Baseline Fasting Glucose Level, Age, Sex, and Body Mass Index and the Development of Diabetes in US Adults

Understanding the interplay between diabetes risk factors and diabetes development is important to develop individual, practice, and population-level prevention strategies.

To evaluate the progression from normal and impaired fasting glucose levels to diabetes among adults.

This retrospective community-based cohort study used data from the Rochester Epidemiology Project, in Olmsted County, Minnesota, on 44 992 individuals with at least 2 fasting plasma glucose (FPG) measurements from January 1, 2005, to December 31, 2017. People who met criteria for diabetes on or before their first FPG measurement were excluded. Data were electronically retrieved in December 2019 with analyses finalized in November 2024.

The exposure was baseline FPG level, with covariates including the following measures that are consistently recorded in the electronic health record: body mass index (BMI), age, and sex.

The cumulative probability of freedom from diabetes was estimated and presented graphically using a Kaplan-Meier curve. Multivariable Cox proportional hazards regression modeling was used to estimate the partial hazard ratios (HRs) for variables of interest. Diabetes was defined as an FPG level greater than 125 mg/dL.

A total of 44 992 individuals (mean [SD] age at baseline, 43.7 [11.8] years; 26 025 women [57.8%]) were included. The baseline mean (SD) BMI was 28.9 (6.6). Over a median follow-up of 6.8 years (IQR, 3.6-9.7 years), 3879 individuals (8.6%) developed diabetes. The Kaplan-Meier 10-year cumulative risk of incident diabetes was 12.8% (95% CI, 12.4%-13.2%). All initial FPG levels outside a range of 80 to 94 mg/dL were associated with increased risk for diabetes (ie, FPG <70 mg/dL: HR, 3.49 [95% CI, 2.19-5.57]; FPG 120-125 mg/dL: HR, 12.47 [10.84-14.34]). Other independent risk factors were male sex (HR, 1.31 [95% CI, 1.22-1.40]), older age (≥60 years: HR, 1.97 [95% CI, 1.71-2.28]), and any abnormal category of BMI, including underweight (BMI <18.5: HR, 2.42 [95% CI, 1.77-3.29]; BMI ≥40: HR, 4.03 [95% CI, 3.56-4.56]). There was a significant additive association of variables, particularly FPG level and BMI. For instance, a woman aged 55 to 59 years with a BMI of 18.5 to 24.9 and an FPG level of 95 to 99 mg/dL had an estimated 10-year diabetes risk of 7.0%. However, an almost doubling of risk to 13.0% was observed if the BMI was 30.0 to 34.9, and risk more than doubled again to 28.0% if FPG level also increased to 105 to 109 mg/dL. A nomogram was generated to facilitate individual classification into one of four 10-year risk categories.

This retrospective cohort study of 44 992 individuals suggests that FPG level, age, BMI, and male sex were all associated with development of diabetes, with significant interaction between these variables. These data contribute to understanding the clinical course of diabetes and highlight the substantial individual variation in diabetes risk according to commonly measured clinical variables. The findings facilitate lifestyle and pharmacologic interventions to treat those at highest risk of diabetes to reduce future morbidity and mortality. Further work is needed to validate this risk categorization tool for different populations.