Recruitment strategy that pairs use of electronic medical records with postal mailings shows cost-effectiveness and increased participation of under-represented groups
Boston – As electronic medical records (EMR) become ubiquitous in health care settings, scientists are increasingly turning to electronic-based recruitment methods to encourage participation in clinical trials. However, little is known about how this use of technology compares to more traditional clinical trial recruitment strategies, and some researchers worry that an overreliance on technology has the potential to exclude an eligible and interested diverse participant population.
In a new study published in the journal Clinical Trials, researchers led by Stephen Juraschek, MD, PhD, Assistant Professor of Medicine at Beth Israel Deaconess Medical Center (BIDMC) and Hailey Miller, RN, PhD, Postdoctoral Associate at Duke University School of Nursing, compared four electronic-based recruitment methods and four traditional recruitment methods to determine how different strategies may impact enrollment of groups traditionally under-represented in the medical literature.
“Black Americans are under-represented in clinical trials—something I’m reminded of every time I counsel Black patients using data from largely white adults,” said Juraschek. “Electronic medical records hold tremendous promise to offer patients with relevant conditions the opportunity to participate in clinical trials. However, the impact of electronic-based recruitment methods on participant demographics was unknown.”
Juraschek and colleagues used a variety of direct recruitment approaches, both electronic and traditional, to identify adults with gout for a clinical trial examining the links between diet and gout. The team used EMR records to identify prospective enrollees, and messaged them through patient portals—secure online medical messaging services providers use to communicate with patients. They also used non-targeted approaches such as community mailings and advertisements on social media and in newspapers.
After calculating the response rates, cost efficacy and demographic characteristics of eventual enrollees generated by the various strategies, Juraschek and colleagues found important demographic differences in how white and Black adults enrolled in the study. For example, two-thirds of Black participants had been identified through the electronic medical record but did not have an active patient portal. Instead, these enrollees learned about the study through a brochure sent by mail. Black participants otherwise would not have found out about the trial, which was pertinent to their health condition.
These findings build on Juraschek and Miller’s previous research that demonstrated that whites disproportionately use patient portals. While the use of patient portals for recruitment may boost overall enrollment in clinical trials, Juraschek said, overdependence on electronic recruitment could perpetuate the underrepresentation of Black patients in clinical research.
Ultimately, the researchers determined that a hybrid strategy using EMRs to identify patients, followed up with postal mailings to potential enrollees yielded the best results. The hybrid outreach method was both cost-effective and increased participation of underrepresented groups, including Black participants and women.
“All patients have the right to be offered medical therapies informed by studies that represent them,” said Juraschek, who is also an Instructor of Medicine at Harvard Medical School. “These findings also demonstrate the importance of a range of recruitment approaches to enroll a representative study population.”
Co-authors included Jeanne Charleston, Beiwin Wu, Kelly Gleason, Karen White, Cheryl Dennison Himmelfarb, Daniel E. Ford, Lawrence J. Appel, Allan C. Gelber and Edgar R. Miller III of Johns Hopkins University; and Timothy B. Plante of University of Vermont.
This work was supported by The Rheumatology Research Foundation and the Johns Hopkins Institute for Clinical and Translational Research, which is funded in part by the National Center for Advancing Translational Sciences (NCATS) (ULITR003098). Juraschek was supported by the National Heart, Lung and Blood Institute (7K23HL135273). Miller was supported by the National Institute of Nursing Research (T32NR012704).
The authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this manuscript.