Background: Waist circumference (WC) is used to indirectly measure abdominal adipose tissue and the associated risk of type 2 diabetes mellitus (T2DM) and cardiovascular disease (CVD). Because of its easy implementation and low cost, self-measured WC is commonly used as a screening tool. However, discrepancies between self-measured and objectively measured WC may result in misclassification of individuals when using established cut-off values. The aim of this study was to determine the accuracy of self-measured WC in adults at risk of T2DM and/or CVD, and to determine the anthropometric, demographic and behavioural characteristics associated with bias in self-measured WC.
Methods: Self-measured and objectively measured WC was obtained from 622 participants (58.4% female; mean age 43.4 ± 5.3 years) in the Hoorn Prevention Study. The associations of gender, age, educational level, body mass index, smoking status, dietary habits, physical activity and sedentary behaviour with the discrepancies between self-measured and objectively measured WC were analysed using independents t-test and one-way ANOVA. Bland-Altman plots were used to plot the agreement between the two measures.
Results: On average, self-measured WC was overestimated by 5.98 ± 4.82 cm (P < 0.001). Overestimation was consistent across all subgroups, but was more pronounced in those who were younger and those with lower educational attainment.
Conclusions: The results support self-measured WC as a useful tool for large-scale populations and epidemiological studies when objective measurement is not feasible, but overestimation should be taken into account when screening adults at risk of T2DM and/or CVD.