A cross-over, randomised feasibility study of digitally-printed versus hand-painted artificial eyes in adults (PERSONAL-EYE-S):health economic findings

Abstract Background Technology advances mean alternatives to hand-painted artificial eyes are possible, but the feasibility of conducting a large-scale trial is unknown. The aim was to assess the feasibility of collecting healthcare resource use and associated costs needed to undertake a large-scale randomised controlled trial (RCT) comparing the effectiveness and cost-effectiveness of hand-painted artificial eyes with digitally-printed artificial eyes. Methods Participants wore a digitally-printed artificial eye and a hand-painted artificial eye, for two weeks each, in a random order. Individual patient-level data was used to explore health outcomes (EQ-5D-5L) and resource use. Costs of the two artificial eye services were collected. A full economic evaluation was not conducted. An appropriate economic evaluation framework was developed to identify the relevant health economic data necessary for a future full trial. Results Thirty-five participants were randomised. Response rates were 97–100% for the EQ-5D-5L. Resource use questions were less well completed: 54% complete responses at baseline, 40% partial responses and 6% missing/invalid responses. The two follow-up points had similar rates. Eye services cost data were well completed. Mean utility was 0.77 after wearing the hand-painted eye and 0.83 after the digitally-printed eye. Average manufacturing times were 294 min (digitally-printed) and 355 min (hand-painted). Remake appointments were needed for digitally-printed eyes only. Estimated cost for the digitally-printed eye service is £404 and £347 for the hand-painted eye. Time between fitting and final evaluation was 56 days (digitally-printed eye) and 60 days (hand-painted). Results should be interpreted with caution, as estimates were based on a small sample. Conclusion It was possible to collect EQ-5D-5L, healthcare resource use and manufacturing times allowing a costing to be calculated. Thus, a large-scale RCT with full cost-effectiveness analysis would feasible. Refinements are suggested. Future economic analysis should consider how best to evaluate the service, possibility with modelling rather than within-trial analysis only. Trial registration ISRCTN85921622.