Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Jul 14.
Published in final edited form as: Appetite. 2021 Jun 18;166:105473. doi: 10.1016/j.appet.2021.105473

Examining patterns of postnatal feeding in relation to infant’s weight during the first year

Tiffany M Rybak a,*, Amy R Goetz b, Lori J Stark a
PMCID: PMC9280867  NIHMSID: NIHMS1814499  PMID: 34153422

Abstract

Background:

There is mixed evidence regarding specific infant feeding behaviors and later risk for overweight and obesity. We sought to detect underlying patterns in duration of breastfeeding, introduction of solid foods and sweetened beverages, in order to understand the relation to later weight.

Methods:

Patterns of postnatal feeding were examined among infants enrolled in the Infant Feeding Practices Study II (N = 3033). At monthly intervals, mothers reported on the duration of any and exclusive breastfeeding, age of solid food and sweetened beverage introduction, and reported infant weight at ages 9- and 12-months. Latent profile analysis was used to empirically derive patterns of postnatal feeding and examine associations with weight z-scores at 9 and 12 months.

Results:

Two profiles emerged: (1) Short breastfeeding duration and early introduction to solid foods and sweetened beverages (Short BF/Early Introduction; 53%) and (2) longer breastfeeding duration and later introduction to solid foods and sweetened beverages (Longer BF/Later Introduction; 43%). Infants in the Shorter BF/Early Introduction profile had significantly greater weight z-scores at 9 (M = 0.18) and 12 months (M = 0.26), compared to those in the Longer BF/Later Introduction profile (M = −0.21; M = −0.17, respectively).

Conclusions:

Overall, shorter BF duration and earlier food and sweetened beverage introduction is associated with higher weight z-scores at 9 and 12 months. Early intervention should aim to promote breastfeeding and later introduction of solids and discourage consumption of sweetened beverages. These postnatal patterns of feeding behaviors provides important context to inform targeted interventions aimed at reducing risk for later obesity.

Keywords: Infant feeding, Feeding patterns, Pediatric obesity, Breastfeeding, Formula feeding, Introduction to solid foods, Sweetened beverages

1. Introduction

Pediatric obesity is prevalent, costly, and places children and adolescents at greater risk for cardiovascular disease, Type II diabetes, musculoskeletal disorders, as well as some cancers in adulthood (Kopelman, 2007). Data suggests that the prevalence of obesity among infants <2 years of age has increased by >60% over the past 3 decades (Ogden et al., 2008). The early rise of obesity is troublesome given that overweight infants and toddlers are at greater risk for remaining overweight in childhood and adulthood. Furthermore, attempts to prevent and treat obesity with school age and older children have had limited success (Oude Luttikhuis et al., 2009). As such, infancy may be an opportune time for obesity prevention, because it is a critical period of long-lasting metabolic and behavioral consequences (Gluckman & Hanson, 2009; Paul et al., 2009; Wojcicki & Heyman, 2010). In particular, feeding choices (e.g., formula versus breastmilk) and timing (e.g., early introduction of solids) have been implicated as risk factors in the development of poor dietary habits and/or subsequent health issues such as obesity (Birch & Ventura, 2009; Pearce et al., 2013).

The American Academy of Pediatrics emphasizes that breastfeeding is preferable over formula feeding for a variety of short- and long-term health and neurodevelopmental advantages, including lower risk for obesity (Eidelman & Schanler, 2012). Additionally, exclusive breastfeeding is recommended through 6 months of age, with continued breastfeeding and the gradual introduction to solid foods taking place over the remainder of the first year (Briefel et al., 2004; Eidelman & Schanler, 2012). Studies have found that exclusive breastfeeding for 6 months, in comparison to formula feeding or mixed feeding is associated with decreased risk for obesity in childhood (Rossiter et al., 2015; Zheng et al., 2014). Complimentary feeding is defined as the transition from breast milk to the family diet at a time when the baby is both developmentally ready and when breast milk may no longer fulfill the nutritional requirements of the child (Agostoni, Decsi, Fewtrell, & et al, 2008a). Early introduction of complementary foods (i.e., prior to 6 months of age) in developed countries has been associated with gastrointestinal problems, respiratory tract infections, and increased risk for allergies (Burdette et al., 2006). Current research is mixed as to whether early introduction is associated with an increased risk for obesity, though evidence from one systematic review found that introduction prior to 4 months of age was associated with a greater risk for later overweight or obesity (Pearce et al., 2013). Finally, sugar-sweetened beverages are discouraged during the first year of life by the American Academy of Pediatrics (Heyman & Abrams, 2017). Furthermore, drinking sugar-sweetened beverages during infancy has been associated with obesity in early childhood (Pan et al., 2014).

