically ill patients generating a median of 1348 individual data points per day. Information overload has been identified as a key factor in the misinterpretation of data, leading to medical errors such as misdiagnoses. Improving our understanding of information overload-and investigations into new efforts to minimize it-can improve clinician workflow and productivity as well as patient safety. The objective of this study is to explore the impact of current data representation in the EHR on ICU clinicians' cognitive workload, performance, and satisfaction. The research design uses a mixed methods approach, including eye-tracking assessment and surveys, to assess the efficacy of current EHR interfaces for ICU clinicians in live and simulated environments.

The investigators will randomize consented participants into two groups: the control (EHR) group and the intervention (AWARE) group. During the Randomized Controlled Trial (RCT) crossover, providers in each group will review the same patient records and will perform the same tasks, and complete the same survey instruments. Providers in the control groups will review two patient cases in their institution EHR (Epic or Cerner) first and then, two new cases in AWARE, and providers in the intervention group will review two patient cases in AWARE and then two new cases in the EHR. Cases will be randomized to eliminate order bias and selection bias.

For the cases in the EHR, participants will navigate through the EHR as per their usual routine in the ICU, with no added training sessions before the study. For the cases in AWARE, participants will receive a short training presentation by the study team, explaining the functionality and design of AWARE. Also, a day before each session, the RA will send an email to the provider with a short demonstration video of AWARE to become familiar with the tool.

The study will be conducted in simulation or Biobehavioral labs at each site. The PI, or research assistant, will explain the study procedure and obtain consent; participants will be asked to use Tobii Pro Fusion during the session, participants will not need to wear anything, the eye-tracker will be mounted to the monitor. During the session, participants will review 2 patient cases in their institutional EHR and two in AWARE. After the participant completes the patient record review, the research assistant (RA) will ask the participant a series of decision-based Q&A activities requiring verbal responses or task completion in the EHR or AWARE. The provider may use the EHR or AWARE to complete the Q&A activity. Cases will be in random order for each participant to avoid selection or order bias. After usability testing, the investigators will ask the participant to fill out the NASA-TLX survey and the System Usability Scale (SUS). The NASA-TLX measures the perceived workload of using the EHR, and the SUS measures the level of satisfaction as a result of using the EHR. All sessions will be recorded.

• Study Type: Interventional
• Enrollment (Actual): 113
• Phase: Not Applicable

Contacts and Locations

Study Locations

• United States
  • North Carolina
    • Chapel Hill, North Carolina, United States, 27514
      • University of North Carolina at Chapel Hill

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

Description

Inclusion Criteria:

ICU physicians and advanced practice providers (APPs),

• active full time ICU service,
• use an institutional EHR (Epic or Cerner) to deliver care, and
• reads and speaks English;

Residents

• prior ICU rotation experience,
• use an institutional EHR (Epic or Cerner) to deliver care, and
• reads and speaks English;

Combined Exclusion Criteria:

• Non-ICU Physicians or APPs,
• residents with no prior ICU experience

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Diagnostic
• Allocation: Randomized
• Interventional Model: Crossover Assignment
• Masking: Single

