Background: Obesity and physical inactivity are related to the development of insulin resistance, but their relative importance and conjoint effects are unclear.
Methods: We related body mass index (BMI) and self-reported leisure-time physical activity (PA) at the age of 50 years to insulin sensitivity measured with euglycemic insulin clamp technique and the presence of metabolic syndrome (MetS) at a subsequent examination, 20 years later, in 862 men free from diabetes and MetS at baseline.
Results: In a multivariable model including BMI, PA, homeostasis model assessment insulin resistance index, erythrocyte sedimentation rate, and all components of MetS at baseline, both BMI (beta, -0.19 mg/kg bodyweight/min per 1 kg/m; P<0.0001) and PA (adjusted least square means, 5.1, 5.2, 5.4, and 6.2 mg/kg bodyweight/min in individuals with sedentary, moderate, regular, and athletic PA, respectively; P=0.0035) were significant predictors of insulin sensitivity at age 70. When categorizing individuals into four groups by BMI and PA at baseline, insulin sensitivity at the age of 70 years decreased significantly over the following categories: multivariable-adjusted least square means, 5.8 (low BMI/high PA); 5.6 (low BMI/low PA); 5.1 (high BMI/high PA); and 4.6 (high BMI/low PA) mg/kg bodyweight/min, respectively; P value of less than 0.0001.
Conclusion: In our community-based sample of middle-aged men, BMI and PA were independent predictors of insulin resistance after 20 years of follow-up. Our results imply that obesity and physical inactivity may increase insulin resistance and metabolic risk by partly independent pathways, and emphasize the importance of strategies that address both obesity and physical inactivity to achieve increased public health.