Infant feeding is a dynamic process, involves a number of decision-points by mothers, and cannot be simply defined by one type of feeding behavior. As such, it is important to look at profiles of feeding across several feeding behaviors in relation to infant’s weight. In the current study, we sought to detect underlying patterns across several feeding behaviors (e.g., duration of any and exclusive breastfeeding, age of solid food introduction, age of sweetened beverage introduction, and age of other beverage introduction) in order to understand their combined relationship to weight throughout the first year. Examining patterns of feeding behaviors provides important context to inform targeted interventions aimed at modifiable early feeding and dietary behaviors to reduce risk for later obesity.

2. Methods

2.1. Participants and procedure

The data were obtained from the Infant Feeding Practices Study II (IFPS-II), a longitudinal, observational study conducted by the Centers for Disease Control and Prevention (CDC) and Food and Drug Administration from May 2005 to June 2007 (Fein et al., 2008a). The data are publicly available and de-identified and not considered human subjects research, thus institutional review board designated the study as exempt from review. Mothers in the IFPS-II were followed from late pregnancy through their infant’s first year of life. Eligibility included mothers who were >18 years of age and singleton infants born >35 weeks of gestation with a birth weight >2.25 kg (Fein et al., 2008b). Mother-child dyads were excluded from the study if infants had an illness or condition that would affect feeding in the first year of life. After completion of an initial prenatal questionnaire and birth screener, postnatal questionnaires were completed monthly (Neonatal, Months 2, 3, 4, 5, 6, 7) and then approximately every seven weeks (Months 9, 10.5, and 12). Response rates ranged from 63% to 83% on postnatal questionnaires. Additional information regarding methods and procedures for IFPS-II are presented elsewhere (Fein et al., 2008b).

2.2. Measures

2.2.1. Maternal & infant demographic information

Maternal information collected prenatally included age, race and ethnicity, poverty-to-income ratio, pre-pregnancy body mass index (BMI), gestational diabetes, and smoking behaviors. Gestational weight gain was assessed via self-report at the first postnatal survey. The birth screener and neonatal survey gathered information on the health of the mother and infant following delivery, gestational age (in weeks), method of delivery, infant sex, and birth weight (in grams).

2.2.2. Food Frequency Questionnaire

Our indicator variables were created from information gathered by the Food Frequency Questionnaire (FFQ), which was administered each month postnatally. Infants were coded as exclusively breastfed at each month if the FFQ indicated they received breastmilk, but no other food/formula. Duration of any breastfeeding was calculated as the estimation of infant age for the last month mothers indicated that infant received any breastmilk on the FFQ. Age solid foods were introduced was calculated based on the age of infant when mothers indicated solid food was received by infant on each month’s postnatal FFQ and included infant cereal in the bottle. Age of sweetened beverage introduction was calculated based on the age of infant when mothers indicated juice or soft drinks were consumed based on each month’s postnatal FFQ. Finally, age at which other beverage was introduced was calculated based on the age of the infant when mothers indicated cow’s milk was consumed on each postnatal FFQ.

2.2.3. Anthropometrics

Mothers were asked to report on infant weight and length, as measured at their most recent doctor’s visit, on postnatal questionnaires distributed at months 9 and 12. As detailed in other studies, there were concerns regarding validity of the parent-reported length (Li et al., 2008), therefore we opted to use weight-for-age z-scores (weight z-scores) rather than weight-for-length z-score measurements. Weight-for-age z-scores transforms weight data into standardized z-scores that takes into account an infant’s sex and age. Weight-for-age z-scores were calculated according to the U.S. Center for Disease Control and Prevention (CDC) national reference guidelines (Flegal & Cole, 2013). Biologically implausible z-scores (e.g., extreme values determined by 5 SD above or below the mean, per CDC SAS coding program) were flagged and re-coded as missing. In the IFPS-II data set, 15 values were determined to be biologically implausible at 9 months and 5 values were determined to be biologically implausible at 12 months.

