Citation

Chernenko A, Meeks H, Smith KR. BMC Public Health 2019; 19(1): e100.

Affiliation

Department of Family and Consumer Studies, University of Utah, 225 South 1400 East Alfred Emery Building 228, Salt Lake City, UT, 84112, USA.

Copyright

(Copyright © 2019, Holtzbrinck Springer Nature Publishing Group - BMC)

DOI

10.1186/s12889-019-6391-3

PMID

30670035

Abstract

BACKGROUND: Driver license departments in many US states collect data on individuals' height and weight. These data can be useful to researchers in epidemiological and public health studies. As height and weight on driver license are self-reported, they may be prone to reporting bias. We compare height and weight obtained from driver license records and clinically measured height and weight, as well as body mass index (BMI) values calculated using the two data sources for the same individual.

METHODS: We linked individual height and weight records obtained from the Driver License Division (DLD) in the Utah Department of Public Safety to clinical records from one of the largest healthcare providers in the state of Utah. We then calculated average differences between height, weight and BMI values separately for women and men in the sample, as well as discrepancies between the two sets of measures by age and BMI category. We examined how well self-reported height and weight from the driver licenses classify individuals into specific BMI categories based on clinical measures. Finally, we used two sets of BMI values to estimate individuals' relative risk of type II diabetes.

RESULTS: Individuals, on average, tend to overestimate their height and underestimate their weight. Consequently, the value of BMI calculated using driver license records is lower than BMI calculated using clinical measurements. The discrepancy varies by age and by BMI category. Despite the discrepancy, BMI based on self-reported height and weight allows for accurate categorization of individuals at the higher end of the BMI scale, such as the obese. When used as predictors of relative risk of type II diabetes, both sets of BMI values yield similar risk estimates.

CONCLUSIONS: Data on height and weight from driver license data can be a useful asset for monitoring population health in states where such information is collected, despite the degree of misreporting associated with self-report.

Language: en

Keywords

Bias; Body mass index; Driver license; Self-report