Association Between Relative Leg Length and Insulin Resistance in Rural, Urban and Rural–Urban Migrant Populations of Peru

Abstract

Objective

To determine the association between relative leg length and insulin resistance according to rural, urban, and rural–urban migrant groups.

Methods

Cross-sectional study using data from the PERU MIGRANT study (2007–2008). The exposure was relative leg length categorized as short, normal, or long, and the outcome was insulin resistance (logarithm of homeostatic model assessment log-HOMA2-IR). Linear regression models with log transformation, adjusted for sex, age, parental education, hip circumference, and physical activity level, were employed to estimate geometric mean ratios of insulin resistance across leg length categories. Interaction effects of population groups (rural, urban, and migrants) on insulin resistance were explored, along with mediation analysis of central obesity and excess body fat in the main relationship.

Results

Using data from 947 participants, 52.7% female, mean age 47.7 years (SD = 11.9), we found a robust inverse association between relative leg length and insulin resistance. The geometric mean of insulin resistance in subjects with long leg length was 43% (e^β1: 0.57, 95% CI: 0.47–0.69) lower than those in the normal category. A significant interaction effect of the population group on relative leg length categories (p < 0.001) was observed, particularly in the migrant and rural groups. Excess body fat and abdominal obesity explained 33% and 12% of the association between relative leg length and insulin resistance, respectively.

Conclusions

Longer leg length was associated with lower insulin resistance values, with a greater interaction effect observed among the rural–urban migrant and urban groups. These findings support the hypothesis that metabolic disorders in adults may be traceable to nutritional and developmental conditions early in life.