2.3. Data analyses

Latent profile analyses (LPAs) were conducted in three steps using Mplus Version 8 (Muthen & Muthen, Los Angeles, CA) and missing data were addressed using full-information maximum-likelihood (Little et al., 2013) with robust standard errors due to varying response rates across time points (Muthén & Muthén, 1998). First, LPAs were conducted to empirically derive patterns (i.e., latent profiles) of postnatal feeding practices, which included any and exclusive duration of breastfeeding, age of introduction to solid food, age to introduction of sweetened drinks, and age of introduction to other beverages. The best fitting model was evaluated and selected based on goodness-of-fit indices and size of latent profiles. Models were compared on various fit statistics including Bayesian information criterion (BIC) (GJTaos, 1978), Lo-Mendell-Rubin test (LMR) (Lo et al., 2001), bootstrap likelihood ratio test (BLRT) (Geoffrey McLachlan, 2000), and entropy (Berlin et al., 2014; Nylund et al., 2007). The LMR and BLRT compare neighboring classes and determine statistically significant improvement in fit for the inclusion of one more class. Entropy values measure classification accuracy and higher values indicate greater accuracy.

Secondly, predictors of primary outcome variable (e.g., weight z-scores) were determined once the best fitting model was selected using error-corrected logistic regression (R3STEP method) (Asparouhov and Muthén, 2014a). Mothers age, pre-pregnancy BMI, and birth weight were significantly different across latent classes and therefore were included in the final model to rule out the potential confounds of profile difference and difference in other outcomes (i.e., weight z-scores) or their direct effects. The final model was conducted using BCH weights (BCH method) and Wald’s tests (Asparouhov and Muthén, 2014b). The first examined differences in weight z-scores at 9 and 12 months by profile group membership (Asparouhov and Muthén, 2014b).

3. Results

The final sample size consisted of 3033 mother-infant dyads who had complete data on all postnatal feeding variables (e.g., breastfeeding duration, age at solid food introduction and sweetened beverage introduction) for the latent profiles. Participants (n = 1869) who had missing data on postnatal feeding variables were not included in these analyses. T-tests and two-way contingency tables analyses revealed some significant demographic differences between the final sample and excluded cases. Participants in the final sample were more likely to be white (p < .001), married (p =.02), have higher levels of education (p =.007), and live in households with fewer people (p < .001) than the excluded sample. There were no significant group differences on maternal age, maternal BMI, child sex, child birthweight, child gestational age, or percentage of children breastfed for four or more weeks.

In the estimation stage, several latent profile models were fit to the data, specifying one through five latent profiles. The log-likelihood of three-through five-latent profile models did not replicate despite several attempts of increasing amounts of random starts. The information criteria, entropy, and likelihood ratio tests used to determine the best fitting model for one through five latent profiles are presented in Table 1. Fit-statistics indicated that a two-profile solution was the best fitting model (see Table 1).

Table 1.

Fit statistics for latent profile analysis.

AIC BIC Entropy LMR Test BLRT Test
One Latent Profile 83276.51 83360.75 n/a n/a n/a
Two Latent Profiles 91504.62 91606.91 0.96 p < .01 p < .01
Open in a new tab

Note. The loglikelihood was not replicated for the three and four latent profile models, suggesting that a two latent profile model was the best fitting model based on available fit statistics.

AIC = Akaike Information Criterion.

BIC = Bayesian Information Criterion.

LMR = Lo-Mendell-Rubin Test.

BLRT = Bootstrap Likelihood Ratio Tests.

