Examining Sex and Race Differences in Obesity Risk Factors in Older Americans Using Genetic and Lifestyle Data in Regards

Abstract

Obesity is a prevalent chronic condition determined by social, psychological, behavioral, and biological factors. The prevalence of obesity remains a major public health issue and has continued to rise in the United States and around the world. People who are obese face a higher risk of cardiovascular and metabolic diseases. The fat mass and obesity-associated gene (FTO) is one of the most commonly identified genes in obesity-related research.

This three-paper dissertation is aimed at improving understanding of sex and race differences in obesity risk factors in older Americans from genetic (FTO SNPs) and lifestyle (diet and physical activity) perspectives, using cross-sectional (Chapter 2) as well as longitudinal (Chapters 3 & 4) data. African Americans and females reported relatively lower educational attainment, which was significantly associated with higher body mass index (BMI). Five FTO SNPs (rs1558902_A, rs3751812_T, rs9941349_T, rs9930506_G, and rs9922708_T) were significantly associated with higher BMI. Compared to African Americans, allele frequencies of the five SNPs were much higher in European Americans. Regardless of race, BMI potentially mediated the associations between rs1558902_A and hypertension, and rs3751812_T and hypertension. Analysis showed that changes in self-reported daily categories of food consumption, except change in meat consumption, were not significantly associated with changes in obesity indicators in older Americans between two examinations (roughly 9 years). And the 11 FTO SNPs did not significantly moderate the association between changes in self-reported daily meat consumption and changes in obesity indicators. But in cross-sectional analysis, two SNPs (rs1558902_A and rs3751812_T) were significant predictors for BMI in T1. The daily consumption of meat was statistically positively associated with BMI, which varied by race but was not moderated by the two SNPs. In the Chapter 4, the 11 FTO SNPs did not significantly moderate the association between physical activity and changes in obesity indicators. The potential roles of FTO SNPs on changes in obesity indicators and the efficacy of physical activities may vary by study designs, characteristics of study participants, types of exercise interventions, and measurements of physical activity.

In this dissertation, multiple linear regression models combining lifestyle and genetic data were tested in cross-sectional and longitudinal analyses, which we believe adds to the current literature on this complex problem. More studies should focus on discovering risk alleles or risk scores unique in African Americans and exploring other non-genetic influences to inform mechanisms in the prevention and control of obesity. Lifestyle modifications for reducing obesity rate among populations with genetic susceptibility are a promising approach but require further investigation.