In the United States, more than 8% of 2- to 5-year-olds and 17% of 6- to 11-year-olds have obesity.¹ Children with obesity are more likely to have elevated cholesterol and blood pressure levels, breathing and joint problems, and are more likely to become adults with obesity.^2,3 Multiple factors contribute to development of childhood obesity,³ and though no single factor explains all obesity, understanding which factors may play a role has important public health implications.

Early infancy, particularly early nutrition, has been identified as a critical period for preventive strategies aimed at decreasing rates of obesity.⁴ One strategy that has been identified is increasing adherence to guidelines regarding timing of the introduction of solid foods. In previous analyses of the Infant Feeding Practices Study II (IFPS II), we found that 40% of infants in our sample were introduced to solid foods before 4 months of age.⁵ Although there has been inconsistency in recent years regarding recommendations on whether complementary foods should be introduced between 4 and 6 months,^6,7 or not until 6 months,^8–10 most groups agree that introduction to solid foods before 4 months is too early.^6–8,10

A review of 26 studies concluded that introduction of solids ≤4 months is associated with obesity, particularly among formula-fed infants.¹¹ However, of the studies reviewed, the researchers concluded that few were large, good-quality studies. Additionally, some studies suggest that the association may differ for breastfed versus formula-fed infants.^11,12 Thus, our aim, using data from a cohort of US children followed longitudinally throughout the first year of life, and contacted again at 6 years, was to assess how early introduction of solid foods (before 4 months) was associated with obesity at 6 years of age, and test for an interaction with breastfeeding status, controlling for multiple potential infant, child, and maternal confounders.

The IFPS II, conducted from 2005 to 2007, followed mothers from their third trimester of pregnancy through their infant’s first year. Infants born before 35 weeks, weighing less than 5 pounds (lbs), who were a non-singleton, or with medical conditions that would affect feeding were not eligible to participate. A cross-sectional Year 6 Follow Up study (Y6FU) was conducted in 2012 when infants from the original study were approximately 6 years old. Among eligible IFPS II participants (completed the neonatal questionnaire and not subsequently disqualified from IFPS II),¹³ 1542 completed the Y6FU (response rate of 52%). Detailed methodology of Y6FU, including comparison of respondents to nonrespondents and to a nationally representative sample, are presented in a previously published methods article.¹³ Generally, participants of Y6FU were more likely to be white, higher educated, and married compared to a national sample.¹³

Overall prevalence of obesity in our sample at 6 years was 12.0% (Table 1) and varied by maternal race/ethnicity, education, PIR, prepregnancy BMI, marital status, and infant’s birth weight.

The relationship between timing of introduction of solid foods and obesity at 6 years differed significantly between those introduced before 4 months and those introduced at 4–<6 months (15.4% vs. 9.9%; p = 0.007), but not between those introduced at 4–<6 months and those introduced at 6 months or older (Fig. 1).

Odds of obesity at 6 years, by timing of introduction of solid foods during infancy, is shown in Table 2. Compared to infants who first received solid foods at 4–<6 months of age, infants introduced to solids before 4 months had increased odds of obesity at 6 years (odds ratio [OR] = 1.66; 95% confidence interval [CI], 1.15, 2.40) in unadjusted analysis; however, this association was no longer present after adjusting for covariates (OR = 1.18; 95% CI, 0.79, 1.77). Compared to infants introduced to solid foods at 4–<6 months, those introduced to solid foods at 6 months or older did not have different odds of obesity in the unadjusted or adjusted analysis. In the full model, interaction between timing of introduction of solid foods and breastfeeding duration was not significant (p = 0.2); as such, we did not included it in subsequent analyses.

Findings regarding the association between timing of introduction of solid foods and subsequent obesity have been inconsistent, with some evidence suggesting that the association between timing of solid food introduction and subsequent obesity may differ for breastfed and formula-fed infants.¹¹ One explanation for why studies may have found an association between timing of introduction of solids and subsequent obesity among formula-fed, but not breastfed, infants is that infants fed at the breast are better able to self-regulate their intake in response to solid foods being added to their diet by decreasing their consumption of human milk.¹⁶ In contrast, solid foods given to formula-fed infants may be additive, rather than substitutive, potentially because infants fed from a bottle are more dependent on the caregiver’s input, which is likely influenced by external cues, such as feeding until the bottle is empty.¹⁷

Despite this possible mechanism, in our sample, we did not find an interaction between breastfeeding status and timing of solid food introduction and subsequent obesity, nor did we find an association between timing of solid food introduction and subsequent obesity when breastfeeding status was not taken into account.

Findings of an association between early introduction to solids and childhood obesity are difficult to determine, given that when solid foods are introduced is inextricably linked to exclusive breastfeeding duration and formula feeding; therefore, residual confounding cannot be ruled out.¹¹ Inconsistent with our findings, Huh and colleagues found a 6-fold increase in odds of obesity at 3 years among those breastfed less than 4 months and introduced to solids before 4 months, but not among those breastfed longer.¹² The differences found among studies regarding the influence of breastfeeding status on timing and subsequent obesity may be linked to variations in study design, categorizations of infant feeding, and definitions of terms such as “starting solids.”¹¹

Limitations of our study include that the sample population is not nationally representative. Additionally, child obesity at 6 years of age was based on measurements of height and weight taken by the mother, which may be susceptible to measurement error. In the National Health and Nutrition Examination Survey 2011–2012, which assesses measured height and weight, prevalence of obesity was 8.4% among 2- to 5-year-olds and 17.7% among 6- to 11-year-olds, which is generally consistent with our estimate. The major strength of this analysis was the ability to control for many sociodemographic and dietary factors that have often not been controlled for in other studies. Additionally, information on timing of introduction of solid foods and breastfeeding duration were calculated from approximately monthly recalls assessed throughout infancy, limiting recall bias that may occur when asking about these practices when the child is older.

In this study, we did not find a significant association between timing of introduction of solid foods and child obesity at 6 years of age. Despite our findings, optimal infant feeding practices, including exclusive breastfeeding for approximately the first 6 months, timely introduction of complementary foods, and continued breastfeeding for at least the first year,¹⁸ should be promoted and supported for all infants because of the numerous health benefits for infants and mothers.^19–21