Cardiovascular Risk Prediction in Diabetic Men and Women Using Hemoglobin A1c vs Diabetes as a High-Risk Equivalent

Study Questions:

Can glycemic hemoglobin (HgA1c) be used as a predictor of high cardiovascular disease (CVD) risk among diabetic patients?


Women and men, including those with a diagnosis of diabetes at baseline, were followed prospectively for occurrence of CVD events. Female participants were enrollees of the Women’s Health Study (n = 39,896), a randomized trial of vitamin E and aspirin for the prevention of CVD and cancer. Male participants were enrollees of the Physician’s Health Study II (n = 14,641), a randomized trial of beta carotene, ascorbic acid, and vitamin E, and multivitamins. Participants over the age of 80 and those with missing data were excluded from the present study. Multivariate models were created for men and women separately, which included traditional CVD risk factors. HgA1c levels were included in these models only for the diabetic participants. In diabetic participants, the resulting predicted risks were compared with classification of diabetes as a cardiovascular risk equivalent (10-year CVD risk of at least 20%).


A total of 24,674 women (685 diabetic at baseline) and 11,280 men (563 diabetic at baseline) were included in this analysis. During follow-up, 125 CVD events and 170 CVD events were recorded in women and men, respectively. Among the women and men, those with diabetes were older and had a less favorable risk factor profile compared to nondiabetic participants, with the exception of smoking and family history for premature myocardial infarction. In women, the models including HbA1c levels in diabetic participants improved the C statistic by 0.177 (p < 0.001) over the risk equivalence model and showed improved reclassification (net reclassification improvement [NRI] of 26.7% [p = 0.001]). In men, the improvements were more modest, but still statistically significant (C statistic change of 0.039 [p = 0.02]; NRI of 9.2% [p = 0.04]). Including HbA1c levels also improved prediction over a dichotomous term for diabetes in women (NRI of 11.8% [p = 0.03]), but not in men.


The authors concluded that among diabetic men, HgA1c levels improved predictive ability of current CVD risk models.


These data provide an interesting comparison of prediction models for men and women using HgA1c levels to assist in prediction of CVD risk among diabetic patients. The central question is whether these findings ultimately change current management. Does the addition of A1c levels to current risk prediction models improve our ability to identify high-risk patients in which more aggressive treatment will result in reduced CV events? Recent studies suggest identifying diabetics early and then treating risk factors aggressively to improve risk for CVD events; thus, the lack of information on HgA1c over time limits the usability of these findings. In addition, the use of cohorts that are predominately Caucasian and lack diversity from a socioeconomic standpoint adds additional limitations.

Keywords: Myocardial Infarction, Follow-Up Studies, Hemoglobins, Vitamin E, Blood Glucose, Cardiology, European Continental Ancestry Group, Cardiovascular Diseases, Women's Health, Risk Factors, Diabetes Mellitus

< Back to Listings