Sex differences of anthropometric indices of obesity by age among Iranian adults in northern Iran: A predictive regression model

BACKGROUND: Background and Objectives: The biological variation of body mass index (BMI) and waist circumference (WC) with age may vary by gender. The objective of this study was to investigate the functional relationship of anthropometric measures with age and sex. METHODS: The data were collected from a population-based cross-sectional study of 1800 men and 1800 women aged 20-70 years in northern Iran. The linear and quadratic pattern of age on weight, height, BMI and WC and WHR were tested statistically and the interaction effect of age and gender was also formally tested. RESULTS: The quadratic model (age(2)) provided a significantly better fit than simple linear model for weight, BMI and WC. BMI, WC and weight explained a greater variance using quadratic form for women compared with men (for BMI, R(2)=0.18, p<0.001 vs R(2)=0.059, p<0.001 and for WC, R(2)=0.17, p<0.001 vs R(2)=0.047, p<0.001). For height, there is an inverse linear relationship while for WHR, a positive linear association was apparent by aging, the quadratic form did not add to better fit. CONCLUSION: These findings indicate the different patterns of weight gain, fat accumulation for visceral adiposity and loss of muscle mass between men and women in the early and middle adulthood.