Investigating the effect of providing monetary incentives to participants on completion rates of referred co-respondents: An embedded randomized controlled trial

Abby Dunn; James Alvarez; Amy Arbon; Stephen Bremner; Chloe Elsby-Pearson; Richard Emsley; Christopher Jones; Peter Lawrence; Kathryn J Lester; Natalie Morson; Nicky Perry; Julia Simner; Abigail Thomson; Sam Cartwright-Hatton

doi:10.1016/j.conctc.2024.101267

Investigating the effect of providing monetary incentives to participants on completion rates of referred co-respondents: An embedded randomized controlled trial

Contemp Clin Trials Commun. 2024 Feb 8:38:101267. doi: 10.1016/j.conctc.2024.101267. eCollection 2024 Apr.

Affiliations

¹ University of Sussex, Falmer, United Kingdom.
² University Hospitals Sussex NHS Foundation Trust, Brighton, United Kingdom.
³ Brighton and Sussex Medical School, Brighton, United Kingdom.
⁴ King's College London, London, United Kingdom.
⁵ University of Southampton, Southampton, United Kingdom.

Abstract

Background: The use of a second informant (co-respondent) is a common method of identifying potential bias in outcome data (e.g., parent-report child outcomes). There is, however, limited evidence regarding methods of increasing response rates from co-respondents. The use of financial incentives is associated with higher levels of engagement and follow-up data collection in online surveys. This study investigated whether financial incentives paid to index participants in an online trial of a parenting-focused intervention, would lead to higher levels of co-respondent data collection.

Methods: A study within a trial (SWAT) using a parallel group RCT design. Participants in the host study (an RCT of an online intervention) were randomised into one of two SWAT arms: received/did not receive a £15 voucher when referred co-respondent completed baseline measures. Primary outcome was completion (No/Yes) of Spence Children's Anxiety Scale (SCAS or SCAS-Pre) at baseline. Additional analysis explored impact of incentives on data quality.

Results: Intention to treat analysis of 899 parents (183 co-respondents) in the no-incentive arm, and 911 parents (199 co-respondents) in incentive arm. Nomination of co-respondents was similar between incentive arms. The RR for the incentive arm compared to the no incentive arm was 1.13 (95% CI: 0.91 to 1.41, p = 0.264) indicating that incentives did not impact completion of outcomes by consented co-respondents. There were no indications of different data quality between arms.

Discussion: The finding that payment of financial incentives to index participant does not lead to greater levels of co-respondent outcome completion suggests that careful consideration should be made before allocating resources in this way in future trials.

Trial registration: The host study was registered at Study Record | ClinicalTrials.gov and the SWAT study was registered in the SWAT Store | The Northern Ireland Network for Trials Methodology Research (qub.ac.uk): SWAT number 143: Filetoupload,1099612,en.pdf (qub.ac.uk).

Keywords: Co-respondent; Incentives; RCT; SWAT; Trial design.