Abstract
Video Abstract
OBJECTIVES
Breastfeeding practices may protect against offspring obesity, but this relationship is understudied among women with obesity. We describe the associations between breastfeeding practices and child BMI for age z-score (BMIz), stratified by maternal BMI.
METHODS
We analyzed 8134 dyads from 21 cohorts in the Environmental Influences on Child Health Outcomes Program. Dyads with data for maternal pre-pregnancy BMI, infant feeding practices, and ≥1 child BMI assessment between the ages of 2 and 6 years were included. The associations between breastfeeding practices and continuous child BMIz were assessed by using multivariable linear mixed models.
RESULTS
Maternal pre-pregnancy BMI category prevalence was underweight: 2.5%, healthy weight: 45.8%, overweight: 26.0%, and obese: 25.6%. Median child ages at the cessation of any breastfeeding and exclusive breastfeeding across the 4 BMI categories were 19, 26, 24, and 17 weeks and 12, 20, 17, and 12 weeks, respectively. Results were in the hypothesized directions for BMI categories. Three months of any breastfeeding was associated with a lower BMIz among children whose mothers were a healthy weight (−0.02 [−0.04 to 0.001], P = .06), overweight (−0.04 [−0.07 to −0.004], P = .03), or obese (−0.04 [−0.07 to −0.006], P = .02). Three months of exclusive breastfeeding was associated with a lower BMIz among children whose mothers were a healthy weight (−0.06 [−0.10 to −0.02], P = .002), overweight (−0.05 [−0.10 to 0.005], P = .07), or obese (−0.08 [−0.12 to −0.03], P = .001).
CONCLUSIONS
Human milk exposure, regardless of maternal BMI category, was associated with a lower child BMIz in the Environmental Influences on Child Health Outcomes cohorts, supporting breastfeeding recommendations as a potential strategy for decreasing the risk of offspring obesity.
What’s Known on This Subject:
Childhood obesity is increasing in prevalence worldwide. Children with obesity are likely to have obesity as adults, increasing lifetime risk of chronic health conditions. Exclusive breastfeeding may protect against offspring obesity, but this relationship is understudied among women with obesity.
What This Study Adds:
Human milk exposure, regardless of maternal pre-pregnancy BMI category, was associated with lower child BMI for age z-scores in the ECHO cohorts, supporting breastfeeding recommendations as a potential strategy for decreasing the risk of offspring obesity.
For US children aged 2 to 19 years in 2020, the obesity prevalence was 19.7%, impacting 14.7 million individuals.1,2 Children with obesity are likely to have obesity as adults, increasing their risk of chronic conditions, including type 2 diabetes, cardiovascular disease, and cancer.3,4 The identified predictive factors of childhood obesity include genetics and environmental exposures (eg, diet, parent and family behaviors, and the built, natural, and social environment) during the antenatal and postnatal periods.4,5
Exclusive breastfeeding is recommended for 6 months,6 in part because breastfeeding is widely acknowledged to protect against obesity and many chronic conditions over the life course.6,7 Nonetheless, women with obesity are less likely to initiate breastfeeding and are more prone to early cessation compared with women of a healthy weight,8–10 making it difficult to assess the associations between breastfeeding behavior and childhood obesity in this group. Moreover, some reports imply a deficiency in milk quality of obese women, potentially influencing their motivation and support to breastfeed.11
The rising prevalence of obesity raises public health concerns that fewer infants will be breastfed, and those who are breastfed will be breastfed for a shorter duration. Thus, it is important to address the impact of maternal obesity on breastfeeding behaviors, as well as on child health outcomes. The authors of few studies have had sufficient sample sizes with which to explore the relationship among maternal obesity, breastfeeding, and child obesity, making this analysis of >8000 maternal–child dyads from 21 pediatric cohorts an important scientific contribution. Our objectives were to describe the associations between breastfeeding practices and child BMI for age z-score (BMIz), according to maternal pre-pregnancy BMI category. We hypothesized that a longer duration of exclusive breastfeeding would be associated with a lower risk of childhood obesity, even among children born to women with pre-pregnancy obesity.
Methods
Study Participants
The Environmental Influences on Child Health Outcomes (ECHO) Program includes 69 pediatric cohorts designed to evaluate the effects of early life environmental exposures on 5 priority outcome areas, including childhood obesity, described in detail elsewhere.12,13 The inclusion criteria for this analysis were data available on pre-pregnancy maternal BMI, infant feeding practices, and at least 1 child BMI measurement between the ages of 2 and 6 years.14 The exclusion criteria, selected on the basis of their potential to impact infant feeding practices, were nonsingleton births and gestational age at birth <35 weeks, birth weight <2500 g (low birth weight), chronic medical conditions, congenital birth defects, the need for resuscitation or surfactant replacement therapy after delivery, NICU stay, neurologic conditions, neonatal opioid withdrawal syndrome, neonatal sepsis, sickle cell anemia, or significant birth injury. Our analytic sample consisted of 12 997 BMI measurements from 8134 mother–child pairs participating in 21 ECHO pediatric cohorts (Figs 1 & 2). All participants consented in accordance with protocols approved by local institutional review boards and ECHO program protocols.
FIGURE 1.
Participant locations by state across 21 ECHO cohorts.
FIGURE 2.
Flowchart for final analytic sample (n = 8134).
Measures
Infant Feeding
Extant breastfeeding data were harmonized across ECHO cohorts and collected as part of the ECHO-wide data collection protocol. Because we include data collected before the adoption of the standardized ECHO-wide cohort protocol, cohorts collected breastfeeding status and duration in multiple ways, and those data were harmonized into the summary variables used in this analysis (see Online Supplement for harmonization details). We examined 2 binary breastfeeding variables related to the index child: (1) ever breastfed (yes/no), and (2) exclusive breastfeeding at 3 months (yes/no). The continuous breastfeeding measures included the duration of any breastfeeding (allowing missing data on formula or other food, irrespective of supplements) and the duration of exclusive breastfeeding (defined as no formula feeding or other food, irrespective of supplements, allowing missing data on food and supplements). Both variables assume that breastfeeding was initiated at birth. Because of differences in cohort ascertainment of infant feeding practices, each breastfeeding variable had a different sample size. All breastfeeding variables were treated as time-fixed.
