Anthropometric indicators of obesity such as predictors of high blood pressure in the elderly

Keila de Oliveira Diniz, Saulo Vasconcelos Rocha, Antonio Cesar Cabral de Oliveira



High blood pressure is a leading cause of mortality worldwide and a risk factor for several diseases. The aim of this study was to determine the predictive power of anthropometric indicators of obesity and establish their cutoff points as discriminators of hypertension and identify the anthropometric indicator of obesity that best discriminates high blood pressure in the elderly. This is a cross-sectional study with a sample of 300 older adults, 167 (56.5%) women. The following anthropometric indicators of obesity were measured: body mass index (BMI), waist circumference (WC), waist / height ratio (WHtR) and conicity index. Moreover, systolic and diastolic blood pressure measurements were collected. To identify hypertension predictors, the analysis of receiver operating curves (ROC) with 95% confidence interval was adopted. Subsequently, cutoff points with their respective sensitivities and specificities were identified. Analyses were carried out considering 5% significance level. It was observed that some anthropometric indicators of obesity showed area under the curve (AUC) significant with BMI = 0.60 (0.50 to 0.70); WHtR = 0.61 (0.51 to 0.71); conicity index = 0.58 (0.58 to 0.68) in men. The different cutoff points of anthropometric indicators with better predictive power and their respective sensitivities and specificities were identified. The best areas under the ROC curve were for BMI, WHtR and conicity index for men; however, such measures were not satisfactory to predict high blood pressure in women.


Anthropometry; Older adults; Obesity; Arterial pressure

Full Text:



The abbreviated title of the journal is Rev. Bras. Cineantropom. Desempenho Hum, which should be used in bibliographies, footnotes and bibliographical references. E-ISSN 1980-0037, impressa ISSN 1415-8426, Florianópolis, Santa Catarina, Brazil. This work is licensed under a Creative Commons Attribution 4.0 International License.