Objective: The JJ risk engine, developed in 2012, is an algorithm that predicts the 19 incidence of diabetes complications that may develop after 5 to 10 years. However, 20 studies validating the JJ risk engine have not yet been reported;we aimed to verify the 21 JJ risk engine. In 2013, we conducted a retrospective survey using medical records of 22 484 patients with type 2 diabetes. The observed value of coronary heart disease (CHD) 23 complicationsafter 5 years and the predicted value by the JJ risk engine as of 2013 were 24 compared and verified using the discrimination and calibration values.25
Results: Among the total cases analyzed, the C-statistic was 0.588,and the calibration 26 was p <0.05; thus, the JJ risk engine could not correctly predict the risk of CHD. However, 27 in the group expected to have a low frequency of hypoglycemia, the C-statistic was 0.646; 28 the predictability of the JJ risk engine was relatively accurate. Further, in the group of 29 patients using high-dose insulin, segregated from the group expected to have a high 30 frequency of hypoglycemia, the C-statistic was 0.866; thus, the JJ risk engine correctly 31 predicted the risk of CHD. Hence, the above results were not consistent in trend.