Data mart speeds recruitment for COVID research
by Paul Govern
The arrival last March of the COVID-19 pandemic in Tennessee presented challenges for the clinical research enterprise at Vanderbilt University Medical Center. Prompt recruitment of patients newly diagnosed with COVID-19 into clinical trials was, and remains, a public health imperative with top-level backing from the federal government.
“With different VUMC research teams potentially looking to urgently recruit the very same patients, as an institution we were at risk of taxing the good will and patience of acutely ill patients with repeated overtures to join various COVID clinical trials,” said Tara Helmer, research services consultant with Research Support Services and project manager for VUMC’s centralized COVID-19 human subjects recruitment strategy.
Two opposed requirements, speed and coordination, shaped the recruitment innovations introduced on May 1, 2020, for COVID trials at VUMC.
- Within minutes of undergoing testing for COVID at a VUMC facility, patients receive an automated email or text message from Gordon Bernard, MD, Executive Vice President for Research — in English, Spanish, or Arabic — that links to a web form where respondents can indicate whether or not they would like to be directly contacted by COVID-19 research teams at VUMC. In human subject research, this step is known as consent to contact, and it currently applies to cold calls at VUMC, that is, to direct overtures from research teams not involved in the patient’s care. Patient responses to these messages populate a REDCap (Research Electronic Data Capture) data-base.
- The moment a positive COVID-19 test result comes back, data elements from the patient’s electronic health record (EHR) are automatically sent to a second REDCap database known as the COVID-19 data mart: demographics, laboratory results, vitals, medications, allergies, problem list, inpatient/outpatient status. A program checks these data mart records against the consent to contact database, and against the clinical scheduling system (because having an upcoming clinic appointment is an inclusion criterion for one of the participating studies).
- The REDCap team has worked with researchers to partially translate study inclusion criteria into lines of computer code, allowing instant awareness of which patients in the data mart match criteria for which COVID studies, to a first approximation. The data mart’s focus, for now, is on federally funded, multicenter outpatient interventional trials.
- As the program identifies prospective matches, an electronic dashboard allows the REDCap team to release patient records, with a click, to the appropriate study coordinators. To participate in the data mart, COVID researchers must agree to cede right of refusal to each other on a rotating basis, according to the time of day or day of the week: when patients appear to meet criteria for more than one study (usually the case), records are released to whichever study currently has right of refusal.
- As records are released to researchers, study coordinators examine the EHR in more detail for complete inclusion criteria, then make recruitment calls as appropriate. When patients decline to join a given study, provided they don’t also revoke their consent to contact, they may be approached again as their records are reviewed anew by research coordinators from the next study in line. Once a patient joins an interventional COVID trial, overtures from the other interventional studies using the data mart are halted.
To date, five studies have participated in the COVID data mart.
“I think the strategy has saved a ton of time and resources,” said Bev Woodward, MSN, RN, a research projects manager with the Division of Infectious Diseases.
“This saves a lot of time for us,” said Cassandra Reynolds, clinical trials manager with the Division of Cardiovascular Medicine. “Going through all the records from the hospital and trying to find these patients is a huge time commitment. This streamlines that and brings the patients to us that meet our high-level eligibility criteria.”
The strategy is the work of VUMC human subject protection experts, Privacy Office leadership, Patient Education, and, not least, the REDCap team and Research Support Services staff at the Vanderbilt Institute for Clinical and Translational Research (VICTR).
“We know that well-orchestrated, large-scale, rigorous multicenter trials are the only way we will efficiently learn what therapies are safe and efficacious for patients,” said Bernard, who is one of two program directors at VICTR. “VUMC’s prioritization of these trials through informatics tools such as the data mart has helped us contribute to the growing body of knowledge about what works, and importantly, what doesn’t.”
REDCap is a secure web application geared to support data capture for clinical and translational research. It was conceived and developed at VUMC by Paul Harris, PhD, professor of Biomedical Informatics and director of the Office of Research Informatics, and launched in 2004.
Under Harris and Rob Taylor, manager of application development, REDCap has in recent years been modified to allow data to flow directly into research projects from an institution’s EHR servers (based, as always, on privacy protections and well-defined permissions granted to each researcher-user).
This clinical data interoperability provides the linchpin for the new human subject recruitment strategy.
Starting in 2006, Vanderbilt made REDCap freely available for non-commercial use, and today it’s used in 141 countries at more than 4,800 organizations comprising the REDCap Consortium.
“I should stress that these technical solutions can be easily adapted at any clinical research institution that uses REDCap,” Harris said. “And they can be applied to other diseases. At Vanderbilt and elsewhere, we’re going to see many more projects using these tools.”