#4011 LESSONS LEARNT FROM VIRTUAL RECRUITMENT TO A MULTICENTRE RANDOMISED CONTROLLED TRIAL EVALUATING A DIGITAL HEALTH INTERVENTION FOR CKD

Abstract

Background and Aims: The post-COVID era has seen an escalation in the use of digital health interventions (DHIs), which have the potential to provide a widely accessible and cost-efficient approach to aspects of health service delivery such as patient education. Robust effectiveness evaluation is essential for all novel interventions to enable appropriate and sustained implementation. We have developed an education/self-management DHI named 'My Kidneys & Me' (MK&M) for chronic kidney disease (CKD). A multi-centre randomised controlled trial (RCT) (SMILE-K) with virtual study processes allowing for minimal face-to-face participant contact was conducted to pragmatically deliver and evaluate MK&M. Here we report the recruitment metrics of SMILE-K to inform and improve future design of virtual trials and delivery of DHIs. Method(s): The SMILE-K virtual study design involved outcome measures collected via an online survey, and 2:1 randomisation to intervention or control group The recruitment process is summarised in Figure 1. Briefly, eligible adult CKD patients were identified by their clinical teams and given or sent a study invitation. If interested, patients emailed the study team, who then sent sequential emailed links to: 1) the information sheet and consent form; 2) the baseline outcome measure survey; and 3) (if randomised to intervention), MK&M. Result(s): 26 hospital sites in England recruited to SMILE-K between May 2021 and December 2022. 6693 invitations were issued: 4942 (74%) by mail, 883 (13%) in-person, 616 (9%) by mail following remote consultation; and 252 (4%) by email. 13% (875/ 6693 invited) invitees expressed interest, of which 60% (533/875) consented. Of the 533 consented, 5% (265/4942) had been invited via mail, 14% (120/883) in-person, 17% (103/616) via mail following remote appointment, and 2% (5/252) via email. Of those consented, 80% (424/533) completed the baseline survey and were subsequently randomised. The mean age of recruited participants was 74 years (+/-9.0; range:20-88) and 94% were White British. The median time from expression of interest to consent was 1 day (IQR:0-5), and consent to randomisation was 6 days (IQR:2.5-11). Conclusion(s): Our experience shows that a virtual strategy for DHI RCT recruitment provides the opportunity to reach a high number of eligible participants. However, only 13% of invited patients expressed interest, and just 7% were recruited into the study. Most invitations were sent via mail with no prior approach or explanation. This method resulted in a markedly lower response rate than invitations issued after discussion with a healthcare professional, either remotely or in-person, underlining the importance of personal contact. Substantial attrition occurred at the consent stage, indicating that reading the information sheet and completing the online consent form unassisted was a barrier. Despite the need for internet access and digital skills, there was a wide participant age range, including those in their 80s. However, we noted a disproportionate representation of White ethnicity. This highlights the need to embed specific strategies to engage minority ethnic communities and other disadvantaged groups in the design of DHIs and future virtual trials to maximise reach and access to their potential benefits.