Randomization Protocol Development and Implementation for a Multi-Site Randomized Controlled Trial

• In Event: A New Model of Intentional Pretrial Supervision: Development, Implementation, and Multi-Site Randomized Controlled Trial

Abstract

Randomization procedures in multi-site research designs involve many complexities. Randomization protocols must account for variation in site-level practices, while maintaining consistent protocols and equal assignments. As part of a larger multi-site randomized controlled trial of Intentional Pretrial Supervision, we developed a site-administered randomization mechanism with external monitoring for randomization fidelity. We leveraged survey management platform Qualtrics (Qualtrics, 2005) to randomize individuals assigned to pretrial monitoring across ten counties in Indiana. We opted to use Qualtrics for randomization purposes due to the ability to create site-specific randomization forms while maintaining a shared directory to automatically screen for duplicate individuals. To protect randomization fidelity, we conducted weekly review of randomized participants to link randomized participants to local court cases within each county and triage randomization issues. By externally monitoring site-level randomization, we could closely track randomization and document exclusions for fidelity. To contribute to the knowledge base on randomization techniques in multi-site randomized controlled trials, we discuss randomization protocols, including commonly encountered issues and training development for randomization with local pretrial officers.

Authors

• Chelsea M. A. Foudray, George Mason University

• Evan Marie Lowder, George Mason University

• Peyton Frye, George Mason University

• Eric Grommon, Paul H. O’Neill School of Public and Environmental Affairs