Accuracy of self-reported height, weight, and waist circumference in poor, rural areas of China

Objectives: There are differing opinions and fi ndings about the accuracy of self-reported data. This study sought to assess the accuracy of self-reported height, weight, and waist circumference (WC) data among adult residents in poor, rural areas of China. Methods: Multi-stage stratified cluster random sampling was used to sample rural adult residents in Jinzhai County. Intraclass correlations (ICCs), Bland-Altman plots with limits of agreements (LOAs), and Cohen's kappa were used to examine the accuracy of self-reported data. Multiple linear regression was used to explore the effect of demographic and socioeconomic variables. Results: Height was over-reported by an average of 0.67 cm, weight and WC were both under-reported, by an average of 0.004 kg and 1.09 cm, respectively. The ICC for height was 0.92, the ICC for weight was 0.93, and the ICC for WC was 0.93. Bland-Altman plots showed no trend in variability with increasing height, weight, and WC, but 95% of the LOAs were wide, especially for WC. The percentage agreement (kappa) for BMI was 82.10% (0.71), and the percentage agreement for WC was 83.56% (0.66). Overall, obesity, overweight, and increased WC were underestimated by 0.56%, 2.64%, and 6.89%, respectively. Assessments of which demographic and socioeconomic factors were associated with statistically significant differences between self-reported and measured data showed that education (junior school vs. primary or below: /3 = - 0.56, 95% CI [-1.01 to - 0.11]) and occupation (others vs. professionals: /3 = 2.16, 95% CI [0.62-3.70]) were associated with differences between measured and self-reported height; occupation (industrial and commercial service vs. professionals: /3 = 1.57, 95% CI [0.52-2.61]) was associated with differences between measured and self-reported weight; and gender (women vs. men: /3 = - 1.74, 95% CI [-2.51 to - 0.97]) and age (60- vs. 18-44: /3 = - 1.62, 95% CI [-2.88 to - 0.36]) were associated with differences between measured and self-reported WC. Conclusions: Self-reported height and weight were shown to be reliable in poor, rural areas of China, but self-reported WC should be used with caution. To improve the accuracy of self-reported data, certain subgroups should be targeted for educational interventions.