The first profile (see Fig. 1) consisted of 53% of mother-infant dyads (n = 1612), and was characterized by shorter breastfeeding duration (on average 1 week exclusive and 7 weeks any breastfeeding) as well as earlier introduction to solid foods (15 weeks on average) and sweetened beverages (31 weeks on average). Hereafter, this profile is referred to as Short BF/Early Food and Drink. The second profile (see Fig. 1) comprised 47% of dyads (n = 1421) and was marked by longer breastfeeding duration (on average 13 weeks of exclusive and 48 weeks of any breastfeeding), and later introduction to solid foods (20 weeks on average) and sweetened beverages (36 weeks on average), and is hereafter referred to as the Longer BF/Later Food and Drink profile. Wald’s tests were conducted to determine that each indicator in the profile significantly differed across both groups. Introduction to other beverages (e.g., cow’s milk) did not significantly vary by profile.

Fig. 1.

Fig. 1.

Open in a new tab

Two Latent Profiles of Postnatal Feeding Practices Emerged via Latent Profile Analysis.

Note. Profiles did not significantly differ by age other beverages (e.g., cow’s milk) were introduced.

Table 2 displays descriptive information by each profile, including demographics (e.g., age, race, poverty-to-income ratio), prenatal factors (e.g., gestational weight gain, gestational diabetes, prenatal smoking, and cesarean delivery), and child factors (e.g., birth weight and sex); factors that significantly differ by profile are indicated. Mothers in the Short BF/Early Food and Drink profile were younger and more likely to be below the <185% threshold for income-to-poverty ratio compared to the Longer BF/Later Food and Drink profile. With regard to prenatal factors, mothers in the Short BF/Early Food and Drink profile reported greater prepregnancy BMI, were more likely to have gestational diabetes, engage in prenatal smoking, and deliver by cesarean compared to mothers in the Longer BF/Later Food and Drink profile, p < .05. Infants in the Short BF/Early Food and Drink profile had lower birth weight compared to those in the Longer BF/Later Food and Drink profile. There were no significant profile group differences on race/ethnicity, gestational weight gain, or infant sex.

Table 2.

Descriptive statistics characterizing each latent profile.

Shorter BF and Earlier Introduction to Food Profile (n = 1612) M (SD) or % Longer BF and Later Introduction to Food Profile (n = 1421) M (SD) or % p-value
Mother
 Age 27.97 (5.7) 29.56 (5.2) <.001
 Caucasian, % 84.9% 85.7% .60
 <185% PIR, %) 49.8% 40.1% <.001
 Pre-pregnancy BMI 27.05 (7.6) 26.18 (6.6) <.001
 Gestational weight gain, kg 13.94 (6.6) 14.27 (6.3) .27
 Gestational diabetes 7.6% 5.3% <.001
 Prenatal smoking 13.2% 3.3% <.001
 Cesarean delivery 25.1% 22.9% .027
Child
 Birth weight (g) 3415.7 (513.7) 3503.8 (462.7) <.001
 Sex, female 52.1% 51.4% .79
Open in a new tab

Note. PIR=Poverty to income ratio.

Because maternal age, pre-pregnancy BMI, and birth weight were significantly associated with weight z-scores at 9 and 12 months, they were included as covariates in the final models examining profile membership in relation to weight z-scores. Overall, infants in the Shorter BF/Early Food profile had significantly higher weight z-scores at 9 (M = 0.08, SE = 0.04) and 12 months (M = 0.31, SE = 0.04), compared to those in the Longer BF/Later Food profile (9 months: M = −0.23, SE = 0.04; 12 months: M = −0.12, SE 0.04) (see Table 3). Effect size calculations indicated that these postnatal feeding practice patterns account for a low to moderate effect on weight z-scores at 9 and 12 months (g = 0.20 and g = 0.27, respectively).

Table 3.

Differences in weight z-scores between profiles.

Shorter BF and Earlier Introduction to Food Profile M (S.E.) Longer BF and Later Introduction to Food Profile M (S.E.) p-value
Weight z-score at 9 months 0.08 (0.04) −0.23 (0.04) <.01
Weight z-score at 12 months 0.31 (0.04) −0.12 (0.04) <.01
Open in a new tab

