ipopba - stock.adobe.com
From 156 clicks to 1 hover bubble: How Carilion Clinic operationalized clinical trial data
Carilion Clinic transformed a labor-intensive screening process into a streamlined digital workflow.
After finishing her hospital rounds, Lana Wahid, M.D., vice chair of research at Carilion Clinic, often spent evenings reviewing EHRs to identify patients who might qualify for clinical trials.
When her team later analyzed one of those workflows, they found that determining whether a single patient qualified for a COVID-19 trial required 156 clicks within the EHR.
"The challenge wasn't a lack of data. It was having too much data in too many places," Wahid said.
To address the issue, Carilion partnered with Nordic, a healthcare IT consulting firm, to develop EHR-based dashboards that help researchers and clinicians identify relevant information more quickly.
Too much data, too many places
The challenge emerged in 2021 when Wahid expanded a clinical research program focused on hospitalized patients.
Identifying patients for studies required manually reviewing laboratory results, clinical histories and documentation to evaluate inclusion and exclusion criteria. Initially, Wahid handled much of the screening work herself. Later, as coordinators joined the program, she noticed similar patients were being screened repeatedly across multiple studies.
"I kept thinking there's got to be a different way. In the era of data and data informatics, how can we leverage the electronic health record?" Wahid said. "When I got my next round of funding, I then decided, instead of hiring more people to chart search, to hire IT specialists."
Carilion brought in Nordic, whose advisors help healthcare organizations optimize clinical, operational and EHR workflows.
Building dashboards inside the EHR
Together, Carilion and Nordic developed dashboards within the clinic's Epic EHR, tailored to studies in specific therapeutic areas, including heart failure and pneumonia.
The dashboards compare patient data against study criteria and display visual indicators for potential candidates. Patients meeting criteria appear with green icons, exclusions appear in red and patients requiring additional evaluation appear in yellow.
According to Jason Griffin, managing director of digital health strategy and cybersecurity practice leader at Nordic, understanding how clinicians and researchers gathered information was one of the most important parts of building the dashboards.
"The biggest challenge was understanding the workflow," Griffin said. "We spent a lot of time working with stakeholders to understand how clinicians and researchers gather information, where bottlenecks exist and which information would be most valuable."
According to Wahid, the new workflow also allows research teams to identify patients eligible for multiple studies. This gives those patients the opportunity to select which study interests them most, rather than requiring investigators to make that decision.
Faster screening changed research operations
The redesigned workflow reduced the 156-click screening process to a single hover bubble containing the information that researchers needed.
Before the dashboards were implemented, research coordinators often spent entire mornings screening patients before heading to the bedside to discuss enrollment opportunities. By then, some patients had been discharged, and physicians had already completed rounds.
However, after implementing the new dashboards, screening across over 20 clinical trials could be completed in about an hour.
"By 9 o'clock in the morning, the messages are going out to the attendings and then by 9:30, 10 o'clock we're at the bedside," Wahid said.
"Our consent rate skyrocketed once we understood the impact of operationalizing the screening process [to make it] faster," Wahid said. "It allowed us to become the top-enrolling site on many international NIH trials."
From screening to clinical impact
The project soon expanded beyond clinical trial screening into the patient safety arena. This move involved adding new features to the dashboards.
During the COVID-19 pandemic, Wahid recalled a patient enrolled in an anticoagulation trial who experienced a significant bleeding event. While the clinical team was aware of the change and managing the patient's care appropriately, the research team did not learn about it until later in the day.
The patient later died from a retroperitoneal bleed despite the care team's efforts.
For Wahid, the incident underscored how difficult it could be for research teams to maintain real-time awareness of important changes in a study participant's condition. In response, Carilion and Nordic added real-time safety notifications that alert research staff through Epic in-basket messages and email when predefined clinical thresholds are met.
Wahid said the notifications have also improved adverse-event reporting in clinical trials. Before implementation, she occasionally learned about serious adverse events after required reporting deadlines had passed.
"Since we've implemented this tool, ... we've had zero protocol deviations as it relates to delayed reporting around adverse events," she said.
As the dashboards evolved, specialists across the organization began requesting access to similar information for clinical use. Endocrinologists, gastroenterologists, hematologists, cardiologists and infectious disease physicians often wanted access to the same data elements already being tracked for research monitoring.
In response, Carilion and Nordic developed specialty-specific dashboards that surfaced information tailored to each group's workflow. For cardiologists, for example, the heart failure dashboard displays guideline-directed medical therapy recommendations based on a patient's ejection fraction.
"The original plan was for research," Wahid said. "But then it just blew up into something a lot more productive for others."
Workflow gains exposed the limits of EHR-centric design
The dashboards brought together information that previously required researchers to navigate multiple screens, but they also highlighted the amount of important clinical data that remains outside the EHR.
For instance, at Carilion, certain echocardiography data used in cardiology and structural heart research does not flow into Epic, and therefore, does not flow into the dashboards.
The dashboards project succeeded in making EHR data easier to use, but it also exposed the limitations of building solutions around a single platform. Wahid said that if she could go back and create the dashboards differently, she would pursue a more platform-agnostic approach that could draw from multiple data sources rather than relying primarily on the EHR.
The experience has highlighted a broader challenge facing healthcare organizations. While EHR-based tools can simplify access to information, their effectiveness still depends on access to data from other systems.
The project began with a simple question: why did identifying a single research candidate require 156 clicks? The answer led Carilion to redesign not only screening processes, but also how clinical information is accessed across research and care delivery.
Elizabeth Stricker, BSN, RN, comes from a nursing and healthcare leadership background, and covers health technology and leadership trends for B2B audiences.
