Abstract
This cross-sectional investigation examined the relationship between sitting time and insulin resistance in 6931 U.S. adults. The mediating effects of several covariates were evaluated. Self-reported sitting time, measured in minutes per day, was the exposure variable. Insulin resistance (IR), indexed using the natural log of the homeostatic model assessment of insulin resistance (L-HOMA-IR), was the outcome variable. This study used data collected from the 2011-2018 National Health and Nutrition Examination Survey (NHANES). Results showed a strong, positive, dose-response association between sitting time and insulin resistance after adjusting for age, sex, race, and year of assessment (F = 12.6, p < 0.0001). Across the sitting time tertiles (low, moderate, and high), the L-HOMA-IR means (±SE) each differed from each other (0.37 ± 0.008, 0.40 ± 0.012, and 0.43 ± 0.012). Further controlling for cigarette smoking and physical activity did not alter the significance of the relationship. Adding body mass index (BMI) to the demographic covariates weakened the relationship, but it remained significant. However, the association was no longer significant after adjusting for the demographic covariates and waist circumference (F = 1.1, p = 0.3349). None of the L-HOMA-IR means (±SE) differed from each other (0.40 ± 0.007, 0.41 ± 0.009, and 0.41 ± 0.008). Overall, waist circumference was a powerful mediating variable between sitting time and insulin resistance. Apparently, time spent sitting is a powerful predictor of IR. However, much of the association between sitting time and IR is a function of differences in waist size. As a strong measure of abdominal adiposity and a significant predictor of multiple metabolic diseases, managing waist size is a health practice to consider when insulin resistance is a concern.