Experience of linking to the NHS diabetic eye screening programme records in the ASCEND-eye randomized trial and recommendations for improvement.

Background: The ASCEND-Eye sub-study of the large, double-blind, 2x2 factorial design, placebo-controlled ASCEND trial compared the effects of aspirin and, separately, omega-3 fatty acids on diabetic retinopathy outcomes derived from NHS Diabetic Eye Screening Programmes (DESP) in England and Wales, in adults aged 40 years or older with diabetes and no pre-existing atherosclerotic cardiovascular disease. ASCEND-Eye was unprecedented in what it set out to achieve; no previous studies had successfully obtained linked DESP data for research purposes on a national scale in England and Wales before. Objective: To describe our experience of linking DESP records to help other researchers wishing to use them. We explain the application process, lead times and resources required, and how these data were governed. Results: The process of gaining regulatory and ethics committee approval for ASCEND-Eye through to data acquisition took four years. Several challenges were encountered, including a lack of documentation defining the governance of the NHS screening service, the absence of a single central data repository, the inherent complexity of liaising with multiple data controllers, and a lack of responsiveness to invitations to collaborate by nearly half of the DESPs in England. Conclusion: Routinely collected healthcare data is a valuable source of outcome measure information in clinical trials. However, researchers frequently face barriers to accessing these datasets despite having written informed consent from trial participants to do so. We hope to encourage more NHS DESPs to take part in research.