Do Non-participants at Screening have a Different Threshold for an Acceptable Benefit-Harm Ratio than Participants? Results of a Discrete Choice Experiment

Abstract

Objective: The objective of the study was to investigate non-participants' preferences for cardiovascular disease screening programme characteristics and whether non-participation can be rationally explained by differences in preferences, decision-making styles and informational needs between non-participants and participants.

Methods: We conducted a discrete choice experiment at three screening sites between June and December 2017 among 371 male non-participants and 830 male participants who were asked to trade different levels of five key programme characteristics (chance of health benefit, risk of overtreatment, risk of later regret, screening duration and screening location). Data were analysed using a multinomial mixed-logit model. Health benefit was used as a payment vehicle for estimation of marginal substitution rates.

Results: Non-participants were willing to accept that 0.127 (95% confidence interval 0.103-0.154) fewer lives would be saved to avoid overtreatment of one individual, whilst participants were willing to accept 0.085 (95% confidence interval 0.077-0.094) fewer lives saved. This translates into non-participants valuing health benefits 7.9 times higher than overtreatment. The corresponding value of participants is 11.8. Similarly, non-participants had higher requirements than participants for advanced technology and a quicker screening duration. With regard to their participation decision, 64% of the non-participants felt certain about their choice compared with 89% among participants.

Conclusions: This study shows that non-participants have different preferences than participants at screening as they express relatively more concern about overtreatment and have higher requirements for a high-tech screening programme. Non-participants also report to be more uncertain about their participation decision and more often seek additional information to the standard information provided in the invitation letter. Further studies on informational needs and effective communication strategies are warranted to ensure that non-participation is a fully informed choice.