Factors Associated With Enrollment to a Decentralized Study

Objective: To assess whether the mode and formatting of invitations affect enrollment in a large, decentralized study.

Patients and methods: Between July 1, 2022, and October 30, 2022, we prospectively compared various approaches to enroll patients in the Tapestry DNA Sequencing Research Study, a decentralized exome-sequencing study. In phase 1, patients were randomized to receive invitations via the electronic health record (EHR) patient portal or email (cohort 1, 69,852 patients). Phase 2 randomized in a 2×2 factorial design to receive (by portal or email) standard or enhanced (ie, more visually appealing) invitations (cohort 2, 26,198 patients). Factors that predicted enrollment rates were analyzed.

Results: The enrollment rate was greater in cohort 2 (1,785 of 24,550, 7.27%) than 1 (1,758 or 69,765, 2.52%) and remained significant after multivariable adjustment (odds ratio, 1.31; 95% CI, 1.19-1.45). Enrollment rates were greater in women than men, patients 50 to 70 years of age than younger patients, White or non-Hispanic or Latino patients than those in other racial categories, urban than rural residents, and patients who had more health care encounters or more recent health care before this study (P<.02). The enrollment rate was also greater when invitations were delivered via EHR than email (odds ratio, 1.56; 95% CI, 1.44-1.68; P<.001).

Conclusion: Invitations via EHR rather than email facilitate enrollment to large, decentralized studies. Enhanced display of invitation material did not increase enrollment. Lower enrollment rates in men, younger individuals, non-White and Hispanic individuals, and rural residents highlight a continued need to focus enrollment strategies on these subgroups.