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Experimental: EHR (Control), then AWARE; In this hour-long task, participants in this arm will first review and complete tasks using the institutional EHR. Then after an approximate 5 minute rest, participants will then review and complete tasks using AWARE.	Device: AWARE; AWARE is a platform with a visualization dashboard that sits on top of pre-existing, comprehensive EHR systems. AWARE provides clinical decision support tools for the ICU, Emergency Departments and other critical care areas right at the patient's bedside.; Device: Electronic Health Record (EHR); An Electronic Health Record (EHR) is an electronic version of a patients medical history, that is maintained by the provider over time, and may include all of the key administrative clinical data relevant to that persons care under a particular provider, including demographics, progress notes, problems, medications, vital signs, past medical history, immunizations, laboratory data and radiology reports.
Experimental: AWARE Intervention, then EHR (Control); In this hour-long task, participants in this arm will first review and complete tasks using the AWARE intervention. Then after an approximate 5 minute rest, participants will then review and complete tasks using the institutional EHR.	Device: AWARE; AWARE is a platform with a visualization dashboard that sits on top of pre-existing, comprehensive EHR systems. AWARE provides clinical decision support tools for the ICU, Emergency Departments and other critical care areas right at the patient's bedside.; Device: Electronic Health Record (EHR); An Electronic Health Record (EHR) is an electronic version of a patients medical history, that is maintained by the provider over time, and may include all of the key administrative clinical data relevant to that persons care under a particular provider, including demographics, progress notes, problems, medications, vital signs, past medical history, immunizations, laboratory data and radiology reports.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Number of Clicks Per Case	To measure efficiency, participants' number of mouse clicks to complete case review and tasks will be recorded.	Duration of Single Usability Session, up to 25 minutes
Time to Task Completion	To assess the amount of time required to make a decision (i.e., complete a task), participants will be asked to find the answers to clinical questions and the average amount of time needed to answer will be reported.	Duration of Single Usability Session, up to 25 minutes
Overall Mean Task Scores	To measure performance, participants will be scored on the overall average responses to 8 questions using the following: 0 points for an incorrect response, 0.5 points for a partially correct response, and 1 point for correct responses. Overall mean score for the 8 questions combined is reported.	Duration of Single Usability Session, up to 25 minutes
Time to Case Completion	To measure efficiency, participants' time to complete a patient case (chart review and tasks) will be recorded and reported in minutes.	Duration of Single Usability Session, up to 25 minutes
Mean Number of Fatigue Instances Per Case	Physiologic fatigue will be measured using an eye tracking device to capture changes in pupil size (which represents cognitive fatigue). A fatigue instance was recorded whenever the participant's pupil diameter dropped below the subject-specific fatigue threshold. For every provider-interface combination, the investigators calculated the mean and standard deviation of these one-second pupil diameter values and defined the subject-specific fatigue threshold as the individual mean minus 1.5 standard deviations.	Duration of Single Usability Session, up to 25 minutes
System Usability Scale Survey Score	Perceived satisfaction measured using the System Usability Scale survey. Values range from a minimum of 0 to a maximum of 10. Higher scores indicate higher satisfaction (better). The overall score is reported and ranges from 0 to 100. It is expected to take 2-3 minutes to complete this survey.	Immediately following each intervention, within approximately 2-3 minutes

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Perceived Workload Score	The NASA-Task Load Index (TLX) measures perceived workload using 6 subscale questions. Scores are reported as an overall total ranging from a minimum of 0 to a maximum of 100. Higher scores indicate higher workload (worse). It is expected to take 1-2 minutes to complete this index.	Immediately following each intervention, within approximately 1-2 minutes
Number of Eye Fixations on Screen	Information seeking load measures the amount of information being processed by participants using an eye-tracking device to measure fixation points (which represents thought-processing).	Duration of Single Usability Session, up to 25 minutes

Collaborators and Investigators

• Sponsor: University of North Carolina, Chapel Hill
• Collaborators: Mayo Clinic; National Library of Medicine (NLM); University of Pittsburgh Medical Center; MedStar Health
• Investigators: Principal Investigator: Saif Khairat, PhD, University of North Carolina, Chapel Hill

Publications and helpful links

General Publications

• Khairat S, Morelli J, Boynton MH, Bice T, Gold JA, Carson SS. Investigating Information Visualization to Combat Information Overload in Electronic Health Records: Protocol for a Randomized Controlled Trial. JMIR Res Protoc. 2025 Sep 9;14:e74247. doi: 10.2196/74247.

Study record dates

Study Major Dates

• Study Start (Actual): July 24, 2023
• Primary Completion (Actual): May 30, 2025
• Study Completion (Actual): May 30, 2025

Study Registration Dates

• First Submitted: June 30, 2023
• First Submitted That Met QC Criteria: June 30, 2023
• First Posted (Actual): July 10, 2023

Study Record Updates

• Last Update Posted (Actual): May 5, 2026
• Last Update Submitted That Met QC Criteria: May 1, 2026
• Last Verified: March 1, 2026

More Information

Terms related to this study

• Keywords: Visualization
• Additional Relevant MeSH Terms: Behavior; Communication; Information Seeking Behavior; Organization and Administration; Health Services Administration; Health Care Quality, Access, and Evaluation; Investigative Techniques; Epidemiologic Methods; Data Collection; Health Care Evaluation Mechanisms; Quality of Health Care; Public Health; Environment and Public Health; Records; Medical Records Systems, Computerized; Medical Records; Electronic Health Records
• Other Study ID Numbers: 20-3384; R01LM013606 (U.S. NIH Grant/Contract)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: YES
• IPD Plan Description: Deidentified individual data that supports the results will be shared beginning 12 to 24 months following publication provided the investigator who proposes to use the data has a methodologically sound proposal, approval from an Institutional Review Board (IRB), Independent Ethics Committee (IEC), or Research Ethics Board (REB), as applicable, and executes a data use/sharing agreement with UNC.
• IPD Sharing Time Frame: beginning 12 months after and continuing for 24 months following publication
• IPD Sharing Access Criteria: Investigator has IRB, IEC, or REB approved methodologically sound proposal and an executed data use/sharing agreement with UNC.

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: Yes
• product manufactured in and exported from the U.S.: Yes