Research conducted at Imperial College London has found that financial incentives designed to encourage patients to attend diabetic eye screening result in fewer attendees, compared to no incentive at all. The randomised study demonstrated that participants receiving any incentive were actually less likely to attend for appointment compared to participants receiving no incentive. The researchers consequently advise that, “financial incentives should not be used to promote screening unless tested in context, as they may negatively affect attendance rates.” Diabetic retinopathy (DR) is an enormous global health challenge and DR screening is a vital tool in the clinical management of the disease. In the UK, almost 1,300 new cases of DR are reported annually and healthcare strategies to increase screening uptake have an important role in reducing the burden of disease.
The provision of financial incentives is thought to be a useful tool in behavioural economic studies and, in the case of DR screening, may provide a cost-effective strategy to increasing the early detection and treatment of disease. The UK based research team set up a three-cohort randomized controlled trial in which 1,051 participants >16 years of age were randomized in the ratio of 1.4:1:1 to receive a standard invitation letter (control), an offer of £10 cash for attending a screening appointment, or the offer of winning £1,000 in a “prize draw”, with a 1 in 100 chance of winning. The participants had previously been invited for appointment over a prior 24-month period but had failed to attend and had not re-scheduled. The primary outcome for the study was the proportion of participants that attended screening. Results showed that the primary outcome was met by 7.8% of the control group (no incentive), 5.5% of the group offered a once-off payment of £10 cash, and 3.3% of the group offered the £1,000 lottery incentive. Analysis of the results showed that participants who received any incentive were significantly less likely to attend their appointment, compared with controls (risk ratio (RR)=0.56; 95% CI 0.34 to 0.92). Further, participants in the £1,000 lottery incentive group (RR=0.42; 95% CI 0.18 to 0.98) were significantly less likely to attend the screening appointment than those in the control group.
The results of the study clearly contradicted some of the received wisdom of behavioural economic theory whereby participants receiving an incentive may be more likely than not to participate in a given activity. However, the UK research supports previous research in the USA which also found a negative incentive effect associated with lower screening rates in a DR screening study. Notwithstanding such, the UK group conclude that the effect size in the London-based study was in fact quite small and not clinically significant, and therefore the results may not be generalizable to a broader population, especially given the existing high ethnic diversity of the study population used in the central London research. Accounting for the counter-intuitive observations, the researchers suggested that a potential lack of understanding on the importance of screening, coupled with a general resistance to medical intervention, may in part explain the lack of impact for the incentives.