Modelos preditivos de (in) segurança alimentar e nutricional, segundo indicadores socioeconômicos, demográficos e nutricionais de famílias brasileiras
Arquivos
Data
2018-02-27
Autores
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Federal de Viçosa
Resumo
Segurança alimentar e nutricional é definida como garantia de acesso regular e permanente a alimentos de qualidade, em quantidade suficiente, sem comprometer o acesso a outras necessidades essenciais, de indivíduos e famílias. A avaliação da (in) segurança alimentar e nutricional demanda o emprego de diferentes indicadores, que abarquem as dimensões de acesso, disponibilidade e utilização biológica dos alimentos, visando contemplar seu amplo conceito. Este estudo objetivou avaliar a (in) segurança alimentar e nutricional segundo indicadores de percepção, renda, disponibilidade calórica, consumo alimentar e estado nutricional, por localização e situação dos domicílios, e construir modelos de predição de insegurança alimentar e nutricional de famílias brasileiras, a partir destes indicadores. Empregou-se dados secundários referentes às famílias avaliadas na Pesquisa de Orçamentos Familiares (POF), realizada pelo Instituto Brasileiro de Geografia e Estatística - IBGE, em 2008- 2009. Para a avaliação da insegurança alimentar e nutricional utilizou-se os seguintes indicadores: percepção sobre a quantidade insuficiente de alimento consumida pela família; renda familiar mensal per capita inferior a meio salário mínimo; renda familiar per capita inferior a dois dólares/dia; disponibilidade calórica no domicílio inferior ao somatório das necessidades energéticas de todos os integrantes da família, presentes no período da pesquisa; disponibilidade calórica no domicílio inferior a 2.500 calorias per capita/dia; consumo inadequado (abaixo ou acima das recomendações) de pelo menos um dos macronutrientes (lipídios, carboidratos ou proteínas) por pelo menos um dos moradores do domicílio; e presença de distrofia nutricional (baixo peso, baixa estatura e excesso de peso) em pelo menos um dos moradores do domicílio. Calculou-se as prevalências de insegurança alimentar e nutricional, pelos distintos indicadores, e seus respectivos intervalos de confiança, bem como razão de prevalência pela regressão de Poisson. Para a construção dos modelos de predição utilizou-se regressão de Poisson, bivariada e multivariada, com variância robusta. As análises foram estratificadas por localização (macrorregião) e situação (urbana ou rural) dos domicílios. Visando fortalecer a utilização da questão de percepção sobre a suficiência de alimentos para a família, presente na POF, como um indicador de segurança alimentar e nutricional, verificou-se a sensibilidade, valor preditivo negativo e concordância das questões isoladas da Escala Brasileira de Insegurança Alimentar (EBIA) na detecção desta situação, presentes na Pesquisa Nacional por Amostra de Domicílios (PNAD), nas edições de 2009 e 2013. Este estudo foi aprovado pelo Comitê de Ética em Pesquisa da Universidade Federal de Viçosa (parecer no 1.269.063/2015). As prevalências de insegurança alimentar e nutricional, pelos 11 indicadores, confirmam a mesma como um problema de saúde pública, com maior probabilidade de ocorrência nas macrorregiões Nordeste, Norte e zona rural do país. Considerando a dimensão de acesso e disponibilidade de alimentos, verificou-se que as maiores prevalências de insegurança alimentar e nutricional, em famílias brasileiras, foram detectadas pelos indicadores de disponibilidade calórica segundo necessidade energética e percepção quanto a insuficiência de alimentos para a família. Já em relação a dimensão de utilização biológica dos alimentos, verificou-se os indicadores de consumo inadequado de lipídios ou de carboidratos e de presença de excesso de peso ou de baixa estatura, por pelo menos um dos moradores do domicílio, apresentaram as maiores prevalências de insegurança. Ao avaliar todos os indicadores, de forma conjunta nos modelos de predição, constatou-se que a insegurança alimentar e nutricional foi melhor explicada, estatisticamente, pelos modelos que apresentavam como variáveis dependentes os indicadores de renda (em salário mínimo e em dólar) e de consumo inadequado de proteínas. Ao considerar os indicadores mantidos no modelo, explicando a concomitância da insegurança pelas distintas propostas, verificou-se que os modelos de percepção, disponibilidade calórica segundo necessidade energética, de renda familiar per capita (em salário mínimo e em dólar) e consumo inadequado de carboidratos por pelo menos um dos moradores do domicílio foram explicados por pelo menos um indicador de cada vertente da insegurança alimentar e nutricional. Estes modelos podem ser considerados como completos por permitirem avaliar a insegurança por diferentes dimensões, abrangendo as vertentes de acesso, disponibilidade e utilização biológica dos alimentos. Ressalta-se que, ao avaliar as questões isoladas da EBIA, presentes nas PNAD, certificou-se que as primeiras quatro questões da escala apresentaram bons resultados de valores preditivos negativos, sensibilidade e concordância, sendo que a primeira pergunta, referente a “preocupação com que o alimento acabasse antes de ter acesso a mais alimentos” destacou-se em relação as demais. Estes resultados confirmam que a questão relativa à preocupação quanto a quantidade de alimentos pode ser considerada como proxy de insegurança alimentar em estudos populacionais. Além disso, os achados do presente estudo corroboram com a recomendação de que a (in) segurança alimentar e nutricional deve ser avaliada por indicadores que abarquem as dimensões alimentar e nutricional desta situação, de forma conjunta, contemplando as vertentes de acesso, disponibilidade e utilização biológica. A avaliação sistêmica, por modelos de predição utilizando distintos indicadores, no presente estudo, permitiu maior aproximação dos instrumentos de análise com o amplo conceito de segurança alimentar e nutricional, adotado no Brasil.
