Aplicabilidade das equações de estimativa de peso e estatura em idosos residentes em instituições de longa permanência no município de Natal-RN

Abstract

The anthropometric assessment is important to detect changes in nutritional status and subsidize health interventions. Estimation methods can be employed in the absence of measurement of weight and height in the elderly. However, such methods are not always applicable. Objective: To compare and validate weight and stature estimation methods in the elderly living in nursing homes, Natal-RN. Methodology: The equations were evaluated qualitatively by the reproducibility of anthropometric measurements that made it up and quantitatively by mean differences analysis and plot, the coefficient of determination (R²), root mean square error (RMSE), intraclass correlation coefficient (ICC), and graphic residual analysis. In stature estimation methods were done cluster analysis to verify similarity of the different methods evaluated. It was adopted as significant p <0.05. It was considered applicable when the estimation method presented R²> 0.7; the lowest RMSE among the evaluated methods; ICC> 0.7; and its 95% confidence interval with less distance between the upper and lower limits. Results: It were evaluated 315 elderly from 10 nursing homes by Natal-RN. The average body weight was greater in the younger elderly and without restricting mobility. Height was measured only in the elderly without restricting mobility. Qualitative analysis showed a weight equation Eq.2 to Rabito et al. (2008) as the best reproducibility, since it is not used skinfold thickness. Quantitative analysis revealed Eq.2 of Rabito et al. (2008) as the best applicability for the whole population and different genders, age groups (60 to 69 years, 70-79 years and 80 years or more) and mobility restriction. In relation to height, it was considered the methods that use the demi-span as low reproducibility. The cluster analysis grouped the Latin American equations in a cluster, the US and in other did not group demi-span and arm length methods. The eq.4 Bermudez; Tucker (2000) was the best applicable for the entire population. The different group’s analysis found that for elderly males and 60-69 years, IV Bermúdez; Tucker (2000) method was applicable. Conclusion: Eq.2 of Rabito et al. (2008) was applicable to estimate weight in this population and the different groups analyzed. The eq.4 Bermudez; Tucker (2000) was applicable only for males and age group 60-69 years. There was no equation applicable to height estimation for female and aged 70 years or older. Thus, it is necessary to carry out further researches to develop specific stature and weight estimation methods for the Brazilian institutionalized elderly population