Strategies to improve retention in randomised trials

Area of Science:

• Clinical Trials Methodology
• Evidence Synthesis
• Health Services Research

Background:

• Participant retention in randomized trials is vital to prevent bias and maintain study power.
• Poor retention compromises the generalizability, validity, and reliability of research findings.
• Many retention strategies exist, but few have undergone rigorous formal evaluation.

Purpose of the Study:

• To systematically quantify the impact of various strategies designed to enhance participant retention in randomized trials.
• To explore whether the effectiveness of retention strategies differs across various trial settings.

Main Methods:

• A comprehensive search of multiple electronic databases (CENTRAL, MEDLINE, Scopus, etc.) and repositories was conducted up to January 2020.
• Included were eligible randomized or quasi-randomized trials evaluating retention strategies embedded within 'host' randomized trials across all healthcare settings.
• Data extraction focused on retention strategies, study characteristics, and participant numbers, with meta-analyses using risk difference (RD) and 95% confidence intervals (CI). Certainty of evidence was assessed using GRADE.

Main Results:

• The review included 69 studies with over 100,000 participants, primarily evaluating interventions for postal questionnaire response.
• No retention strategy comparisons were supported by high-certainty evidence according to GRADE assessments.
• Moderate-certainty evidence supported three strategies (self-sampling kits, monetary rewards with reminders, recruitment pens) and identified one strategy that reduced retention (inclusion of a diary).

Conclusions:

• The majority of evaluated interventions focused on improving postal questionnaire response rates, with limited focus on participant return to trial sites.
• No retention strategies demonstrated high-certainty evidence of effectiveness.
• Future research should prioritize well-designed studies to improve the certainty of evidence for specific retention strategies, particularly those currently supported by single studies.