Child BMI
The sources of child height and weight included those taken at study visits (69.8%), medical records (ie, measured at a clinic or well-child visits; 24.9%), and mother or caregiver report (5.3%). Height and weight data were used to calculate BMI and age- and sex-standardized child BMI z-scores by using the 2000 Centers for Disease Control and Prevention US Growth Charts.15 The outcome of interest was continuous child BMI for age z-score (BMIz), which was time-varying.15 We used all biologically plausible BMI z-score measurements, as defined by the Centers for Disease Control and Prevention,15 available for each child between the ages of 2 and 6 years.
Maternal Pre-Pregnancy BMI
Maternal height and weight were assessed by self-report, study visit, or medical record and used to calculate BMI. Maternal pre-pregnancy BMI categories were underweight (BMI <18.5 kg/m2), healthy weight (18.5 ≤ BMI <25 kg/m2), overweight (25 ≤ BMI <30 kg/m2), and obese (BMI ≥30 kg/m2; Supplemental Fig 4). The maternal pre-pregnancy BMI that was assessed from 12 months before conception through the end of the first trimester (<13 weeks gestational age) was used.
Covariates
The inclusion of covariates was based on their hypothesized associations with breastfeeding and child BMI. Covariates included the sex of the child (male, female), the highest level of education attained by the mother or caregiver (less than high school, high school degree or equivalent, some college or Associate’s degree, Bachelor’s degree or higher), maternal age at delivery (in years), the use of nicotine/tobacco products during pregnancy (yes/no), the age of the child at the time of BMI measurement, maternal race (Black, White, other), and maternal ethnicity (Hispanic, non-Hispanic). Maternal race and ethnicity were included in our models as proxies for structural barriers to breastfeeding and cultural factors that might influence breastfeeding practices and child diet. When race information was missing for the mother, we substituted information about the race of the child, when available. When maternal education information was not available during pregnancy, we substituted information about the highest level of education reported for the mother or other caregiver during the child’s infancy or at any point before age 6 years. For sociodemographic variables still missing (Table 1), we assumed a missing at-random mechanism and performed multivariate imputation with chained equations >35 imputations each with 20 iterations. Imputation models included child BMI z-score, ever breastfed, maternal race, maternal ethnicity, mother or caregiver highest education level, maternal age at delivery, parity, child’s year of birth, gestational diabetes status, delivery mode, the use of nicotine/tobacco products during pregnancy, the age of the child, gestational weight gain,16 and maternal BMI category. All covariates were treated as time-fixed.
TABLE 1.
Sociodemographic Characteristics of Mother–Child Pairs (n = 8134)
| Overall, n = 8134 | Underweight, 207 (2.5%) | Healthy Weight, 3727 (45.8%) | Overweight, 2115 (26.0%) | Obese, 2085 (25.6%) | |
|---|---|---|---|---|---|
| Socio-demographic characteristics of mothers | |||||
| Maternal race, n (%) | |||||
| Black | 1440 (17.7%) | 42 (20.3%) | 454 (12.2%) | 350 (16.5%) | 594 (28.5%) |
| White | 4749 (58.3%) | 105 (50.7%) | 2363 (63.4%) | 1241 (58.7%) | 1040 (49.9%) |
| Other* | 1453 (17.9%) | 53 (25.6%) | 719 (19.3%) | 368 (17.4%) | 313 (15.0%) |
| Missing/unknown | 492 (6.1%) | 7 (3.4%) | 191 (5.1%) | 156 (7.4%) | 138 (6.6%) |
| Maternal Hispanic ethnicity | |||||
| Hispanic | 1743 (21.4%) | 41 (19.8%) | 652 (17.5%) | 533 (25.2%) | 517 (24.8%) |
| Non-Hispanic | 5805 (71.4%) | 145 (70.1%) | 2815 (75.5%) | 1437 (67.9%) | 1408 (67.5%) |
| Missing/unknown | 586 (7.20%) | 21 (10.1%) | 260 (7.0%) | 145 (6.9%) | 160 (7.7%) |
| Maternal age at delivery, y | |||||
| Median (25th, 75th percentile) | 30 (26, 34) | 27 (23, 32) | 31 (26, 34) | 30 (26, 34) | 29 (25, 33) |
| Maternal education, n (%) | |||||
| Less than high school | 361 (4.4%) | 13 (6.3%) | 146 (3.9%) | 93 (4.4%) | 109 (5.2%) |
| Some high school or high school | 710 (8.7%) | 21 (10.1%) | 257 (6.9%) | 192 (9.1%) | 240 (11.5%) |
| Some college or technical school | 1481 (18.2%) | 36 (17.4%) | 500 (13.4%) | 401 (19.0%) | 544 (26.1%) |
| Bachelor’s degree or higher | 2752 (33.8%) | 48 (23.2%) | 1534 (41.2%) | 719 (34.0%) | 451 (21.6%) |
