Type 2 diabetes risk in persons with dysglycemia: the Framingham Offspring Study

Detection of risk of type 2 diabetes mellitus (T2DM) among adults with dysglycemia.

We used a nested case-cohort prospective design to estimate risk of new diabetes (diabetes treatment or FPG ≥7.0 mmol/L) among 1004 Framingham Heart Study Offspring with baseline dysglycemia [fasting plasma glucose (FPG) 5.4-6.9 mmol/L and/or 2-h post glucose load level 7.8-11.0 mmol/L]. Using clinical characteristics previously shown to predict incident T2DM, we used logistic regression to estimate odds ratios (OR), p-values for predictors, and assessment of model discrimination.

At the end of 7 years follow-up there were 118 incident T2DM cases. In a model that included age, sex, elevated blood pressure or blood pressure treatment, lipid-lowering treatment and elevated triglycerides, we found the following additional characteristics to be independently associated with new T2DM: parental history of diabetes (OR 2.28, p=0.004); excess adiposity (BMI ≥30 kg/m(2) or waist circumference ≥101.6 cm) (OR 2.04, p=0.0005), and low HDL-C [<1.0 (men) or <1.3 mmol/L (women)] (OR 2.77, p<0.0001). The multivariable C-statistic for this model was 0.701, and with glycemic category information included, c=0.751.

The key non-glycemic traits that predicted later T2DM in adults with dysglycemia were parental history of diabetes, excess adiposity and low HDL-C.