'Is insulin right for me?': Feasibility of a pilot randomised controlled trial and acceptability of a web-based intervention to reduce psychological barriers to insulin therapy among adults with type 2 diabetes

Aims Acceptable and accessible interventions are needed to address 'psychological insulin resistance', which is a common barrier to insulin uptake among adults with type 2 diabetes (T2D). Our aim was to test the feasibility of a randomised controlled trial (RCT) study design and acceptability of a theoretically grounded, psycho-educational, web-based resource to reduce negative insulin appraisals among adults with T2D. Methods A double-blinded, parallel group, two-arm pilot RCT (1:1), comparing intervention with active control (existing online information about insulin). Eligible participants were Australian adults with T2D, taking oral diabetes medications. Exclusion criteria: prior use of injectable medicines; being 'very willing' to commence insulin. Primary outcomes: study feasibility (recruitment ease, protocol fulfilment, attrition, data completeness); secondary outcomes: intervention acceptability (intervention engagement, user feedback) and likely efficacy (negative Insulin Treatment Appraisal Scale [ITAS] scores at follow-up). Online surveys completed at baseline and 2 weeks. Results During 4-week recruitment, 76 people expressed interest: 51 eligible and 35 enrolled (intervention = 17, control = 18; median[interquartile range] age = 62[53, 69] years; 17 women). Protocol fulfilment achieved by 26 (74%) participants (n = 13 per arm), with low participant attrition (n = 6, 17%). Intervention acceptability was high (>80% endorsement, except format preference = 60%). ITAS negative scores differed between-groups at follow-up (M diff = -6.5, 95% confidence interval: -10.7 to -2.4), favouring the intervention. Conclusions This novel web-based resource ("Is insulin right for me?") is acceptable and associated with a likely reduction in negative insulin appraisals, relative to existing resources. This pilot shows the study design is feasible and supports conduct of a fully powered RCT.