| Missing/unknown | 2830 (34.8%) | 89 (43.0%) | 1290 (34.6%) | 710 (33.5%) | 741 (35.5%) |
| Marital status, n (%) | |||||
| Married/currently living with a partner | 3690 (45.4%) | 73 (35.3%) | 1813 (48.7%) | 997 (47.1%) | 807 (38.7%) |
| Previously married | 191 (2.4%) | 7 (3.4%) | 79 (2.1%) | 44 (2.1%) | 61 (2.9%) |
| Single, never married | 1096 (13.4%) | 31 (14.9%) | 398 (10.7%) | 272 (12.9%) | 395 (18.9%) |
| Missing/unknown | 3157 (38.8%) | 96 (46.4%) | 1437 (38.5%) | 802 (37.9%) | 822 (39.4%) |
| Maternal anthropometry, behavioral, delivery, and metabolic characteristics | |||||
| Pre-pregnancy BMI (kg/m2), (median [25th, 75th percentile]) | 25.25 (22.1 to 30.1) | 17.9 (17.2 to 18.2) | 22.2 (20.8 to 23.5) | 27.1 (26 to 28.4) | 34.5 (32 to 38.8) |
| Parity (categories) | |||||
| Nulliparous (0) | 2274 (27.9%) | 72 (34.8%) | 1207 (32.4%) | 547 (25.9%) | 448 (21.5%) |
| Parous: 1 | 1744 (21.4%) | 34 (16.4%) | 797 (21.4%) | 497 (23.5%) | 416 (19.9%) |
| Parous: ≥2 | 1062 (13.1%) | 23 (11.1%) | 397 (10.6%) | 284 (13.4%) | 358 (17.2%) |
| Missing | 3054 (37.6%) | 78 (37.7%) | 1326 (35.6%) | 787 (37.2%) | 863 (41.4%) |
| Wt change during pregnancy (according to 2009 Institute of Medicine recommendation) | |||||
| Gained less than the recommendation | 1467 (18.0%) | 59 (28.5%) | 833 (22.3%) | 239 (11.3%) | 336 (16.1%) |
| Met the recommendation | 1794 (22.1%) | 69 (33.3%) | 934 (25.1%) | 395 (18.7%) | 396 (19.0%) |
| Gained more than recommendation | 3459 (42.5%) | 47 (22.7%) | 1241 (33.3%) | 1141 (53.9%) | 1030 (49.4%) |
| Missing/unknown | 1414 (17.4%) | 32 (15.5%) | 719 (19.3%) | 340 (16.1%) | 323 (15.5%) |
| Gestational diabetes, n (%) | |||||
| No | 5151 (63.3%) | 133 (64.2%) | 2490 (66.8%) | 1348 (63.7%) | 1180 (56.6%) |
| Yes | 394 (4.8%) | 7 (3.4%) | 121 (3.3%) | 109 (5.2%) | 157 (7.5%) |
| Missing/unknown | 2589 (31.8%) | 67 (32.4%) | 1116 (29.9%) | 658 (31.1%) | 748 (35.9%) |
| Gestational hypertension, n (%) | |||||
| No | 4557 (56.0%) | 119 (57.5%) | 2176 (58.4%) | 1207 (57.1%) | 1055 (50.6%) |
| Yes | 256 (3.2%) | <5 (<2.0%)** | >55 (>1.5%)** | 72 (3.4%) | 121 (5.8%) |
| Missing/unknown | 3321 (40.8%) | >80 (>40.0%)** | <1495 (<42.0%)** | 836 (39.5%) | 909 (43.6%) |
| Use of tobacco products during pregnancy, n (%) | |||||
| No | 7263 (89.3%) | 176 (85.0%) | 3390 (90.9%) | 1881 (88.9%) | 1816 (87.1%) |
| Yes | 663 (8.2%) | >25 (>10.0%)** | 245 (6.6%) | 174 (8.2%) | >210 (>8.0%)** |
| Missing/unknown/declined to answer | 208 (2.5%) | <5 (<5.0%)** | 92 (2.5%) | 60 (2.8%) | <55 (<5.0%)** |
| Mode of delivery, n (%) | |||||
| Vaginal | 4932 (60.6%) | 150 (72.5%) | 2375 (63.7%) | 1298 (61.4%) | 1109 (53.2%) |
| Cesarean | 1237 (15.2%) | 17 (8.2%) | 420 (11.3%) | 323 (15.3%) | 477 (22.9%) |
| Missing/inconsistent | 1965 (24.2%) | 40 (19.3%) | 932 (25.0%) | 494 (23.3%) | 499 (23.9%) |
| Child characteristics | |||||
| Child female sex, n (%) | 3742 (46.0%) | 88 (42.5%) | 1707 (45.8%) | 976 (46.2%) | 971 (46.6%) |
| Total anthropometric measurements per child | |||||
| Median (25th, 75th percentile) | 1 (1 to 2) | 1 (1 to 2) | 1 (1 to 2) | 1 (1 to 2) | 1 (1 to 2) |
| Range | 1 to 17 | 1 to 9 | 1 to 10 | 1 to 17 | 1 to 15 |
| Child BMI z-scores (across all measurements) | |||||
| Median (25th, 75th percentile) | 0.37 (−0.41 to 1.12) | −0.14 (−0.76 to 0.70) | 0.16 (−0.60 to 0.88) | 0.45 (−0.32 to 1.16) | 0.69 (−0.07 to 1.49) |
| Range | −5.38 to 5.18 | −4.63 to 4.43 | −5.38 to 4.98 | −5.26 to 4.74 | −5.17 to 5.18 |
| Children’s age at time of anthropometric measurements, y | |||||
| Median (25th, 75th percentile) | 3.26 (2.48 to 4.57) | 3.43 (2.68 to 4.56) | 3.43 (2.55 to 4.61) | 3.20 (2.43 to 4.55) | 3.16 (2.43 to 4.54) |
| Range | 2.00 to 5.99 | 2.00 to 5.98 | 2.00 to 5.99 | 2.00 to 5.99 | 2.00 to 5.99 |
| Children’s gestational age at birth, wks | |||||
| Median (25th, 75th percentile) | 39 (38 to 40) | 39 (38 to 40) | 39 (39 to 40) | 39 (38 to 40) | 39 (38 to 40) |
| Range | 36 to 43 | 36 to 42 | 36 to 43 | 36 to 43 | 36 to 43 |
| Children’s birth wt, g | |||||
| Median (25th, 75th percentile) | 3410 (3124 to 3727) | 3250 (2945 to 3560) | 3390 (3100 to 3689) | 3445 (3150 to 3774) | 3470 (3190 to 3800) |
| Range | 2500 to 5528 | 2505 to 4710 | 2500 to 5400 | 2504 to 5273 | 2500 to 5528 |
Among the 1453 women in the other race category, 629 (43.3%) identified as Asian, whereas 459 (31.6%) identified multiple races.
