Bigger influence of misreporting weight and height in overweight prevalence than in arithmetic differences among older population

Objective: This cross-sectional study is to examine the difference between direct measured and self-reported height, weight and BMI calculating from them. Under the hypothesis that they under-estimate the obesity prevalence, we compared prevalence of overweight and obesity by measured vs. reported, and characters of participants. Methods and Procedures: All participants (527 people, 301(57.1%) females, 47 to 95 years old) were recruited from the second wave of Korean Longitudinal Study of Ageing (KLoSA) with informed consent in 2008. Self-reported weight, height, and information on socio-demographic variables were from data of KLoSA second wave's questionnaires. Overweight and obesity were defined as BMI>=25 and >=30 kg/m2 respectively. Results: Mean of the individual differences between measured and self-reported weight were 2.8kg (SD 2.8) in men and 2.7(2.6) in women, whereas in height were 3.0cm (2.9) in men, 3.9(3.5) in women. In BMI, it was 1.3kg/m2 (1.2) in men and 1.8(1.5) in women. Differences in weight are more in heavy group while height's differences are evenly distributed in all groups. While prevalence of overweight and obesity by reported BMI was 19.0% in male, 28.6% in female, by measured BMI was 38.1% in male, 46.8% in female. Important factors for overweight and obesity by reported BMI were self-rated health status and income level, meanwhile in that by measured BMI, were age and married state. Conclusion: The prevalence of obesity from measured BMI is about 20% more than using reported, and related factors are different in the study of Korean middle aged and order community dwelling people.