4. Discussion

Overall, two profiles emerged that showcase different decisional points and timing of breastfeeding discontinuation and introduction of food, sweetened drinks, and other beverages (e.g., cow’s milk). The first profile consisted of just over half of our sample (53%) and was characterized by early discontinuation of breastfeeding and early introduction to solid foods and sweetened drinks, diverging from what is recommended by American Academy of Pediatrics (AAP) guidelines (Agostoni et al., 2008; Harder et al., 2005; Kramer & Kakuma, 2012). The second profile (47%) was characterized by later discontinuation of breastfeeding and later food and sweetened drink introduction. Notably, the AAP does not recommend offering sweetened drinks during the first year of life (Pan et al., 2014), yet a majority of mothers in both profiles introduced them on average about midway through the first year; therefore this may be an important area for public health intervention for all infants. The Short BF/Early Food and Drink profile had higher weight-for-age z-scores at 9 and 12 months, compared to the Longer BF/Later Food and Drink profile, despite beginning life with a relatively lower birth weight.

To our knowledge, breastfeeding duration and timing of introduction to solid foods and beverages has not been previously examined in combination. Our results suggest that earlier discontinuation of breastfeeding together with earlier introduction to solid foods and sweetened beverages are associated with a higher weight trajectory, placing infants at greater risk for obesity during childhood (Singhal & Lanigan, 2007). These findings align with previous research that has found that discontinuation of breastfeeding (Eidelman & Schanler, 2012; Hediger et al., 2001) and earlier introduction of solid foods independently contribute to greater weight gain in the first year of life (Agostoni et al., 2008), and extends that research by looking at these patterns of feeding in combination with one another.

Early intervention should aim to promote breastfeeding and later introduction of solids and sweetened beverage consumption. These postnatal feeding profiles provides important context to inform targeted interventions aimed at modifiable early feeding and dietary behaviors to reduce risk for later obesity. By recognizing that there are a variety of barriers related to early discontinuation of breastfeeding, providers can time the message and problem solve effectively. Currently, breastfeeding promotion efforts in the United States have had some success by improving initiation of breastfeeding rates through the Baby Friendly Hospital Initiative, providing education about the benefits, lactation support, and problem solving potential barriers (Gross et al., 2016; Kramer et al., 2001; Philipp et al., 2001; Pugh et al., 2002).

Given that the sample was almost evenly split between these two profiles, those that extend the length of breastfeeding by an average of even one month were associated with healthier infant weight gain on average. This finding indicates that these interventions should be universal and even extending breastfeeding by small increments (i.e., a month) may be beneficial. Currently, lactation support is provided at most delivery hospitals in the United States (Philipp et al., 2001), but mothers may need additional outpatient consultation in the first several weeks to overcome challenges and persist in their efforts to breastfeed. A meta-analysis found that breastfeeding initiation rates and duration in the United States were positively influenced by both prenatal and postnatal interventions (Chung et al., 2008). Some pediatric primary care centers in the United States provide lactation support at the newborn and one month visit, but this is not practiced widely (Chung et al., 2008; Haroon et al., 2013). Mothers who deliver by cesarean section, have a greater BMI, and experience gestational diabetes are noted to have greater difficulty with lactation (Dewey et al., 2003; Mallan et al., 2018; Turcksin et al., 2014). Therefore it has been recommended that mothers who are at higher risk for lactation difficulties and who are motivated to breastfeed, receive follow-up 72–96 h postpartum for additional lactation support after leaving the hospital (Dewey et al., 2003; Mallan et al., 2018; Turcksin et al., 2014).

The profiles that emerged in our study would indicate that early discontinuation of breastfeeding and early introduction to solid foods and sweetened beverages are linked. This finding is consistent with another study that found 70% of infants who started solid foods early (at or before 4 months) were no longer breastfeeding at 6 months (Grummer-Strawn et al., 2008). As such, continuing to emphasize delaying solid food and sweetened beverages during frequent well-baby clinic visits is important given that for 53% of our sample, solid foods and sweetened beverages were introduced earlier into the infant’s diet than American Academy of Pediatrics recommendations (Agostoni et al., 2008; Kramer & Kakuma, 2012).

