RESUMEN
AIMS: Waist circumference (WC) is a convenient measure of abdominal adipose tissue. It itself is a cardiovascular disease (CVD) and diabetes-risk factor and is strongly linked to other CVD risk factors. There are, however, ethnic differences in the relationship of WC to the other risk factors. The aim of this study was to determine the optimal cut-off points of WC and body mass index (BMI) at which cardiovascular risk factors can be identified with maximum sensitivity and specificity in a representative sample of the Tunisian adult population and to investigate any correlation between WC and BMI. METHODS: We used a sample of the Tunisian National Nutrition Survey, a cross-sectional population-based survey, conducted in 1996 on a large nationally representative sample, which included 3435 adults (1244 men and 2191 women) of 20 years or older. WC, BMI, blood pressure and fasting blood measurements (plasma glucose, total cholesterol, triglycerides) were recorded. Receiver operating characteristic (ROC) curve analysis was used to identify optimal cut-off values of WC and BMI to identify with maximum sensitivity and specificity the detection of high blood pressure, hyperglycaemia, high blood cholesterol and hypertriglyceridaemia. RESULTS: ROC curve analysis suggested WC cut-off points of 85 cm in men and 85 cm in women for the optimum detection of high blood pressure, diabetes and dyslipidaemia. The optimum BMI cut-off points for predicting cardiovascular risk factors were 24 kg/m(2) in men and 27 kg/m(2) in women. The cut-off points recommended for the Caucasian population differ from those appropriate for the Tunisian population. The data show a continuous increase in odds ratios of each cardiovascular risk factor, with increasing level of WC and BMI. WC exceeding 85 cm in men and 79 cm in women correctly identified subjects with a BMI of >/=25 kg/m(2), sensitivity of >90% and specificity of >83%. CONCLUSIONS: Based on the ROC analysis, we suggest a WC of 85 cm for both men and women as appropriate cut-off points to identify central obesity for the purposes of CVD and diabetes-risk detection among Tunisians. WCs of 85 cm in men and 79 cm in women were the most sensitive and specific to identify most subjects with a BMI >/=25 kg/m(2).
Asunto(s)
Enfermedades Cardiovasculares/prevención & control , Obesidad/complicaciones , Adulto , Antropometría/métodos , Índice de Masa Corporal , Femenino , Humanos , Masculino , Persona de Mediana Edad , Obesidad/etnología , Obesidad/metabolismo , Oportunidad Relativa , Valor Predictivo de las Pruebas , Curva ROC , Factores de Riesgo , Sensibilidad y Especificidad , Factores Sexuales , Túnez/etnología , Relación Cintura-CaderaRESUMEN
OBJECTIVE: The prevalence of diabetes mellitus is known to be increasing rapidly worldwide, but few population-based surveys have been undertaken in Africa or the Middle East. The aims of this study are to report the prevalence of diabetes mellitus and impaired fasting glucose (IFG) in Tunisia, to compare the prevalence to previous studies and to analyse the relationship between diabetes and age, sex, area of residency and body mass index (BMI). SUBJECTS AND SETTING: We have used data from the Tunisian National Nutrition Survey, a cross-sectional health study providing a large nationally representative sample of the Tunisian population including 3729 adults. We used the American Diabetes Association diagnostic criteria to determine the prevalence of diabetes mellitus and IFG. RESULTS: The overall diabetes prevalence was 9.9% (9.5% in men and 10.1 in women) giving age-adjusted prevalence of diabetes of 8.5% (7.3% in men and 9.6% in women). Step-wise logistic regression showed age of more than 40 years, urban residency and high BMI to each be significantly and independently related to diabetes prevalence. The prevalence of diabetes mellitus has more than doubled in Tunisia over a 15-year period. CONCLUSIONS: Our study has demonstrated a high prevalence of diabetes in the adult population with a wide difference among the rural and urban areas with an increased prevalence compared to previous studies. The results underline the need to increase public awareness and to emphasize the benefit of lifestyle modification in order to prevent type II diabetes.