The interaction of anthropometric variables with fasting blood glucose on progression from normal fasting glucose to overt diabetes (DM), is uncertain. In the Rochester Epidemiology Project (MN, USA), we identified 44,992 individuals with ≥ 2 fasting plasma glucose (FPG) tests from 2005-17, after excluding people who met criteria for DM (FPG > 125mg/dL) on/before their first FPG. The cohort was 58% female, had an age range of 18-65 yrs (44 ± 10, Mean ± SD) and BMI of 29 ± 7 kg/m2. Subjects who did not develop DM were censored at their last FPG value on/before Dec 31, 2017. Those who developed DM were assigned an event date as the first diagnostic value. Over a median follow-up of 7 yrs, 3888 (8.6%) developed DM. The Kaplan-Meier 10-year cumulative risk of incident DM was 12.8% (CI 12.4%-13.3%). As expected, multivariable Cox modeling identified elevated initial FPG [100-109mg/dL (HR 2.0; 1.9-2.2); 110-125mg/dL (HR 5.8; 5.4-6.3)] as conferring increased risk. Other independent risks were male sex (HR 1.30; 95% CI 1.22-1.38) and abnormal BMI [<18 kg/m2 (2.67; 1.96-3.63), 25-29.9 kg/m2 (1.37; 1.23-1.52), 30-34.9 kg/m2 (2.15; 1.92-2.39), ≥ 35 kg/m2 (3.59; 3.23-3.99)]. A very strong, nonlinear, age effect was also noted, with a fitted hazard ratio for age 60 versus 20 of 2.65. In this large cohort, male sex, age and abnormal category of BMI (including underweight) are associated with increased risk of progression to DM and additive to baseline FPG.


A.M. Egan: None. C. Wood-Wentz: None. K.R. Bailey: None. A. Vella: Advisory Panel; Self; vTv Therapeutics, Zealand Pharma A/S. Research Support; Self; Novo Nordisk A/S, Xeris Pharmaceuticals, Inc.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at