In accordance with ECHO’s publication and data use policy, symbols < or > are used to display numbers in which there exists a cell size >0 but <5, and there is a potential risk of reidentifying participants. The cell with the small size and a few surrounding cells are sufficiently suppressed to prevent back calculation of the exact numbers in the cells with the small size.
Statistical Analysis
We examined the univariate statistics of sociodemographic and breastfeeding practices, as well as maternal anthropometric, behavioral, metabolic, and delivery characteristics overall and across maternal pre-pregnancy BMI categories. To evaluate the associations between breastfeeding practices and continuous child BMIz, we used multivariable linear mixed models to estimate the difference in continuous BMI z-scores per increase in breastfeeding predictor. We adjusted for covariates and included random intercepts for the cohort and child to account for the hierarchical nature of the data-clustering within the cohort and repeated measures of BMI.17 We specified an unstructured covariance matrix. All analyses were stratified by maternal pre-pregnancy BMI category. To aid interpretability, continuous breastfeeding predictors were scaled to 4 weeks to represent the results for every additional month of breastfeeding. To obtain P values for the interaction between breastfeeding and maternal pre-pregnancy BMI, we specified separate models, including the cross-product of each breastfeeding variable and maternal pre-pregnancy BMI category, excluding the maternal underweight category because of issues with model instability and nonconvergence resulting from a limited sample size. We also specified separate models, including the cross-product of each breastfeeding variable and continuous child age, to obtain P values for the interaction between breastfeeding and child age at the time of BMI measurement. We conducted sensitivity analyses, including adjustment for pregnancy weight gain category,16 gestational diabetes, and mode of delivery, random slopes for child age at the time of BMI measurement, complete case analyses, limiting outcomes to only those measured at facilities or study visits, and controlling for paternal height and weight. Analyses were performed in Stata version 17. Statistical significance was 2-sided and defined at α level 0.05.
Results
Sociodemographic characteristics of the 8134 mother–child pairs are shown in Table 1. Before the index pregnancy, 2.5% of the mothers were underweight, 45.8% were healthy weight, 26.0% were overweight, and 25.6% were obese. Overall, 17.7% of the women identified as Black, 58.4% identified as white, and 17.9% were classified as other race. Among the 1453 women in the other race category, 629 (43.3%) identified as Asian, whereas 459 (31.6%) identified multiple races. Overall, 21.4% of mothers identified as Hispanic. The median age at the time of children’s BMI measure was 3.26 years (interquartile range, 2.48–4.57), and a total of 12 997 total BMI measurements were available for the longitudinal analysis. Children had an average of 1.6 BMI measurements.
Most children (90.6%) were “ever breastfed,” but this varied by maternal pre-pregnancy weight category (underweight = 90.3%; healthy weight = 92.2%; overweight = 91.9%; obese = 86.3%; Table 2). The median child age at which breastfeeding was stopped was lowest among those in the obese category (17.3 weeks compared with underweight = 18.8 weeks; overweight = 24 weeks; healthy weight = 26 weeks). Similarly, the median duration of exclusive breastfeeding varied by weight category (underweight = 12 weeks; healthy weight = 20 weeks; overweight = 17.3 weeks; obese = 12 weeks). The fraction who exclusively breastfed for at least 3 months also varied across maternal weight.
TABLE 2.
Description of Infant Feeding Practices
| Breastfeeding Variables | Overall, n = 8134 | Underweight, n = 207 | Healthy Weight, n = 3727 | Overweight, n = 2115 | Obese, n = 2085 |
|---|---|---|---|---|---|
| Ever breastfed the child | |||||
| No | 768 (9.4%) | 20 (9.7%) | 291 (7.8%) | 172 (8.1%) | 285 (13.7%) |
| Yes | 7366 (90.6%) | 187 (90.3%) | 3436 (92.2%) | 1943 (91.9%) | 1800 (86.3%) |
| Duration of any breastfeeding, wks | |||||
| n | 3652 | 102 | 1707 | 959 | 884 |
| Median (25th, 75th percentile) | 24 (8 to 52) | 18.75 (6 to 48) | 26 (12 to 52) | 24 (8 to 52) | 17.3 (5.0 to 39) |
| Range, wks | 0 to 282 | 0 to 130 | 0 to 208 | 0 to 282 | 0 to 260 |
| Duration of exclusive breastfeeding (confirmed no formula), wks* | |||||
| n | 1990 | 51 | 902 | 512 | 525 |
| Median (25th, 75th percentile) | 17.3 (4 to 26) | 12 (3 to 24) | 20 (8 to 26) | 17.3 (4 to 28) | 12 (2 to 26) |
| Range, wks | 0 to 208 | 0 to 104 | 0 to 208 | 0 to 156 | 0 to 186 |
| Exclusive breast feeding through 3 mo of age* | |||||
| No | 1413 (17.4%) | 45 (21.7%) | 535 (14.4%) | 375 (17.7%) | 458 (21.9%) |
| Yes | 878 (10.8%) | 13 (6.3%) | 454 (12.2%) | 232 (11.0%) | 179 (8.6%) |
| Missing/unknown | 5843 (71.8%) | 149 (72.0%) | 2738 (73.4%) | 1508 (71.3%) | 1448 (69.5%) |
We define “exclusive breastfeeding” at 3 months as a report of breastfeeding and confirmation of no formula feeding. Supplement use (eg, fluoride, iron, vitamin D, multivitamins, fish oil, DHA, EPA, prebiotics, probiotics, other vitamins, human milk fortifier) is included, and information on supplement use and complementary or solid food intake may be missing. Maternal pre-pregnancy BMI categories were underweight (BMI <18.5 kg/m2), healthy weight (18.5 ≤ BMI <25 kg/m2), overweight (25 ≤ BMI <30 kg/m2) and obese (BMI ≥30 kg/m2).
