Diabetic Retinopathy Progression Prediction Model Validation

  • Research type

    Research Study

  • Full title

    Progression of Diabetic Retinopathy from referral to treatment or vision loss: External Validation, update and net clinical benefit of a Multivariable Prediction Model

  • IRAS ID

    253774

  • Contact name

    Nicola Adderley

  • Contact email

    n.j.adderley@bham.ac.uk

  • Sponsor organisation

    University of Birmingham

  • Clinicaltrials.gov Identifier

    NIHR203608, NIHR RfPB

  • Duration of Study in the UK

    1 years, 6 months, 0 days

  • Research summary

    Patients with diabetes commonly develop complications at the back of the eye, called diabetic retinopathy. This condition can cause blindness if left untreated and therefore diabetic patients are screened every one to two years in the diabetic eye screening programme (DESP). Once more serious disease develops, hospital / surveillance clinics appointments become necessary for monitoring. With diabetes becoming more common, there is increased demand for these appointments, resulting in delays. This means there is a chance of higher risk patients losing their sight while waiting to be seen. It would therefore be safer to have a system by which these higher risk patients could be prioritised according to their risk.

    Using information (data) collected from general practices, we have developed a model which identifies patients at high risk of progressing to diabetic retinopathy requiring treatment or to loss of vision. This model now needs testing in a different setting to ensure it works well in patients using hospital services / surveillance clinics.

    The hospital / surveillance clinics routinely collect data on patients with diabetes > 12 years of age, once referred by the diabetic retinopathy screening services and are now under hospital eye services / surveillance clinics. The data will be extracted by nurse specialists in three NHS trusts (Sunderland, Sussex and Sandwell). This data will then be completely anonymised, encrypted and provided to the University's research team. The model will be updated by including variables not available in the previous data.

    The findings will be presented to the group including patients and experts, to decide on how the model can best be implemented to benefit patients in a consensus meeting to arrive at the final model. This group gave input into the study design in an earlier meeting.

  • REC name

    South Central - Hampshire A Research Ethics Committee

  • REC reference

    22/SC/0425

  • Date of REC Opinion

    24 Nov 2022

  • REC opinion

    Further Information Favourable Opinion