Extension to ten years

The DIfE ─ German Diabetes Risk Score (DRS), developed by the German Institute of Human Nutrition Potsdam-Rehbruecke (DIfE), has thus far enabled the prediction of the individual risk of developing type 2 diabetes within the next five years. Scientists from DIfE and the German Center for Diabetes Research (DZD) have now extended the prediction period to ten years, validated it in two independent study populations and correspondingly updated the DRS.

According to the German Diabetes Association, there are over eight million people with type 2 diabetes in Germany. The number of unreported cases is estimated to be at least two million. In order to be able to identify people with a high risk of developing the disease at an early stage, DIfE scientists developed the DIfE ─ German Diabetes Risk Score (DRS) in 2007, which can be used outside of clinical practice.

The DRS offers adults between the ages of 18 and 79 the opportunity to fill out a questionnaire or take an online test at home to determine their individual risk of developing type 2 diabetes within the next five years, non-invasively and free of charge. The longer follow-up period of the EPIC Potsdam study on which it is based, which is now available, has allowed the DRS to be revised to predict the 10-year risk. "The extension of the prediction period enables even longer-term individual prediction of risk for type 2 diabetes," said Dr. Catarina Schiborn, first author and research associate in the Department of Molecular Epidemiology at DIfE. "This means that people at increased risk can be educated at an early stage and informed about options for risk reduction and lifestyle changes."

Lifestyle has a direct impact on the disease risk

For the further development of the prediction model, the scientists used the health and lifestyle data of around 25,000 participants in the EPIC-Potsdam study. Parameters that are relevant for the prediction of type 2 diabetes were included in the statistical modelling. These include unchangeable factors such as age, gender and family history of type 2 diabetes, as well as modifiable factors such as waist circumference, smoking habits, high blood pressure and physical activity. The model also takes dietary habits into account, such as the consumption of whole grain products and red meat.

Among the participants in the study, type 2 diabetes was diagnosed in 1,367 people within ten years of the initial examination. For the majority of those affected, the 10-year DRS was correct in its prediction. "In EPIC-Potsdam, the score predicted a higher risk for future patients with a probability of 83.4 percent than for those who did not develop diabetes in the future," Schiborn explained. The very good prediction quality could be confirmed in two independent samples – the population-based observational study EPIC-Heidelberg and in a sample representative for Germany, the BGS98 cohort. "This underlines the significance of the score for the German population," said Schiborn.

Wide range of applications due to different test variants

The extension of the risk prediction to ten years enables the updating of the previous DRS versions: the online test, the questionnaire self-test and the patient questionnaire for medical professionals, in which the diabetes risk can be specified together with the long-term blood glucose value (HbA1c value).

"The 10-year DRS offers a wide range of applications and can be used in everyday clinical practice for screening, as well as at home when clinical parameters are not available," said Professor Matthias Schulze, head of the Department of Molecular Epidemiology at DIfE. In addition, the updated DRS, like its predecessor, can be used in educational campaigns and population-wide screening procedures.

Further information and contact

Press release - German Institute of Human Nutrition Potsdam-Rehbruecke (DIfE)