Adjusted and unadjusted main results of the associations between breastfeeding practices and child BMIz are shown in Fig 3 (with results for continuous breastfeeding exposures further scaled to 3 months to aid interpretability). Overall, the adjusted and unadjusted results are in the same direction, although the adjusted results were attenuated. Ever breastfeeding (vs never breastfeeding) was not statistically significantly associated with child BMIz. Exclusive breastfeeding at 3 months (vs not) was associated with a lower child BMIz among children whose mothers were a healthy weight −0.06 [−0.10 to −0.02], P = .002), overweight (−0.05 [−0.10 to 0.005], P = .07), or obese (−0.08 [−0.12 to −0.03], P = .001). Three months’ duration of any breastfeeding was associated with a lower BMIz among children whose mothers were a healthy weight (−0.02 [−0.04 to 0.001], P = .06), overweight (−0.04 [−0.07 to −0.004], P = .03), or obese (−0.04 [−0.07 to −0.006], P = .02). Three months’ duration of exclusive breastfeeding was associated with a lower BMIz among children whose mothers were a healthy weight (−0.06 [−0.10 to −0.02], P = .002), overweight (−0.05 [−0.10 to 0.005], P = .07), or obese (−0.08 [−0.12 to −0.03], P = .001). Supplemental Table 3 (unadjusted) and Supplemental Table 4 (adjusted) reveal these results with the continuous breastfeeding exposures in 1-month increments. Adjusting for birth weight and parity did not substantially change the magnitude and direction of observed associations (not shown).
FIGURE 3.
(A) Associations between binary breastfeeding variables and child BMI z-scores (BMIz) according to maternal pre-pregnancy BMI categories, (B) associations between continuous breastfeeding variables (scaled to 3 months) and child BMIz according to maternal pre-pregnancy BMI categories.
A*, adjusted; BF, breastfeeding; ex., exclusive; U*, unadjusted. Estimates in the figures reflect adjustment for maternal race, maternal Hispanic ethnicity, maternal/caregiver highest education level, maternal age at delivery, use of nicotine/tobacco during pregnancy, sex of the child, child current age, and include random intercepts for the cohort and child. In addition, each 6 months of any breastfeeding was associated with lower BMIz among children whose mothers were healthy weight (−0.04 [−0.08 to 0.002], P = .06), overweight (−0.06 [−0.12 to −0.007], P = .03); or obese (−0.07 [−0.13 to −0.01], P = .02). Each 6 months of exclusive breastfeeding was associated with lower BMIz among children whose mothers were healthy weight (−0.11 [−0.18 to −0.04], P = .002), overweight (−0.09 [−0.19 to 0.009], P = .07), or obese (−0.15 [−0.24 to −0.06], P = .001).
The interaction terms between breastfeeding and maternal pregnancy BMI were not statistically significant for ever breastfeeding (interaction term = -0.01, P = .82), exclusive breastfeeding at 3 months (interaction term = 0.01, P = .90), the duration of any breastfeeding (interaction term = −0.002, P = .42), or the duration of exclusive breastfeeding (interaction term = −0.003, P = .51). The interaction terms between continuous child age at BMI measurement and breastfeeding were statistically significant for exclusive breastfeeding at 3 months (interaction term = 0.06, P = .01), the duration of any breastfeeding (interaction term = 0.004, P = .001), and the duration of exclusive breastfeeding (interaction term = 0.005, P = .02) but not for ever breastfeeding (interaction term = −0.01, P = .81). The interaction results suggest that from 2 to 6 years of age, there is a statistically significant divergence in the rates of change in BMIz between those children exposed to each specific breastfeeding practice compared with those who were not exposed to that specific breastfeeding practice. Although children exposed to each respective breastfeeding practice have a lower BMIz between the ages of 2 and 6 years compared with those who were not exposed to that specific breastfeeding practice, the slope of change in BMIz over time is slightly more positive for children exposed to the breastfeeding practices. Thus, there may be a future point in time at which breastfeeding practices are no longer associated with reduced child BMIz. However, our data are limited to children <6 years of age, at which this limit has not yet been met.
Sensitivity analyses in which we adjusted for the category of gestational weight gain,16 gestational diabetes, and mode of delivery produced similar results to adjusted models (Supplemental Table 5). Likewise, including random slopes for child age did not substantially change the results (Supplemental Table 6). The results of a complete case analysis were slightly stronger compared with the imputed results (Supplemental Table 7). Limiting to BMI z-scores obtained from height and weight measurements obtained from medical records or measured at a study visit by cohort staff did not substantially change the results (Supplemental Table 8). The results of continuous breastfeeding variables scaled to 24 weeks (∼6 months) are provided in Supplemental Table 9. Even after adjustment for paternal height and weight, the associations of the continuous breastfeeding variables and child BMIz were statistically significant among women with a pre-pregnancy healthy weight and obesity (Supplemental Table 10). Although not statistically significant, maternal caregivers with the highest educational level (bachelor’s degree or higher vs less than high school) had a lower child BMI z-score, and this difference was comparable to that of exclusive breastfeeding at 3 months (Supplemental Tables 11 & 12).