Food and nutritional security is defined as ensuring regular and permanent access to quality food in sufficient quantity without compromising access to other essential needs of individuals and families. The evaluation of food and nutritional security requires the use of different indicators, which cover the dimensions of access, availability and biological use of food, in order to contemplate its broad concept. This study aimed to evaluate food and nutritional insecurity according to indicators of perception, income, caloric availability, food consumption and nutritional status, by location and situation of households, and to construct predictive models of food and nutritional insecurity of Brazilian families, based on these indicators. Secondary data concerning the families evaluated in the Family Budget Survey (POF), carried out by the Brazilian Institute of Geography and Statistics - IBGE, in 2008-2009 were used in this study. For the evaluation of food and nutritional insecurity, the following indicators were used: perception about the insufficient amount of food consumed by the family; monthly household income per capita less than half a minimum wage; family income less than two dollars per capita day; caloric availability at home less than the sum of the energy needs of all family members present during the research period; caloric availability in the household of less than 2,500 calories per capita/day; inadequate consumption (below or above recommendations) of at least one of the macronutrients (lipids, carbohydrates or proteins) by at least one of the residents of the household; and presence of nutritional dystrophy (low weight, short stature and excess weight) in at least one of the residents of the household. The prevalence of food and nutritional insecurity, by the different indicators, and their respective confidence intervals, as well as prevalence ratio by Poisson regression were calculated. For the construction of the prediction models, we used Poisson regression, bivariate and multivariate, with robust variance. The analyzes were stratified by location (macroregion) and location (urban or rural) of households. The sensitivity, predictive value and agreement of the isolated issues of the Brazilian Scale of Food Insecurity (EBIA) were verified in order to strengthen the use of the perception question about the adequacy of food for the family, present in the POF, as an indicator of food and nutritional security in the detection of this situation, present in the National Survey of Household Samples (PNAD), in the editions of 2009 and 2013. This study was approved by the Research Ethics Committee of the Federal University of Viçosa (1,269,063/2015). The prevalence of food and nutritional insecurity, by the 11 indicators, confirms this as a public health problem, which is more likely to occur in the Northeast, North and rural macro-regions of the country. Considering the size of access and availability of food, it was verified that the highest prevalence of food and nutritional insecurity in Brazilian families were detected by the caloric availability indicators according to the need energy and perception regarding the insufficiency of food for the family. Regarding the biological food use dimension, the indicators of inadequate consumption of lipids or carbohydrates by at least one of the residents of the household and presence of overweight or short stature were verified in at least one member of the household family, presented the highest prevalence of insecurity. When evaluating all the indicators, together in the prediction models, it was verified that food and nutritional insecurity was better explained, statistically, by the models that showed as dependent variables the indicators of income and inadequate consumption of proteins. When considering the indicators maintained in the model, explaining the concomitance of the insecurity by the different proposals, it was verified that the models of perception, caloric availability according to energy need and per capita family income (in minimum wage and in dollar) were explained by at least an indicator of each food insecurity and nutritional aspect. These models can be considered complete because they allow to evaluate the insecurity by different dimensions, covering the access, availability and biological use of food. It is noteworthy that, when evaluating the isolated issues of EBIA, present in the PNAD, it was verified that the first four questions of the scale presented good results of negative predictive values, sensitivity and agreement, and the first question, that the food would end before having access to more food "stood out in relation to the others. These results confirm that the concern regarding the quantity of food can be considered as a proxy for food insecurity in population studies. In addition, the findings of the present study corroborate with the recommendation that food and nutritional security should be evaluated by indicators that cover the food and nutritional dimensions of this situation, taking into account the access, availability and use biological. The systemic evaluation, using predictive models using different indicators of the present study, allowed a closer approximation of the instruments of analysis with the broad concept of food and nutritional insecurity adopted in Brazil.
