Abstract Objective The International Diabetes Federation (IDF) has proposed ethnicity specific cut-off values for waist circumference, as an essential criterion for the diagnosis of the metabolic syndrome (MetS). However, before introducing waist circumference measurement as part of the clinical examination, or an obligatory criterion for MetS, it is important to ensure that the measurement is reliable and reproducible. Study design The inter- and intra-operator variability in measuring waist circumference was assessed in a preliminary study of 10 health professionals and the inter-operator variability in clinical practice was assessed in a study of 102 patients. Repeated measures analysis of variance (ANOVA) was used to assess the significance of the inter- and intra-operator variability. Results Intra-operator variability was not found to be significant in the preliminary study (F = 0.15, p = 0.764). However, the inter-operator variability was significant in both studies (preliminary: F = 4.16, p<0.001, clinical practice study: F = 14.06, p<0.001). In clinical practice, this could lead to disagreement among operators regarding the presence of central obesity in 9% of the patients. After giving written instructions on how to measure waist circumference, the coefficient of variation was not significantly altered (p = 0.202) but the inter-operator variability lost significance (F = 2.34, p = 0.11), suggesting a reduction in systematic error. Conclusions Measuring waist circumference is subject to significant inter-operator variability and could potentially lead to misclassifying patients as having the MetS, or not. Better training of health professionals on how to measure waist circumference properly is needed in order to ensure that patients are not misclassified and that international comparisons of the prevalence of the MetS are reliable.