Discussion
In ECHO pediatric cohorts, the results suggest a protective association between breastfeeding and childhood obesity regardless of maternal pre-pregnancy BMI category. Across most breastfeeding exposures, the associations were stronger among children with mothers who had obesity at pre-pregnancy compared with those whose mothers were categorized as overweight at pre-pregnancy; similar results were observed when comparing children of mothers with pre-pregnancy obesity to mothers with a healthy weight across all breastfeeding practices, except for exclusive breastfeeding at 3 months. Lower BMI z-scores were observed for ever breastfed children (vs never breastfed children), but the associations were not significant. Exclusive breastfeeding at 3 months was associated with a lower child BMIz but only significant among women with a healthy pre-pregnancy BMI. Each additional month of any or exclusive breastfeeding was associated with a significantly lower child BMIz among mothers who were overweight (for any breastfeeding) or obese (for any and exclusive breastfeeding). Our results suggest that 6 months of exclusive breastfeeding would lower the mean BMI by 0.15 SD among children born to obese women (Supplemental Table 9). Nationally, this could affect the children of the 4 million women of childbearing age with obesity.18 More than 35.2% of measurements available for children born to women with pre-pregnancy obesity in our sample indicated overweight or obesity compared with 27.6% and 19.9% among children of overweight and healthy weight mothers, in line with the higher incidence of obesity among children of women with pre-pregnancy obesity.19,20
Findings are corroborated by the authors of several reviews reporting exposure to breastfeeding as being protective against childhood obesity.4,21–24 When comparing ever to never breastfeeding on the risk of child obesity, stronger protective associations for a breastfeeding duration of at least 6 months were found in the World Health Organization European child obesity surveillance study from 22 countries.21 Similarly, the 2020 Dietary Guidelines Advisory Committee concluded that there is moderate evidence of a reduced risk of overweight and obesity at age 2 or older with a breastfeeding duration of at least 6 months.22 However, results are in contrast to the only randomized controlled trial (RCT) of breastfeeding, a study that led to a prevalence of 43% exclusive breastfeeding at 3 months in the experimental arm versus 6% in the control arm (ie, PROBIT, a cluster-RCT of breastfeeding promotion in Belarus in 1996 and 1997),25 which did not reveal differences between trial arms in the measures of childhood adiposity at child age 626,27 or 1628 years. A 2021 systematic review of breastfeeding and risk of overweight in childhood that included the PROBIT RCT and 6 cohorts with sibling-pair analyses (to rule out unmeasured confounding) revealed evidence to suggest that breastfeeding was associated with a lower risk of overweight and obesity, more strongly with longer breastfeeding duration.29
The authors of one report conducted in a single ECHO cohort examined breastfeeding and child percentage body fat in children aged 4 to 8 years.24 The findings indicated that the percentage of body fat among children was inversely associated with breastfeeding duration. Our data are unique because the large sample size from varied cohort populations allowed us to stratify by maternal pre-pregnancy weight status and assess the associations between breastfeeding and childhood obesity, specifically for infants born to mothers with pre-pregnancy obesity. Mothers with pre-pregnancy obesity are more likely to experience pregnancy and delivery complications (eg, gestational diabetes mellitus, hypertension, preeclampsia, cesarean delivery) influencing breastfeeding intentions and resulting in reduced breastfeeding incidence and duration.10,30 In addition, physiologic barriers (eg, delayed lactogenesis, breast engorgement, breast size) are related to breastfeeding difficulties among this group.10,30 In our results, a protective association was observed among women who were overweight and obese before pregnancy, even after adjusting for known barriers that increase cessation among this group of women. Maternal or caregiver education (bachelor’s degree or higher vs less than high school) had similar, although not statistically significant, effect estimates to exclusive breastfeeding at 3 months. The Hashemi-Nazari et al systematic review and meta-analysis indicates a significant linear association between maternal BMI and the risk of early breastfeeding cessation.10 As the prevalence of maternal obesity increases, it appears less likely that the majority of US women will initiate and sustain breastfeeding. Women who are obese should receive additional focused breastfeeding education addressing breastfeeding-related problems and receive assistance from health care professionals, given their low likelihood of breastfeeding success when such supports are not in place.10,31 Although there is evidence of individual variation in breastmilk composition, the potential mechanisms and role of these factors in the transmission of obesity risk from mother to child are still to be understood.32 The limited literature and inconsistent results consequently present harmful and discouraging messaging portraying the breastmilk of obese women to be “inferior” to that of women without obesity. Recent studies have suggested that higher concentrations of “bad” fats, various inflammatory markers, or differences in breastmilk composition are associated with accelerated weight gain in infancy.11,33,34 This study provides evidence and an opportunity for health professionals to protectwomen who want to breastfeed, especially women who are obese, from discouraging messaging. Health care professionals should encourage and promote breastfeeding among these women, despite known barriers because it may reduce the burden of childhood obesity because early life BMIz is strongly predictive of later life obesity.23
The strengths of the current study include maximizing the sample size by combining data from multiple pediatric ECHO cohorts representing a diverse sample of the US population. A related strength was the ability to adjust for important confounding variables in the analyses. Even as we note the strength of harmonizing variables among 21 cohorts, one considerable limitation is the variety of methods by which each of the contributing cohorts collected infant feeding information. ECHO is a consortium of cohorts that were established before ECHO existed, thus extant data from the various cohorts were inconsistent in the definition of exclusive breastfeeding and did not capture information on pumped milk. The use of parental-reported data on a child’s anthropometric measures could increase the potential for outcome misclassification. In addition, BMI is an imperfect measure of adiposity, and thus, the results may not reflect the true relationship between breastfeeding and child adiposity. Future data collected under the established ECHO-wide protocol will reduce these ambiguities and improve our ability to assess similar patterns with subsequent generations of ECHO participants.
It is difficult to separate the effects of breastfeeding or exposure to human milk itself from the potential confounding effects associated with the characteristics of women who desire to breastfeed and can overcome obstacles that may otherwise prevent longer breastfeeding duration (eg, cesarean delivery, gestational diabetes mellitus, preeclampsia). Our findings did not materially change after adjusting for these confounders (Supplemental Table 4). In addition, some maternal characteristics (eg, pre-pregnancy BMI, gestational weight gain) have been reported to have stronger effects than breastfeeding on child obesity,35,36 and any effects of breastfeeding have been shown to weaken over the life course.36,37 Our study did not account for additional exposures that children experience during early childhood (eg, dietary patterns, physical activity, chemical exposures), which are also correlated with child obesity risk.
Alternatively, our results may indicate the physiologic impacts of obesity on ability to breastfeed. If women with more severe obesity are less likely to breastfeed, it is possible that more severe obesity is also driving the higher BMIz in their offspring. Conversely, women with obesity who were able to breastfeed may have had a different metabolic profile during pregnancy, which also impacts offspring health. It is possible that breastfeeding simply reflects the upstream obesity-related factors, rather than being protective itself. With the difficult and entangled nature of assessing causality, studies assessing physiologic and behavioral factors (including both mother and child) have been recommended,4,29 and more rigorously controlled studies with physiologic, not just anthropometric, phenotyping are justified.
