Body fatness and waist circumference are independent predictors of the age-associated increase in fasting insulin levels in healthy men and women.

Objective: Elevated levels of fasting insulin increase the risk for coronary artery disease. In this study, we considered the independent contributions of age, total body fatness, abdominal fat distribution, peak aerobic capacity, leisure time physical activity, dietary intake, and fasting glucose levels to the age-associated increase in fasting insulin levels in healthy men and women.
Design: A cross-sectional analysis.
Subjects: A total of 427 healthy men and 293 healthy women age 18 to 90 years.
Measurements: Plasma glucose and insulin levels were measured in the fasted state and subjects were characterized for body fatness from underwater weighting, abdominal fat distribution from the waist circumference, peak VO2 from a treadmill test to exhaustion, leisure time physical activity from a structured interview and dietary intake from a 3-day food diary. The independent predictors of fasting insulin levels were determined by stepwise multiple regression analysis.
Results: Fasting insulin concentrations increased with age in both men (0.09 +/- 0.02 pmol/yr, unadjusted slope +/- SEE) and women (0.14 +/- 0.03 pmol/yr). In men, the major independent predictors of fasting insulin levels were percent body fat (r2 = 28%, P < 0.001) and waist circumference (r2 = 3% P < 0.001). Age, glucose concentration, and peak VO2 were small (1% each), but independent contributors to the variance in insulin levels. In women, the waist circumference (r2 = 18%, P < 0.001) and the fasting glucose concentration (r2 = 7%, P < 0.001) were the independent correlates of fasting insulin concentrations.
Conclusions: Our data suggest that total adiposity and central body fat distribution are significant determinants of the increase in fasting insulin levels, whereas age, dietary intake, and levels of aerobic fitness and physical activity appear to be less important contributors to the variation in fasting insulin concentrations in healthy men and women.