AUTHOR=Soe Phyumar , Sadarangani Manish , Naus Monika , Muller Matthew P. , Vanderkooi Otto G. , Kellner James D. , Top Karina A. , Wong Hubert , Isenor Jennifer E. , Marty Kimberly , De Serres Gaston , Valiquette Louis , McGeer Allison , Bettinger Julie A. , for the Canadian Immunization Research Network TITLE=Impact of recruitment strategies on individual participation practices in the Canadian National Vaccine Safety Network: prospective cohort study JOURNAL=Frontiers in Public Health VOLUME=12 YEAR=2024 URL=https://www.frontiersin.org/journals/public-health/articles/10.3389/fpubh.2024.1385426 DOI=10.3389/fpubh.2024.1385426 ISSN=2296-2565 ABSTRACT=Background

The Canadian National Vaccine Safety (CANVAS) network conducted a multi-center, prospective vaccine safety study to collect safety data after dose 1 and 2 of COVID-19 vaccines and follow up safety information 7 months after dose 1.

Objective

This study aimed to describe and evaluate the recruitment methods used by CANVAS and the retention of participants by each modality.

Methods

CANVAS deployed a multi-pronged recruitment approach to reach a larger sample, without in-person recruitment. Three primary recruitment strategies were used: passive recruitment, technology-assisted electronic invitation through the vaccine booking system (auto-invitation), or auto-registration through the vaccine registries (auto-enrollment).

Results

Between December 2020 and April 2022, approximately 1.3 million vaccinated adults either self-enrolled or were auto-enrolled in CANVAS, representing about 5% of the vaccinated adult Canadian population. Approximately 1 million participants were auto-enrolled, 300,000 were recruited by auto-invitation, and 5,000 via passive recruitment. Overall survey completion rates for dose 1, dose 2 and the 7-month follow-up surveys were 51.7% (681,198 of 1,318,838), 54.3% (369,552 of 681,198), and 66.4% (452,076 of 681,198), respectively. Completion rates were lower among auto-enrolled participants compared to passively recruited or auto-invited participants who self-enrolled. However, auto-enrolled samples were much larger, which offset the lower completion rates.

Conclusion

Our data suggest that auto-enrollment provided an opportunity to reach and retain a larger number of individuals in the study compared to other recruitment modalities.