Conclusions
Based on the associations identified here, future studies and public health prevention efforts should continue focusing on addressing 2 highly prevalent problems that disproportionately affect marginalized populations resulting in adverse health outcomes: shortened duration of breastfeeding and maternal overweight and obesity. These results support the encouragement of all women, including women who are overweight or obese before conception, to breastfeed as a preventive measure against the development of childhood obesity.
Supplementary Material
Acknowledgments
The authors wish to thank our ECHO colleagues, the medical, nursing, and program staff, and the children and families participating in the ECHO cohorts. We also acknowledge the contribution of the following ECHO program collaborators:
ECHO Components: Coordinating Center: Duke Clinical Research Institute, Durham, North Carolina: Smith PB, Newby KL; Data Analysis Center: Johns Hopkins University Bloomberg School of Public Health, Baltimore, Maryland: Jacobson LP; Research Triangle Institute, Durham, North Carolina: Catellier DJ; Person-Reported Outcomes Core: Northwestern University, Evanston, Illinois: Gershon R, Cella D.
ECHO Awardees and Cohorts: Rhode Island Hospital, Providence RI: Koinis Mitchell D; Avera Health Rapid City, Rapid City, SD: Elliott A; Kaiser Permanente Northern California Division of Research, Oakland, CA: Croen L; Boston Medical Center, Boston MA: Bacharier L; O’Connor G; Children’s Hospital of New York: New York, NY: Bacharier L; Kattan M; Johns Hopkins University, School of Medicine, Baltimore, MD: Wood R; Bacharier L; Washington University in St Louis, St Louis, MO: Rivera-Spoljaric K; Oregon Health and Science University, Portland, OR: McEvoy C; Indiana University, Riley Hospital for Children: Indianapolis, IN, Tepper R; University of Pittsburgh Medical Center, Magee Women’s Hospital, Pittsburgh, PA: Simhan H; University of Rochester Medical Center Rochester, NY: O’Connor T; Harvard Pilgrim Health Care Institute, Boston, MA: Oken E; University of Illinois, Beckman Institute, Urbana, IL: Schantz S; University of California, San Francisco:, San Francisco, CA: Woodruff T; Boston Children’s Hospital, Boston MA: Bosquet-Enlow M.
Glossary
- BMIz
BMI for age z-score
- ECHO
Environmental Influences on Child Health Outcomes
- RCT
randomized controlled trial
Footnotes
Dr Shipp conceptualized and designed the study, interpreted the data, and drafted the initial manuscript; Dr Wosu conducted the analyses, interpreted the data, and drafted the initial manuscript; Dr Knapp conceptualized and designed the study, conducted the analyses, interpreted the data, and drafted the initial manuscript; Dr Sauder conceptualized and designed the study, acquired the data, and interpreted the data; Drs Dabelea, Dunlop, Deoni, Gern, Karagas, Carroll, Johnson, Meeker, and Cordero acquired the data; Drs Perng, Zhu, Aris, Sathyanarayana, and Hockett interpreted the data; Drs Ferrara, Porucznik, O’Connor, and Wright acquired the data and interpreted the data; Dr Paneth conceptualized and designed the study and interpreted the data; Drs Comstock and Kerver conceptualized and designed the study, interpreted the data, and drafted the initial manuscript; and all authors critically reviewed and revised the manuscript, approved the final manuscript as submitted, and agree to be accountable for all aspects of the work.
Deidentified data from the ECHO Program are available through NICHD’s Data and Specimen Hub (DASH). DASH is a centralized resource that allows researchers to access data from various studies via a controlled-access mechanism. Researchers can now request access to these data by creating a DASH account and submitting a Data Request Form. The NICHD DASH Data Access Committee will review the request and provide a response in ∼2 to 3 weeks. Once granted access, researchers will be able to use the data for 3 years. See the DASH Tutorial for more detailed information on the process.
FUNDING: Funded by the National Institutes of Health (NIH). Research reported in this publication was supported by the Environmental influences on Child Health Outcomes (ECHO) program, Office of the Director, NIH, under Award Numbers U2COD023375 (Coordinating Center), U24OD023382 (Data Analysis Center), U24OD023319 with co-funding from the Office of Behavioral and Social Science Research (PRO Core), UH3OD023251 (Alshawabkeh), UH3OD023248 (Dabelea), UH3OD023313 (Koinis Mitchell), UH3OD023318 (Dunlop), UH3OD023279 (Elliott), UH3OD023289 (Ferrara), UH3OD023282 (Gern), UH3OD023365 (Hertz-Picciotto), UH3OD023275 (Karagas), UH3OD023271 (Karr), UH3OD023288 (McEvoy), UH3OD023349 (O’Connor), UH3OD023286 (Oken), UH3OD023285 (Kerver), UH3OD023272 (Schantz), UH3OD023249 (Stanford), UH3OD023337 (Wright). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. The funder did not participate in the work.
CONFLICT OF INTEREST DISCLOSURES: Dr Dunlop reports NIH grant funding as a source of salary to perform aspects of this work. Dr Deoni reports salary and grant support, and speaker honorarium from Nestle Research and Wyeth Nutrition. Dr Porucznik served on the Board of Directors for the International Board of Lactation Consultant Examiners from 2016 to 2022.
