Feeding Style Patterns Are Relatively Stable Across Infancy and Are Associated with Feeding and Weight Outcomes in African American Infants (P11-120-19)

Abstract

ObjectivesInfant feeding styles (FS), feeding beliefs and practices, have been associated with diet and growth outcomes. However, FS are usually considered individually. Using latent class analysis (LCA), we test whether FS cluster, the stability of these classes over infancy, the factors predicting class assignment, and whether classes differ in infant diet and weight outcomes. MethodsData come from 428 African-American women and infants participating in the Mothers and Others Study, a home-based intervention to prevent infant obesity. LCA was used to categorize FS across five constructs (laissez-faire, pressuring, restrictive, responsive and indulgent) measured prenatally at 28-weeks and postnatally at 3 and 15 months. Multinomial regression was used to test the characteristics associated with class membership. Adjusted regression models were used to test associations with breastfeeding and infant weight-for-age z-scores (WAZ). ResultsWe identified two classes (positive and non-responsive) prenatally, an additional class (high controlling) at 3 months and a fourth class (high indulgent) at 15 months. Those in the positive group at baseline were less likely to be in any of the other less-responsive groups at 3 or 15 months. At baseline, the non-responsive group were more likely to have depressive symptoms, but less likely to be obese pre-pregnancy than the positive group. At 3 months, the high controlling group was less likely to have a college education than the positive group. At 15 months, the high controlling group was less likely to have girls or be in the treatment group, the high indulgent group was less likely to have a college education, girls, or be in the treatment group, and the non-responsive group were more likely to be low income and less likely to have a college education than the positive group. Infants with mothers in the high controlling and non-responsive group were less likely to be breastfed at 3 months. Infants with mothers in the high controlling group at 3 months had higher WAZ at 15 months. ConclusionsOur analysis documents that FS cluster. These groups persist across infancy and are associated with infant feeding and growth outcomes, suggesting that early intervention to improve FS is important. Funding SourcesNational Institute of Child Health and Human Development [R01HD073237].