Relationship between indicators of postpartum depression and risk of obesity in Puerto Rican infants enrolled in Special Supplemental Nutrition Program for Women, Infants, and Children (WIC)

Abstract

Background - Childhood obesity has increased significantly over the years, and we are currently experimenting epidemic levels. Most recent studies among group of children living in Puerto Rico revealed that 1 out of 4 children, or approximately 25% of the study population, presented obesity (Elías-Boneta et al., 2015 & Santiago H. et al., 2021). Causal studies have observed variability in the relationship between maternal depressive symptoms and children at risk of obesity (Lampard et al., 2014). Objectives - This study assesses the association between maternal depression indicators and the WFL z-score in 12-month-old infants participating in the Women, Infants, and Children (WIC) Program. Additionally, it aims to evaluate potential mediators such as maternal age, race, ethnicity, level of education, social vulnerability, infant’s physical activity, and feeding practices. Methods – This study is a secondary database analysis which obtained data from an ongoing community-based lifestyle intervention (Baby-Act Trial). The main variables of interest are infant WFL z-score at 12 months old and maternal depression indicators, measured through the EDPS questionnaire. In addition, confounding variables were evaluated to assess their mediating relationship with the principal variables. A generalized linear regression analysis was conducted to evaluate the relationship between the variables. Results – For 192 eligible participants, 4.17% of mothers showed indicators of depression and the infant WFL z-score mean at 12 months old was 1.29 ± 1.73. The unadjusted model showed that there was no significant association between maternal depression indicators and infants’ WFL z-score (p-value = 0.601, 95% CI: -1.40, 0.81). However, we found an overall reduction of 0.30 in WFL z-scores in infants of mothers experiencing possible maternal depression in comparison with non-depressed mothers. Conclusions - Maternal depression indicators were not significantly associated with infant WFL z-score during the first year of life. Moreover, socio-demographics, social vulnerability, infant feeding practices, and physical activity were not mediators in the relationship between maternal depression indicators and the WFL z-score. Sample size determination showed that the prevalence that met inclusion criteria, is below the population level estimate. This might result in a lack of association and high levels of error within the results.