Avaliação dos fatores de risco individuais e familiares para desnutrição energético protéica em crianças com até 60 meses de idade, em Berilo, MG

Abstract

A case control study was conducted to know individual and familiar risk factors in malnourished children at up to 60 months of age, from 24 rural communities, in Berilo, Jequitinhonha`s Valley, state of Minas Gerais, Brazil. The cases group (n=60) was composed by children with a weight/age or height/age or weight/height index below -2 z score of the NCHS/WHO reference. The control group (n=120), at the same sex and age group, were constituted by a weight/age, height/age andweight/height index above -1 z score children. Cases and controls were compared in relation to socioeconomic, demographic, environmental, maternal reproductive, maternal child health, and morbidity factors. The mothers factors studied were the nutritional status, mother/child bond and depression. Initially, the data was analyzed by the univariate method, Odds ratios (OR) and 95% Confidence Intervals were calculated and those variables that p <0,20 were included in a model of multivariate logistic regression analysis. The Multivariate model analysistook into account a hierarchical structure of the risk factors for undernourished. The independent variables were classified in following levels: Level 1-Sociodemographic variables: family income, mother's and father's education levels, eletroelectronic domestic equipments, mothers work; Level 2 - Maternal reproductive variables and environmental variables: prenatal consultations, birth interval, local of the childbirth, number of children; Level3 - Mother's biological variables: nutritional status and stature; Level 4 - Birth weight; Level 5 - Maternal depression; Level 6 - Mother/Child Bond and Level 7-Mothers care variables andmorbidity: Immunization and Anemia. The variables selected in a certain level stayed in the subsequent levels and they were considered as risk factors for the malnutrition those with a level of significance < 0,10, to avoid the exclusion of variables potentially important. This means that some of 95% Confidence Intervals include the unit. Logistic regression analysis identified the significant risk factors associated with occurrence of undernourished, after adjusting for other variables: low family income (OR =4.0; IC: 0.9-17.4); mother's stature < 1,45m(OR =6.6; IC: 0.03-0.7); lack of immunization (OR=8.8; IC:0.7-108.5); weak mother/child bond(OR =2.2; IC:0.2-1.1); lack of prenatal consultation (OR =15.6; IC: 1.6-153.1).Then, data available can be used effectively to priority, in this study population, interventions addressed to the socioeconomic factors, maternal child care, and mother/child bond.