Predicting the Development of Myocardial Infarction in Middle-Aged Adults with Type 2 Diabetes: A Risk Model Generated from a Nationwide Population-Based Cohort Study in Korea
- Author:
Seung-Hwan LEE
1
;
Kyungdo HAN
;
Hun-Sung KIM
;
Jae-Hyoung CHO
;
Kun-Ho YOON
;
Mee Kyoung KIM
Author Information
- Publication Type:Original Article
- From:Endocrinology and Metabolism 2020;35(3):636-646
- CountryRepublic of Korea
- Language:English
-
Abstract:
Background:Most of the widely used prediction models for cardiovascular disease are known to overestimate the risk of this disease in Asians. We aimed to generate a risk model for predicting myocardial infarction (MI) in middle-aged Korean subjects with type 2 diabetes.
Methods:A total of 1,272,992 subjects with type 2 diabetes aged 40 to 64 who received health examinations from 2009 to 2012 were recruited from the Korean National Health Insurance database. Seventy percent of the subjects (n=891,095) were sampled to develop the risk prediction model, and the remaining 30% (n=381,897) were used for internal validation. A Cox proportional hazards regression model and Cox coefficients were used to derive a risk scoring system. Twelve risk variables were selected, and a risk nomogram was created to estimate the 5-year risk of MI.
Results:During 7.1 years of follow-up, 24,809 cases of MI (1.9%) were observed. Age, sex, smoking status, regular exercise, body mass index, chronic kidney disease, duration of diabetes, number of anti-diabetic medications, fasting blood glucose, systolic blood pressure, total cholesterol, and atrial fibrillation were significant risk factors for the development of MI and were incorporated into the risk model. The concordance index for MI prediction was 0.682 (95% confidence interval [CI], 0.678 to 0.686) in the development cohort and 0.669 (95% CI, 0.663 to 0.675) in the validation cohort.
Conclusion:A novel risk engine was generated for predicting the development of MI among middle-aged Korean adults with type 2 diabetes. This model may provide useful information for identifying high-risk patients and improving quality of care.
