Waist circumference is an independent correlate of errors in self-reported BMI

Obesity (Silver Spring). 2010 Nov;18(11):2237-9. doi: 10.1038/oby.2010.178. Epub 2010 Aug 19.

Abstract

Although there are issues of reporting bias surrounding the use of self-reported BMI, it is frequently the method employed to establish the prevalence of obesity. The goal of this study was to assess whether, independently of measured BMI, waist circumference (WC) was associated with the magnitude of the difference between self-reported and measured BMI within a large sample of European-American (EA) and African-American (AA) adults. Self-reported height and weight, and measured height, weight, and WC were collected on 12,809 adults (61% women, 66% EA) aged 18-65 years. Mean negative BMI differences (self-reported minus measured BMI) were identified in all race-by-sex groups (AA men: -0.55; EA men: -0.63; AA women: -0.91; EA women: -0.67). WC was negatively associated with the BMI difference such that a higher WC was associated with greater under-reporting of BMI. However, after adjusting for age and measured BMI, WC was positively associated with the BMI difference in all race-by-sex groups. These results suggest that WC could be useful in gaining an insight into people's awareness of their own body size and fatness.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Adult
  • Aged
  • Bias
  • Black or African American
  • Body Mass Index*
  • Female
  • Humans
  • Linear Models
  • Longitudinal Studies
  • Male
  • Middle Aged
  • Obesity / epidemiology*
  • Prevalence
  • Self Report / standards*
  • Waist Circumference*
  • White People
  • Young Adult