Differences in accuracy of height, weight, and body mass index between self-reported and measured using the 2018 Korea Community Health Survey data

Abstract

OBJECTIVES This study aimed to determine an effective survey method for the accurate calculation of obesity prevalence by comparing the self-reported and measured height, weight, and body mass index (BMI) using the 2018 Korea Community Healthy Survey (CHS) data. METHODS Raw data from the 2018 CHS were used to analyze the differences, correlation, and agreement between self-reported and measured height, weight, and BMI. RESULTS The self-reported height was over-reported than the measured height (0.59 cm greater for men and 0.71 cm greater for women), while the self-reported weight was under-reported than the measured weight (0.55 kg less for men and 0.67 kg less for women). Subsequently, the self-reported BMI was under-estimated (0.35 kg/m 2 lower for men and 0.49 kg/m 2 lower for women) compared with the measured BMI. The kappa statistic and agreement between measured and self-reported values per BMI category (underweight, normal, overweight, and obesity) were 0.82 and 79.6%, respectively. CONCLUSIONS The prevalence of obesity should be calculated using the measured values provided in the CHS in order to promote local health projects based on accurate evidence.

Validity of Measured vs. Self-Reported Weight and Height and Practical Considerations for Enhancing Reliability in Clinical and Epidemiological Studies: A Systematic Review

“…When compared with the measured values, the reported height values for men and women were 0.59 and 0.71 cm greater, respectively. In other words, males as well as females reported greater height values than their real measures as they grew older [ 11 ]. In a similar vein, the study by Fry et al (2022) showed a similar trend, indicating that as males and females age, they tend to overestimate their weight and height [ 10 ].…”

“…Another study revealed that compared with the measured height, which was 0.71 cm more for women and 0.59 cm greater for men, the self-reported weight was 0.67 kg less for women and 0.55 kg less for males. Consequently, when compared with the measured BMI, the self-reported BMI was understated (0.35 kg/m 2 lower for males and 0.49 kg/m 2 lower for women) [ 11 ].…”

Actividad física durante el embarazo y su relación con la ganancia de peso gestacional

“…Esta es una circunstancia habitual en este tipo de estudios e inherente a la dificultad de averiguar el peso pre-gestacional. En todo caso, se han observado anteriormente altas correlaciones (r>0,92) entre el peso medido y reportado (36) , y diferencias reales entre ambos tipos de medición menores a 1 Kilogramo (37) que justifican la sustitución de los datos medidos por los auto-reportados. Por último, este estudio no incluye algunos factores potencialmente confusores como las náuseas del 1 er trimestre o el resultado de los test de tolerancia a la glucosa que pueden subestimar o sobreestimar la influencia de la AF sobre la GPG.…”

“…This is a usual circumstance in this type of studies and inherent to the difficulty determining pre-gestational weight. Nevertheless, high correlations (r > 0.92) have already been observed between measured and reported weight (36) , as well as actual differences lower than 1 kilogram between both types of measurement (37) , which justify substituting the measured data with the self-reported ones. Finally, this study does not include some potentially confounding factors such as 1 st trimester nausea or the result of the glucose tolerance tests, which can under- or over-estimate the influence exerted by PA on GWG.…”

Tobacco consumption, sales, and output as monitoring indicators in the era of the tobacco endgame: a Korean example

“…Non-parametric Spearman rank analyses were used to seek correlations between indicators. Bland–Altman plots were drawn to explore the extent of agreement between pairs of indicators [ 15 , 16 ]. The results from the Bland–Altman analyses are rather robust regardless of data normality [ 17 ].…”