In a previous blog we discussed how “One Size Doesn´t Always Fit All” when it comes to screening for diabetic retinopathy, one of the leading causes of blindness in the global working age population. This is what our founders had in mind when they designed the RetinaRisk algorithm that calculates the individualized risk of people with diabetes of developing sight-threatening eye disease. They proposed a variable interval screening, based on each patient’s risk profile, instead of a fixed screening interval for all. This can ensure that the right person receives the right treatment at the right time – truly personalized medicine.
The RetinaRisk algorithm was developed following extensive international research on risk factors known to affect the progression of diabetic retinopathy, such as the type and duration of diabetes, gender, blood pressure and HbA1c levels, as well as a diagnosis of retinopathy. Clinical validation in over 20.000 diabetic patients in five countries has confirmed that the RetinaRisk algorithm can predict the risk of retinopathy progression with a very high precision (ROC curve with AUC over 0.8).
We are very pleased to now present a five-year real-life use of RetinaRisk. A recent article in Acta Ophthalmologica describes the use of the RetinaRisk algorithm in a diabetic eye-screening programme in Norway between 2014 and 2019, where screening intervals were individualized, and clinical outcomes, safety and cost-eﬀectiveness documented. The diabetes cohort was divided on a voluntary basis into two groups: one with variable screening intervals based on their personal risk proﬁle and the other group with conventional ﬁxed interval diabetic eye screening.
The key findings of this ´real world´ study was that the RetinaRisk algorithm was safe and eﬀective in an ophthalmology clinic´s diabetic screening program. It showed that the use of RetinaRisk reduces the mean frequency of screening visits and liberates valuable time in the ophthalmic practice. This time savings can be used on high-risk diabetic patients or other patient groups, since they otherwise would have been screened at ﬁxed annual visits. RetinaRisk helps to summarize the clinical data for each patient and makes it easier to ﬁnd the high-risk diabetic patients. These patients can then be screened at shorter intervals.
The Norwegian experience demonstrated that patients appreciate the focus on the relevant risk factors and the way that RetinaRisk can visualize the risk proﬁle for each patient. The RetinaRisk app helps to increase the interest and motivate patients to improve their risk proﬁle.
Overall, RetinaRisk was found to be a valuable tool for improving the quality and economics of diabetic retinopathy screening. We thank Dr. Estil for sharing his invaluable experience with using the RetinaRisk algorithm and look forward to our continued fruitful collaboration in eliminating preventable diabetic blindness.