Objectives: To identify dietary patterns in pre-school children and to investigate its determinants.
Methods:Cross-sectional study nested in a cohort born between 2004 and 2005 in the municipality of Diamantina, state of Minas Gerais, Brazil. The instrument used to explore the dietary patterns was a frequency questionnaire for food consumption. Factor analysis was used for the identification of the dietary patterns. The study also assessed personal and socioeconomic data and anthropometric measurements of 232 preschoolers in the period between July 2009 and July 2010. The determinants of food patterns were identified by logistic regression.Results: Three patterns were obtained, which were called: "mixed diet," "snack," and "unhealthy." The children of mothers with low education are more likely (odds ratio -OR = 3.80; confidence interval 95% -95%CI = 1.90-7.60) to consume food from the "mixed diet" pattern, and less likely (OR = 0.31; 95%CI = 0.15-0.61) to consume food from the "snack" pattern, while those with higher per capita income are more likely to consume food from the "unhealthy" pattern (OR = 2.43; 95%CI = 1.13-5.24).
Conclusion:Three dietary patterns were identified in the studied population, which were determined by the level of maternal education and per capita income. Resultados: Três padrões foram obtidos, denominados de "dieta mista", "lanche" e "não saudável". Os pré-escolares filhos de mães com menor escolaridade têm mais chance (odds ratio -OR = 3,80; intervalo de confiança de 95% -IC95% = 1,90-7,60) de consumir alimentos do padrão "dieta mista" e menor chance (OR = 0,31; IC95% = 0,15-0,61) de consumir alimentos do padrão "lanches", enquanto os de maior renda per capita têm mais chance de consumir alimentos do padrão "não saudável" (OR = 2,43; IC95% = 1,13-5,24).
J Pediatr (Rio J)
Conclusão:Foram identificados três padrões alimentares na população estudada, os quais foram determinados pelo nível de escolaridade materna e renda per capita. J Pediatr (Rio J). 2012;88(2):129-36: Padrões alimentares, pré-escolares, análise fatorial, estudos transversais.