Capacidade preditiva e associação de indicadores antropométricos com diabetes mellitus, hipertensão arterial e comprometimento da capacidade funcional em idosos

Abstract

Measurements and anthropometric indices of adiposity are simple and inexpensive useful tools to identify individuals at high risk of disability and morbidity, especially those related to cardiometabolic alterations. However, there are still gaps in knowledge and controversies of the anthropometric indicators more appropriate for the assessment of the elderly. The aim of this study was to investigate the predictive ability and the association of anthropometric indicators with diabetes mellitus (DM), hypertension arterial (HA) and impaired functional capacity in the elderly. This cross-sectional study evaluated a representative sample of 621 elderly in Viçosa, Minas Gerais, Brazil selected by simple random sampling without replacement. The dependent variables were self-reported hypertension, diabetes mellitus and impaired functional capacity as latent variable obtained from basic and instrumental activities of daily living. They were taken the following anthropometric measurements: weight, height, waist, hip, arm and calf circumferences and the following anthropometric indices were calculated: a body shape index (ABSI), body roundness index (BRI), conicity index (CI) index body adiposity (BAI), body mass index (BMI), the waist-height ratio (WHtR), waist-calf ratio (WCR), waist-hip ratio (WHR), and waist-hip-height (WHHtR). The following covariates were obtained: age, sex, education, income elderly, physical exercise, smoking, alcohol intake and diet quality. Regarding the anthropometric characteristics, it was found that women had higher mean BMI, WHtR, BAI and arm circumference (p <0.001) and men of weight, height and WHR (p <0.001). Weight, calf and arm circumferences were lower in older age groups (p <0.001) and the same was true for BMI and height among men (p <0.05). The anthropometric indices with the strongest association, positive and statistically significant with HA were WHHtR in men and the BAI and BMI in women. For DM outcome, the index with the strongest association, positive and statistically significant was WHR in men and WCR in women. The hip circumference showed statistically significant inverse association with HA in men and DM in both sexes. Regarding the predictive ability, ABSI, BRI, BAI and WHHtR did not show better predictive ability than traditional anthropometric indices (BMI, WC, WHR, WHHtR) to discriminate diabetes mellitus and hypertension in both sexes, except WCR that was better than indices traditional to predict DM in women. Cutoff points of anthropometric measurements were identified by sex to discriminate HA and DM, but the sensitivity and specificity were low. Through factor analysis of principal components were identified two anthropometric patterns: global adiposity and central adiposity. After adjustment for confounders, Poisson regression with robust variance showed that the only association that remained significant was that between anthropometric measures of central adiposity and DM in women. Regarding the impairment of functional capacity, the analysis adjusted for confounding factors showed a positive and statistically significant association of the same with BMI (standardized coefficient: 0.211; p <0.001) and waist circumference (standardized coefficient: 0.163; p <0.001). This study showed that HA, DM and impaired functional capacity were associated with global and central adiposity in elderly and that BMI, waist circumference, WHR and WCR are most suitable for anthropometric assessment of the elderly.