This study is not without limitations. Firstly, this study is observational in nature and therefore lack of causality and possible confounding variables are important methodological concerns. Although the current study controlled for prenatal and sociodemographic characteristics, there are likely other factors that could not be adjusted for that may impact these results. Additionally, data were collected via maternal report, which may contain inaccuracies, specifically with report of weight at 9 and 12 months of age. Generalization of these findings is cautioned. Compared with women who participated in other nationally representative surveys (Fein et al., 2008c), mothers in the IFPS-II were more likely to be older, employed, and more highly educated. They were more likely to breastfeed for longer durations, to have fewer children, and were less likely to be low income or smoke. Finally, these data are more than 10 years old and although breastfeeding and early introduction to foods and beverages have been largely stable over recent years, additional contemporary analysis is needed to enhance understanding of current postnatal feeding patterns. A strength of this study was the mixed modeling approach that allowed for examination of various postnatal feeding types that allowed for distinct patterns to emerge.

In conclusion, shorter breastfeeding duration and earlier food and sweetened beverage introduction is associated with higher weight z-scores at 9 and 12 months. Early intervention should aim to promote breastfeeding and delay introduction of solid foods until approximately age 6 months, and provide education on the reasons for which sugar-sweetened beverages should be avoided until toddler age. These postnatal feeding profiles indicate patterns of feeding practices that cluster together and over half the sample had a feeding profile that was associated with higher infant weight at 9 and 12 months of age. These findings can inform targeted universally delivered interventions aimed at modifiable early feeding and dietary behaviors to reduce risk for later obesity.

Acknowledgement

This work was supported by funding from the National Institute of Diabetes and Digestive and Kidney Diseases (T32 DK063929).

Footnotes

Ethical statement

These data were collected as part of the Infant Feeding Practices Study – II conducted by the CDC. Informed consent was conducted with participants prior to data collection. This is a secondary analysis of these data and therefore the authors do not have access to the ethical review committee or institutional review board number associated with the study.

