Abstract:
This study sought to identify eating patterns among children and to verify their association with maternal characteristics. We studied 1,185 children aged 13 to 35 months. Food consumption was investigated using the 24h dietary recall and the eating patterns were identified using principal components factor analysis, followed by orthogonal varimax rotation. We carried out a hierarchical modeling using poisson regression with robust variance estimates in order to estimate prevalence ratios. We identified four eating patterns: common Brazilian, dairy/porridge, healthy and unhealthy. The variance explained by these factors was of 34.3% and the common Brazilian pattern was the one that most contributed to proportional variance. We highlight the dairy/porridge pattern, in which breast milk had negative factorial loading, suggesting its substitution by cow and modified milk. The data reveal that the characterization of children’s eating patterns diverges according to maternal characteristics, such as age, educational level and number of children. Three of the four patterns we found are represented by foods rich in starch, saturated fat and simple sugar and poor in vitamins, minerals and fibers. Multiparity, lower educational levels and maternal age under 20 years were associated with a lower consumption of foods considered to be healthy and important in child development.
Keywords:
Child Nutrition; Food Consumption; Mothers; Statistical Factor Analysis