A factorial randomised trial investigating factors influencing general practitioners’ willingness to prescribe aspirin for cancer preventive therapy in Lynch syndrome: a registered report

Abstract

Background: The National Institute for Health and Care Excellence (NICE) recommends aspirin for colorectal cancer prevention for people with Lynch syndrome. Strategies to change practice should be informed by understanding the factors influencing prescribing. Aim: To investigate the optimal type and level of information to communicate with GPs to increase willingness to prescribe aspirin. Design and setting: We recruited GPs in England and Wales (n=672) to an online survey with a 23 factorial design. GPs were randomised to one of eight vignettes describing a hypothetical patient with Lynch syndrome recommended to take aspirin by a clinical geneticist. Method: Across the vignettes, we manipulated the presence or absence of three types of information: 1) existence of NICE guidance; 2) results from the CAPP2 trial; 3) information comparing risks/benefits of aspirin. We estimated the main effects and all interactions on the primary (willingness to prescribe) and secondary outcomes (comfort discussing aspirin). Results: There were no statistically significant main effects or interactions of the three information components on willingness to prescribe aspirin or comfort discussing harms and benefits. In total, 80.4% (540/672) of GPs were willing to prescribe, with 19.7% (132/672) unwilling. GPs with prior awareness of aspirin for preventive therapy were more comfortable discussing the medication than those unaware (p=0.031). Conclusion: It is unlikely that providing information on clinical guidance, trial results and information comparing benefits and harms will increase aspirin prescribing for Lynch Syndrome in primary care. Alternative multilevel strategies to support informed prescribing may be warranted.