AIMS: To compare two proposed definitions of the metabolic syndrome and to determine the clinical importance of the syndrome with respect to its association with coronary heart disease (CHD). METHODS: Cross-sectional study of 3770 women aged 60-79 years randomly selected from 23 British towns. RESULTS: The prevalence of the metabolic syndrome was high in this population and similar with both definitions: 28.2% (95% confidence interval 26.8, 29.7%) of the women had metabolic syndrome according to a modified version of the WHO definition, and 29.2% (27.7, 30.7%) had the ATP III-defined syndrome. There was reasonable agreement between the two definitions, with 79% of the participants being similarly classified by both definitions. The syndrome was associated with prevalent CHD, with the magnitude of the association with CHD being similar for both definitions. The odds ratio (95% confidence interval) for the age, smoking, physical activity, adult and childhood social class adjusted association of the WHO defined syndrome with prevalent CHD was 1.45 (1.19, 1.75) and for the ATP III-defined syndrome was 1.53 (1.27, 1.85). Insulin resistance alone, hypertension alone and dyslipidaemia alone were all associated with CHD, with the magnitudes of these associations being similar to those for the WHO and ATP III-defined syndrome with CHD. CONCLUSIONS: The prevalence of the metabolic syndrome is high in older British women and is associated with CHD. There is reasonable agreement between a modified version of the WHO definition and the ATP III definition of the syndrome, and both are similarly associated with CHD. Single components of the syndrome are associated with CHD to a similar magnitude as the syndrome.