References

  1. Agostoni C, Decsi T, Fewtrell M, Goulet O, Kolacek S, Koletzko B, Michaelsen KF, Moreno L, Puntis J, Rigo J, & Shamir R (2008). Complementary feeding: a commentary by the ESPGHAN Committee on Nutrition. Journal of pediatric gastroenterology and nutrition, 46(1), 99–110. [DOI] [PubMed] [Google Scholar]
  2. Asparouhov T, & Muthén B (2014a). Auxiliary variables in mixture modeling: Three-step approaches using M plus. Structural Equation Modeling: A Multidisciplinary Journal, 21(3), 329–341. [Google Scholar]
  3. Asparouhov T, & Muthén BJMWN (2014b). Auxiliary variables in mixture modeling: Using the BCH method in Mplus to estimate a distal outcome model and an arbitrary secondary model. Structural Equation Modeling: A Multidisciplinary Journal, 21(2), 1–22.31360054 [Google Scholar]
  4. Berlin KS, Williams NA, & Parra GRJJ, P. P. (2014). An introduction to latent variable mixture modeling (part 1): Overview and cross-sectional latent class and latent profile analyses. Journal of Pediatric Psychology, 39(2), 174–187. [DOI] [PubMed] [Google Scholar]
  5. Birch LL, & Ventura A (2009). Preventing childhood obesity: What works? International Journal of Obesity, 33(S1), S74. [DOI] [PubMed] [Google Scholar]
  6. Briefel RR, Reidy K, Karwe V, & Devaney BJJ, A. D. A. (2004). Feeding infants and toddlers study: Improvements needed in meeting infant feeding recommendations. Journal of the American Dietetic Association, 104, 31–37. [DOI] [PubMed] [Google Scholar]
  7. Burdette HL, Whitaker RC, Hall WC, & Daniels SRJTA (2006). Breastfeeding, introduction of complementary foods, and adiposity at 5 y of age–. The American Journal of Clinical Nutrition, 83(3), 550–558. [DOI] [PubMed] [Google Scholar]
  8. Chung M, Raman G, Trikalinos T, Lau J, & Ip S (2008). Interventions in primary care to promote breastfeeding: An evidence review for the US preventive services task force. Annals of Internal Medicine, 149(8), 565–582. [DOI] [PubMed] [Google Scholar]
  9. Dewey KG, Nommsen-Rivers LA, Heinig MJ, & Cohen RJ (2003). Risk factors for suboptimal infant breastfeeding behavior, delayed onset of lactation, and excess neonatal weight loss. Pediatrics, 112(3), 607–619. [DOI] [PubMed] [Google Scholar]
  10. Eidelman AI, & Schanler RJ (2012). Breastfeeding and the use of human milk. Pediatrics, 129(3), e827–e841. [DOI] [PubMed] [Google Scholar]
  11. Fein SB, Grummer-Strawn LM, & Raju TNJP (2008a). Infant feeding and care practices in the United States: Results from the infant feeding practices study II. Pediatrics, 122(Supplement 2), S25–S27. [DOI] [PubMed] [Google Scholar]
  12. Fein SB, Labiner-Wolfe J, Shealy KR, Li R, Chen J, & Grummer-Strawn LMJP (2008b). Infant Feeding Practical Study II: Study Methods, 122 (Supplement 2), S28–S35. [DOI] [PubMed] [Google Scholar]
  13. Fein SB, Labiner-Wolfe J, Shealy KR, Li R, Chen J, & Grummer-Strawn LM (2008c). Infant feeding practices study II: Study methods. Pediatrics, 122(Suppl 2), S28–S35. [DOI] [PubMed] [Google Scholar]
  14. Flegal KM, & Cole TJ (2013). Construction of LMS parameters for the Centers for Disease Control and Prevention 2000 growth charts (No. 63). US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics. [PubMed] [Google Scholar]
  15. Geoffrey McLachlan D (2000). Finite Mixture models Hoboken. NJ: John Wiley & Sons, Inc. [Google Scholar]
  16. GJTaos S (1978). Estimating the dimension of a model. The Annals of Statistics, 6(2), 461–464. [Google Scholar]
  17. Gluckman P, & Hanson MJI (2009). Developmental and epigenetic pathways to obesity: An evolutionary-developmental perspective. International Journal of Obesity, 32(S7), S62. [DOI] [PubMed] [Google Scholar]
  18. Gross RS, Mendelsohn AL, Gross MB, Scheinmann R, & Messito MJ (2016). Randomized controlled trial of a primary care-based child obesity prevention intervention on infant feeding practices. The Journal of Pediatrics., 174, 171–177. e172. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Grummer-Strawn LM, Scanlon KS, & Fein SB (2008). Infant feeding and feeding transitions during the first year of life. Pediatrics, 122(Suppl 2), S36–S42. [DOI] [PubMed] [Google Scholar]
  20. Harder T, Bergmann R, Kallischnigg G, & Plagemann A (2005). Duration of breastfeeding and risk of overweight: A meta-analysis. American Journal of Epidemiology, 162(5), 397–403. [DOI] [PubMed] [Google Scholar]
  21. Haroon S, Das JK, Salam RA, Imdad A, & Bhutta ZA (2013). Breastfeeding promotion interventions and breastfeeding practices: A systematic review. BMC Public Health, 13(S3), S20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hediger ML, Overpeck MD, Kuczmarski RJ, & Ruan W (2001). Association between infant breastfeeding and overweight in young children. JAMA, 285(19), 2453–2460. [DOI] [PubMed] [Google Scholar]