Food and nutritional security is defined as ensuring regular and permanent access to quality food in sufficient quantity without compromising access to other essential needs of individuals and families. The evaluation of food and nutritional security requires the use of different indicators, which cover the dimensions of access, availability and biological use of food, in order to contemplate its broad concept. This study aimed to evaluate food and nutritional insecurity according to indicators of perception, income, caloric availability, food consumption and nutritional status, by location and situation of households, and to construct predictive models of food and nutritional insecurity of Brazilian families, based on these indicators. Secondary data concerning the families evaluated in the Family Budget Survey (POF), carried out by the Brazilian Institute of Geography and Statistics - IBGE, in 2008-2009 were used in this study. For the evaluation of food and nutritional insecurity, the following indicators were used: perception about the insufficient amount of food consumed by the family; monthly household income per capita less than half a minimum wage; family income less than two dollars per capita day; caloric availability at home less than the sum of the energy needs of all family members present during the research period; caloric availability in the household of less than 2,500 calories per capita/day; inadequate consumption (below or above recommendations) of at least one of the macronutrients (lipids, carbohydrates or proteins) by at least one of the residents of the household; and presence of nutritional dystrophy (low weight, short stature and excess weight) in at least one of the residents of the household. The prevalence of food and nutritional insecurity, by the different indicators, and their respective confidence intervals, as well as prevalence ratio by Poisson regression were calculated. For the construction of the prediction models, we used Poisson regression, bivariate and multivariate, with robust variance. The analyzes were stratified by location (macroregion) and location (urban or rural) of households. The sensitivity, predictive value and agreement of the isolated issues of the Brazilian Scale of Food Insecurity (EBIA) were verified in order to strengthen the use of the perception question about the adequacy of food for the family, present in the POF, as an indicator of food and nutritional security in the detection of this situation, present in the National Survey of Household Samples (PNAD), in the editions of 2009 and 2013. This study was approved by the Research Ethics Committee of the Federal University of Viçosa (1,269,063/2015). The prevalence of food and nutritional insecurity, by the 11 indicators, confirms this as a public health problem, which is more likely to occur in the Northeast, North and rural macro-regions of the country. Considering the size of access and availability of food, it was verified that the highest prevalence of food and nutritional insecurity in Brazilian families were detected by the caloric availability indicators according to the need energy and perception regarding the insufficiency of food for the family. Regarding the biological food use dimension, the indicators of inadequate consumption of lipids or carbohydrates by at least one of the residents of the household and presence of overweight or short stature were verified in at least one member of the household family, presented the highest prevalence of insecurity. When evaluating all the indicators, together in the prediction models, it was verified that food and nutritional insecurity was better explained, statistically, by the models that showed as dependent variables the indicators of income and inadequate consumption of proteins. When considering the indicators maintained in the model, explaining the concomitance of the insecurity by the different proposals, it was verified that the models of perception, caloric availability according to energy need and per capita family income (in minimum wage and in dollar) were explained by at least an indicator of each food insecurity and nutritional aspect. These models can be considered complete because they allow to evaluate the insecurity by different dimensions, covering the access, availability and biological use of food. It is noteworthy that, when evaluating the isolated issues of EBIA, present in the PNAD, it was verified that the first four questions of the scale presented good results of negative predictive values, sensitivity and agreement, and the first question, that the food would end before having access to more food "stood out in relation to the others. These results confirm that the concern regarding the quantity of food can be considered as a proxy for food insecurity in population studies. In addition, the findings of the present study corroborate with the recommendation that food and nutritional security should be evaluated by indicators that cover the food and nutritional dimensions of this situation, taking into account the access, availability and use biological. The systemic evaluation, using predictive models using different indicators of the present study, allowed a closer approximation of the instruments of analysis with the broad concept of food and nutritional insecurity adopted in Brazil.
Descrição
Palavras-chave
Segurança alimentar - Brasil, Alimentos - Consumo - Pevisão, Familia - Nutrição - Brasil
Citação
MORAIS, Dayane de Castro. Modelos preditivos de (in) segurança alimentar e nutricional, segundo indicadores socioeconômicos, demográficos e nutricionais de famílias brasileiras. 2018. 227 f. Tese (Doutorado em Ciência da Nutrição) - Universidade Federal de Viçosa, Viçosa. 2018.