Incentives in Diabetic Eye Assessment by Screening (IDEAS) trial: a three-armed randomised controlled trial of financial incentives

Background: The UK national diabetic eye screening (DES) programme invites diabetic patients aged > 12 years annually. Simple and cost-effective methods are needed to increase screening uptake. This trial tests the impact on uptake of two financial incentive schemes, based on behavioural economic principles. Objectives: To test whether or not financial incentives encourage screening attendance. Secondarily to understand if the type of financial incentive scheme used affects screening uptake or attracts patients with a different sociodemographic status to regular attenders. If financial incentives were found to improve attendance, then a final objective was to test cost-effectiveness. Design: Three-armed randomised controlled trial. Setting: DES clinic within St Mary’s Hospital, London, covering patients from the areas of Kensington, Chelsea and Westminster. Participants: Patients aged ≥ 16 years, who had not attended their DES appointment for ≥ 2 years. Interventions: (1) Fixed incentive – invitation letter and £10 for attending screening; (2) probabilistic (lottery) incentive – invitation letter and 1% chance of winning £1000 for attending screening; and (3) control – invitation letter only. Main outcome measures: The primary outcome was screening attendance. Rates for control versus fixed and lottery incentive groups were compared using relative risk (RR) and risk difference with corresponding 95% confidence intervals (CIs). Results: A total of 1274 patients were eligible and randomised; 223 patients became ineligible before invite and 1051 participants were invited (control, n = 435; fixed group, n = 312; lottery group, n = 304). Thirty-four (7.8%, 95% CI 5.29% to 10.34%) control, 17 (5.5%, 95% CI 2.93% to 7.97%) fixed group and 10 (3.3%, 95% CI 1.28% to 5.29%) lottery group participants attended. Participants offered incentives were 44% less likely to attend screening than controls (RR 0.56, 95% CI 0.34 to 0.92). Examining incentive groups separately, the lottery group were 58% less likely to attend screening than controls (RR 0.42, 95% CI 0.18 to 0.98). No significant differences were found between fixed incentive and control groups (RR 0.70, 95% CI 0.35 to 1.39) or between fixed and lottery incentive groups (RR 1.66, 95% CI 0.65 to 4.21). Subgroup analyses showed no significant associations between attendance and sociodemographic factors, including gender (female vs. male, RR 1.25, 95% CI 0.77 to 2.03), age (≤ 65 years vs. > 65 years, RR 1.26, 95% CI 0.77 to 2.08), deprivation [0–20 Index of Multiple Deprivation (IMD) decile vs. 30–100 IMD decile, RR 1.12, 95% CI 0.69 to 1.83], years registered [mean difference (MD) –0.13, 95% CI –0.69 to 0.43], and distance from screening location (MD –0.18, 95% CI –0.65 to 0.29). Limitations: Despite verification, some address details may have been outdated, and high ethnic diversity may have resulted in language barriers for participants. Conclusions: Those receiving incentives were not more likely to attend a DES than those receiving a usual invitation letter in patients who are regular non-attenders. Both fixed and lottery incentives appeared to reduce attendance. Overall, there is no evidence to support the use of financial incentives to promote diabetic retinopathy screening. Testing interventions in context, even if they appear to be supported by theory, is important. Future work: Future research, specifically in this area, should focus on identifying barriers to screening and other non-financial methods to overcome them. Trial registration: Current Controlled Trials ISRCTN14896403. Funding: This project was funded by the National Institute for Health Research (NIHR) Health Services and Delivery Research programme and will be published in full in Health Services and Delivery Research; Vol. 5, No. 15. See the NIHR Journals Library website for further project information