Zinc intake and its association with lipid and glycemic profile: American NHANES data 2011-2014

Abstract

Hypozincemia results in changes in glycemic and lipid profiles and others have found improvements in these profiles after zinc supplementation in humans and animal models. We propose to analyze dietary/plasma zinc, plasma lipids and glycemia in ~4500 children and adults from the U.S. NHANES survey. The aims of this cross-sectional study are to: (a) evaluate the relationship between zinc intake and lipid and glycemia profiles; (b) investigate the effect of zinc-nutrients interactions on lipid and glycemia profiles; (c) characterize different food patterns based on zinc content associated with lipid and glucose profiles. We will use the cross-sectional data from the 2011/2014 National Health and Nutrition Examination Surveys (NHANES) which is a nationally representative cross-sectional survey conducted by the US National Center for Health Statistics (Centers for Disease Control and Prevention, Atlanta, GA, USA. The NHANES data set that will be included in the analysis will be anthropometric, laboratory results, and diet intake. Inclusion criteria will be age of 6 years and older, in participants with available demographic, with laboratory results and nutrient intakes derived from at least 1-day dietary recall questionnaire.Spearman and partial correlations between zinc intake and outcome variables will be carried out. Multinomial logistic regression models and multiple linear regressions will be used to better explore the associations between zinc intake (independent variable) and dependent variables such as lipid profile, glycemia and serum zinc stratified by age (6-12 years; 13-19 years; above 19 years), and gender (male and female). Several confounding factors, such as race, poverty:income ratio (PIR), TV, computer, and video game use, C-reactive protein, phytate:zinc ratio in the diet, and caloric intake, will be adjusted in multinomial logistic regressions and multiple linear regressions. Principal component analysis and reduced rank regression will be used to generate different dietary patterns that may or may not contain zinc. We also will use a regression method to calculate the factor scores of each nutrient pattern for each study participants. A p-value < 0.05 will be considered as statistically significant. Bonferroni-adjusted p-value will be applied for multiple comparisons in logistic regression models. (AU)