Body mass index mediates the association between meat intake and insulin sensitivity

Abstract

The purpose of this study was to examine the possible inverse relationship between total meat intake and processed meat intake and insulin sensitivity as mediated by body mass index (BMI). The Quantitative Insulin Sensitivity Check Index (QUICKI) was used to assess participants' insulin sensitivity: decreased insulin sensitivity indicated at a score of <0.36. Cross-sectional data were used from Visit 1 of the Atherosclerosis Risk in Communities (ARIC) cohort. Mediation analysis was conducted using the PROCESS macro for SPSS. The mean BMI was 27.3 ± 5.0 kg/m², total meat intake was 1.76 ± 0.89 servings per day, processed meat intake was 0.44 ± 0.46 servings per day, and the mean QUICKI score was 0.34 ± 0.039. This suggested decreased insulin sensitivity was prevalent in this population. As hypothesized and previously shown, both total meat intake and processed meat were inversely and significantly associated with insulin sensitivity after adjusting for key covariates including BMI. In mediation analysis, the significant inverse direct association model between total meat and QUICKI was -0.0017 (95% CI: -0.0026, -0.0008). BMI had a significant indirect association with QUICKI of -0.0028 (95% CI: -0.0033, -0.0023) which accounted for 62% of the association between total meat intake and insulin sensitivity. In the processed meat intake and QUICKI mediation model, BMI accounted for a significant 43% of the relationship. Our results suggest that the association between total meat and processed meat intake and insulin sensitivity is mediated by BMI.

Keywords: body mass index; diet surveys; insulin sensitivity; meat; meat products; regression analysis.