  23. Heyman MB, & Abrams SA (2017). Fruit juice in infants, children, and adolescents: Current recommendations. Pediatrics, 139(6), Article e20170967. [DOI] [PubMed] [Google Scholar]
  24. Kopelman P (2007). Health risks associated with overweight and obesity. Obesity Reviews, 8, 13–17. [DOI] [PubMed] [Google Scholar]
  25. Kramer MS, Chalmers B, Hodnett ED, Sevkovskaya Z, Dzikovich I, Shapiro S, … Shishko G (2001). Promotion of breastfeeding intervention trial (PROBIT): A randomized trial in the republic of Belarus. JAMA, 285(4), 413–420. [DOI] [PubMed] [Google Scholar]
  26. Kramer MS, & Kakuma R (2012). Optimal duration of exclusive breastfeeding. Cochrane Database of Systematic Reviews., (8). [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Li R, Fein SB, & Grummer-Strawn LMJP (2008). Association of breastfeeding intensity and bottle-emptying behaviors at early infancy with Infants9 risk for excess weight at late infancy. Pediatrics, 122(Supplement 2), S77–S84. [DOI] [PubMed] [Google Scholar]
  28. Little TD, Jorgensen TD, Lang KM, & Moore EWGJJ (2013). On the joys of missing data. Journal of Pediatric Psychology, 39(2), 151–162. [DOI] [PubMed] [Google Scholar]
  29. Lo Y, Mendell NR, & Rubin DBJB (2001). Testing the number of components in a normal mixture. Biometrika, 88(3), 767–778. [Google Scholar]
  30. Mallan KM, Daniels LA, Byrne R, & Susan J (2018). Comparing barriers to breastfeeding success in the first month for non-overweight and overweight women. BMC Pregnancy and Childbirth, 18(1), 1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Muthén L, & Muthén BJM (1998). BO 1998-2012 (p. 7).
  32. Nylund KL, Asparouhov T, & Muthén BOJS (2007). Deciding on the number of classes in latent class analysis and growth mixture modeling. A Monte Carlo Simulat. study, 14(4), 535–569. [Google Scholar]
  33. Ogden CL, Carroll MD, & Flegal KMJJ (2008). High body mass index for age among US children and adolescents. JAMA, 299(20), 2401–2405, 2003–2006. [DOI] [PubMed] [Google Scholar]
  34. Oude Luttikhuis H, Baur L, Jansen H, Shrewsbury VA, O’malley C, Stolk RP, & Summerbell CD (2009). Cochrane review: Interventions for treating obesity in children. Evidence-Based Child Health: A Cochrane Review Journal, 4(4), 1571–1729. [DOI] [PubMed] [Google Scholar]
  35. Pan L, Li R, Park S, Galuska DA, Sherry B, & Freedman DS (2014). A longitudinal analysis of sugar-sweetened beverage intake in infancy and obesity at 6 years. Pediatrics, 134(Supplement 1), S29–S35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Paul IM, Bartok CJ, Downs DS, Stifter CA, Ventura AK, & Birch LL (2009). Opportunities for the primary prevention of obesity during infancy. Advances in Pediatrics, 56(1), 107–133. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Pearce J, Taylor M, & Langley-Evans SJI (2013). Timing of the introduction of complementary feeding and risk of childhood obesity: A systematic review. International Journal of Obesity, 37(10), 1295. [DOI] [PubMed] [Google Scholar]
  38. Philipp BL, Merewood A, Miller LW, Chawla N, Murphy-Smith MM, Gomes JS, … Cook JT (2001). Baby-friendly hospital initiative improves breastfeeding initiation rates in a US hospital setting. Pediatrics, 108(3), 677–681. [DOI] [PubMed] [Google Scholar]
  39. Pugh LC, Milligan RA, Frick KD, Spatz D, & Bronner Y (2002). Breastfeeding duration, costs, and benefits of a support program for low-income breastfeeding women. Birth, 29(2), 95–100. [DOI] [PubMed] [Google Scholar]
  40. Rossiter MD, Colapinto CK, Khan MK, McIsaac JLD, Williams PL, Kirk SF, & Veugelers PJ (2015). Breast, formula and combination feeding in relation to childhood obesity in Nova Scotia, Canada. Maternal and Child Health Journal., 19(9), 2048–2056. [DOI] [PubMed] [Google Scholar]
  41. Singhal A, & Lanigan J (2007). Breastfeeding, early growth and later obesity. Obesity Reviews., 8, 51–54. [DOI] [PubMed] [Google Scholar]
  42. Turcksin R, Bel S, Galjaard S, & Devlieger R (2014). Maternal obesity and breastfeeding intention, initiation, intensity and duration: A systematic review. Maternal and Child Nutrition, 10(2), 166–183. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Wojcicki JM, & Heyman MB (2010). Let’s move—childhood obesity prevention from pregnancy and infancy onward. New England Journal of Medicine, 362(16), 1457–1459. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Zheng J-S, Liu H, Li J, Chen Y, Wei C, Shen G, … Li D (2014). Exclusive breastfeeding is inversely associated with risk of childhood overweight in a large Chinese cohort. Journal of Nutrition, 144(9), 1454–1459. [DOI] [PubMed] [Google Scholar]

RESOURCES