JMIR Publications

ABSTRACT

Background:

Background:

Peers are known to influence the health behaviours and attitudes of adolescents, yet recruitment of these networks is challenging. Previous studies have used web-based respondent-driven sampling (WebRDS) methods to recruit this population, yet none have experimentally investigated the impact of financial reimbursements.

Objective:

Objectives: This study aimed to (1) compare the effectiveness of two financial reimbursement strategies for recruiting adolescents and their peer networks, and (2) explore factors associated with successfully recruiting peers.

Methods:

Methods:

A parallel design randomised controlled trial was conducted in which participants (seeds) were randomly allocated to a fixed cash reimbursement (control) or scaled reimbursement (experimental) group as a strategy to be recruited into a web-based peer network study. Seeds aged 16-18 were recruited online through social media advertisements and through an online student panel. They completed a web-based survey which assessed eligibility and included questions about their friends (peers). Allocation occurred through the survey platform using a simple randomisation method. In the fixed group, all participants in a peer network received AUD $5 (USD $3.29); in the scaled group, all participants in a peer network received an additional AUD $5 (USD $3.29) per peer who successfully completed the survey (to a maximum of AUD $30 each; USD $19.72). Participants and researchers were not blinded to intervention groups. The intervention was automated and did not require direct researcher contact. The primary outcome was recruitment of peers to complete the web-based survey (proportion of nominated peers). The number of peers recruited was a secondary outcome.

Results:

Results:

Of 463 seeds allocated to an intervention (scaled n=221; fixed n=242), 319 (68.9%) had complete data for analysis (scaled n=157 (71.0%); fixed n=162) (67.0%). A total of 11.9% of seeds successfully referred peers (18.5% scaled group; 5.6% fixed group). Those in the scaled reimbursement intervention were 3.80 times more likely to successfully recruit their peers than those in the fixed reimbursement intervention (proportion ratio (PR): 3.80, 95% CI 1.78–8.09). Similarly, the average number of peers recruited differed by 0.19 per seed between the scaled and fixed intervention groups (95% CI = 0.11–0.28).

Conclusions:

Conclusions:

Despite the relative effectiveness of scaled reimbursements, recruiting peers through WebRDS remained challenging. Clinical Trial: International Registered Report Identifier (IRRID): DERR1-10.2196/44813