Skip to main content

Research and Projects

PenCHORD - Diabetic retinopathy screening interval assesment

Who is involved?

Theme: PenCHORD
Status: Complete

Read a BITE sized summary of this project.


We developed a simulation model to predict the impact on patient vision of screening patients with Type 2 diabetes for diabetic retinopathy every two years, rather than annually, for those who had not yet developed retinopathy.  We used agent-based modelling to simulate the retinopathy progression of individual patients, their attendance at screening appointments and their treatment. The model was populated with data obtained from the Royal Devon and Exeter NHS Foundation Trust.  Our model predicted that by screening patients who have not yet developed retinopathy every two years until they are identified with retinopathy, there would be no impact on the proportion of patients who would lose their vision but there could be a 25% saving on screening costs.


The prevalence of diabetes continues to grow globally.  In the UK, screening services are facing increasing strain because of significantly increasing demand coupled with an unchanging investment in resources.  Current NICE Guidelines specify that diabetic patients should be screened for retinopathy at least annually.  However, for many, diabetic retinopathy is a slow-progressing disease and there are concerns that the current screening policy is excessive, particularly in light of the growing strain on resources.
The project sought to determine, via simulation modelling, whether patients with type 2 diabetes who hadn’t yet developed retinopathy could be screened every two years, rather than annually.  PenCHORD collaborated with a Diabetic Retinal Screening Manager, a Consultant Endocrinologist and an Operational Performance Manager at the Royal Devon and Exeter Hospital.


We developed an agent-based model that simulated the retinopathy progression of type 2 diabetes patients and their concurrent screening and treatment attendances.  The model was built using AnyLogic framework software.  The model was parameterised using values obtained from analysis of anonymised data of 3,537 patients and 33,810 screening appointments from 1991 to 2011, supplied by the Royal Devon and Exeter NHS Foundation Trust.  Fifteen year forecasts were generated from the model.

This flowchart summarises the process steps that patients in the model follow.

Project outputs

The simulation model predicts that implementing a two year screening interval for type 2 diabetic patients without evidence of diabetic retinopathy does not increase their risk of vision loss. Furthermore, the model estimates that screening costs could be reduced by 25%.

Screening people with type 2 diabetes, who have not yet developed retinopathy, every two years rather than annually, is a safe and cost-effective strategy. Our findings support those of other studies and recommend a review of the current National Institute for Health and Clinical Excellence (NICE) guidelines for diabetic retinopathy screening in the U.K.

Evaluation and impact

The results of the study have been published in the high-impact journal Diabetes Care and both the Royal Devon and Exeter NHS Foundation Trust and Cornwall and Isles of Scilly NHS have expressed an interest in piloting a two-year screening programme on the back of this evidence.  The study is also being considered as part of a national review of evidence for changing the diabetic retinopathy screening interval.

Related publications

View more PenCHORD projects


  • Diabetes Care Full Text Journal Article - PDF