Obesity Heterogeneity and Dyslipidemia in Urbanizing China

Public Deposited
Analytics

Downloadable Content

Download PDF
Request Version for Screen Reader
Last Modified
  • April 27, 2020
Creator
  • Jane Williford
    • Gillings School of Global Public Health, Department of Biostatistics
Abstract
  • Background: Although obesity is a major risk factor for dyslipidemia, little is known about the heterogeneity of obesity, including differences in body fat distribution, the duration and timing of weight change, and whether they are associated with differential risk of dyslipidemia, especially in Asian populations. Methods: Using mixture models, latent classes were derived based on cross-sectional measures of waist circumference, % body fat in 2015 and longitudinal BMI data (from 1991 to 2015), for 6,258 adults (aged 18–66 years in 2015) from the China Health and Nutrition Survey. Descriptive statistics, as well as ANOVA and Chi-squared tests, were used to test for unadjusted differences by classes. We then used multivariable general linear mixed effects models to test for associations between these latent classes and both obesity measures and metabolic markers of dyslipidemia (LDL, HDL, total cholesterol, triglycerides, apo A, apo B, Lp a, and alanine aminotransferase). We controlled for age and sex in all models with random intercepts to control for community-level and household-level correlations. In the dyslipidemia-specific models we additionally controlled for BMI in 2015, alcohol use and smoking status. Results: Four of the five classes had comparable average BMI values in 2015 (from 23.17 kg/m2 (0.26 standard error (s.e.)) to 23.94 kg/m2 (0.94 s.e.)) after adjustment for age and sex but noticeable differences in change in BMI prior to 2015, waist circumference, and % body fat in 2015. Classes differed by sex, age, wave of entry, and number of survey waves. An association between latent class membership and HDL was identified, with the high overall obesity over time class having 3.22 mg higher average HDL values than the class with Lower BMI, High % Body Fat in 2015 class. Conclusion: Evidence of different patterns of obesity was found, with results indicating these patterns may be associated with differences in HDL values.
Date of publication
DOI
Resource type
Rights statement
  • In Copyright
Advisor
  • Annie Green Howard
    • Gillings School of Global Public Health, Department of Biostatistics
    • Carolina Population Center
Degree
  • Bachelor of Science in Public Health
Graduation year
  • 2020
Language
  • English

Relations

Parents:

This work has no parents.

Items

Thumbnail Title Date Uploaded Visibility Actions
file details Jane_Williford_Honors_Thesis_Unsigned_Final.pdf 2020-04-27 Public Download