Low predictability of anthropometric indicators of obesity in metabolic syndrome (MS) risks among elderly women.

While diagnostic criteria for MS may vary depending on ethnicity, obesity remains a key risk factor in its development. In Taiwan, the incidence of obesity and MS among women has been increasing; however cut-off values for defining obesity for the diagnosis of MS among different groups of women have not been clearly established. The goal of this research was to examine the suitability of various anthropometric indicators of obesity in predicting the presence of MS criteria and to determine appropriate cut-off values of these indicators for women of different age and menstrual status. The sample was derived from the 2002 "Taiwan Three High Prevalence Survey" database. Women were divided into three groups based on age and menstrual status. Receiver-operating characteristic (ROC) curves was applied to the anthropometric indicators of obesity including, body mass index (BMI), waist circumference (WC), waist-to-hip ratio (WHR), waist-to-height ratio (WHtR), to ascertain its value in predicting MS. 2848 cases were included. It was found that most MS component values were worse with age and following menopause. Obesity indicators showed poor predictability for MS risks in post-menopausal women over 65 years, but good predictability in women under 65 years; our study revealed the following as ideal cut-off values for non-menopausal female: WHtR<0.49, WC<78 cm, WHR<0.79, BMI<24 kg/m(2); for menopausal women, WHtR<0.54, WC<83 cm, WHR<0.84, BMI<24.4 kg/m(2). It was concluded that obesity alone is not a reliable predictor of MS risks in women over the age of 65, and cut-off values for obesity indicators need to be further reduced in non-menopausal women.