References
- 1. Fryar CD, Carroll MD, Ogden CL. Prevalence of overweight, obesity, and severe obesity among children and adolescents aged 2–19 years: United States, 1963–1965 through 2015–2016. Available at: https://www.cdc.gov/nchs/data/hestat/obesity-child-17-18/obesity-child.htm. Accessed October 4, 2022
- 2. Centers for Disease Control and Prevention. Prevalence of childhood obesity in the United States. Available at: https://www.cdc.gov/obesity/data/childhood.html. Accessed October 4, 2022
- 3. Park MH, Falconer C, Viner RM, Kinra S. The impact of childhood obesity on morbidity and mortality in adulthood: a systematic review. Obes Rev. 2012;13(11):985–1000 [DOI] [PubMed] [Google Scholar]
- 4. Yan J, Liu L, Zhu Y, Huang G, Wang PP. The association between breastfeeding and childhood obesity: a meta-analysis. BMC Public Health. 2014;14(1):1267. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Vos MB, Welsh J. Childhood obesity: update on predisposing factors and prevention strategies. Curr Gastroenterol Rep. 2010;12(4):280–287 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Meek JY, Noble L. Section on breastfeeding; Policy statement: Breastfeeding and the use of human milk. Pediatrics. 2022;150(1):e2022057988. [DOI] [PubMed] [Google Scholar]
- 7. Eidelman AI, Schanler RJ, Johnston M, et al. ; Section on Breastfeeding. Breastfeeding and the use of human milk. Pediatrics. 2012;129(3):e827–e841 [DOI] [PubMed] [Google Scholar]
- 8. Thompson AL. Intergenerational impact of maternal obesity and postnatal feeding practices on pediatric obesity. Nutr Rev. 2013;71(Suppl 1):S55–S61 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Amir LH, Donath S. A systematic review of maternal obesity and breastfeeding intention, initiation and duration. BMC Pregnancy Childbirth. 2007;7(1):9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Hashemi-Nazari S-S, Hasani J, Izadi N, et al. The effect of pre-pregnancy body mass index on breastfeeding initiation, intention and duration: a systematic review and dose-response meta-analysis. Heliyon. 2020;6(12):e05622. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Enstad S, Cheema S, Thomas R, et al. The impact of maternal obesity and breast milk inflammation on developmental programming of infant growth. Eur J Clin Nutr. 2021;75(1):180–188 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Gillman MW, Blaisdell CJ. Environmental influences on child health outcomes, a research program of the National Institutes of Health. Curr Opin Pediatr. 2018;30(2):260–262 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Romano ME, Buckley JP, Elliott AJ, et al. SPR perspectives: scientific opportunities in the Environmental influences on Child Health Outcomes Program. Pediatr Res. 2022;92(5):1255–1261 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Knapp EA, Kress AM, Parker CB, et al. The Environmental influences on Child Health Outcomes (ECHO)-wide cohort. Am J Epidemiol. 2023;192(8):1249–1263 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Centers for Disease Control and Prevention. 2000 CDC Growth Charts for the United States. Hyattsville, MD: Public Health Service, Centers for Disease Control and Prevention, National Center for Health Statistics; 2002 [Google Scholar]
- 16. Rasmussen KM, Catalano PM, Yaktine AL. New guidelines for weight gain during pregnancy: what obstetrician/gynecologists should know. Curr Opin Obstet Gynecol. 2009;21(6):521–526 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Laird NM, Ware JH. Random-effects models for longitudinal data. Biometrics. 1982;38(4):963–974 [PubMed] [Google Scholar]
- 18. Singh GK, DiBari JN. Marked disparities in pre-pregnancy obesity and overweight prevalence among US women by race/ethnicity, nativity/immigrant status, and sociodemographic characteristics, 2012–2014. J Obes. 2019;2019:2419263. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Gaillard R, Felix JF, Duijts L, Jaddoe VW. Childhood consequences of maternal obesity and excessive weight gain during pregnancy. Acta Obstet Gynecol Scand. 2014;93(11):1085–1089 [DOI] [PubMed] [Google Scholar]
- 20. Godfrey KM, Reynolds RM, Prescott SL, et al. Influence of maternal obesity on the long-term health of offspring. Lancet Diabetes Endocrinol. 2017;5(1):53–64 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Rito AI, Buoncristiano M, Spinelli A, et al. Association between characteristics at birth, breastfeeding and obesity in 22 countries: the WHO European Childhood Obesity Surveillance Initiative–COSI 2015/2017. Obes Facts. 2019;12(2):226–243 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Dewey K, Bazzano L, Davis T, et al. The Duration, Frequency, and Volume of Exclusive Human Milk and/or Infant Formula Consumption and Overweight and Obesity: A Systematic Review. Alexandria, VA: USDA; 2020 [PubMed] [Google Scholar]
- 23. Singh AS, Mulder C, Twisk JW, et al. Tracking of childhood overweight into adulthood: a systematic review of the literature. Obes Rev. 2008;9(5):474–488 [DOI] [PubMed] [Google Scholar]
- 24. Hildebrand JS, Ferguson PL, Sciscione AC, et al. Breastfeeding associations with childhood obesity and body composition: Findings from a racially diverse maternal-child cohort. Child Obes. 2022;18(3):178–187 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Patel R, Oken E, Bogdanovich N, et al. Cohort profile: the promotion of breastfeeding intervention trial (PROBIT). Int J Epidemiol. 2014;43(3):679–690 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Kramer MS, Matush L, Vanilovich I, et al. ; PROBIT Study Group. Effects of prolonged and exclusive breastfeeding on child height, weight, adiposity, and blood pressure at age 6.5 y: evidence from a large randomized trial. Am J Clin Nutr. 2007;86(6):1717–1721 [DOI] [PubMed] [Google Scholar]
- 27. Kramer MS, Matush L, Vanilovich I, et al. A randomized breast-feeding promotion intervention did not reduce child obesity in Belarus. J Nutr. 2009;139(2):417S–421S [DOI] [PubMed] [Google Scholar]
- 28. Martin RM, Kramer MS, Patel R, et al. Effects of promoting long-term, exclusive breastfeeding on adolescent adiposity, blood pressure, and growth trajectories: a secondary analysis of a randomized clinical trial. JAMA Pediatr. 2017;171(7):e170698. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Dewey KG, Güngör D, Donovan SM, et al. Breastfeeding and risk of overweight in childhood and beyond: a systematic review with emphasis on sibling-pair and intervention studies. Am J Clin Nutr. 2021;114(5):1774–1790 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Bever Babendure J, Reifsnider E, Mendias E, et al. Reduced breastfeeding rates among obese mothers: a review of contributing factors, clinical considerations and future directions. Int Breastfeed J. 2015;10(1):21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Park S, Jang IS, Min D. Factors associated with the need for breastfeeding information among women with gestational diabetes mellitus: a cross-sectional study. Asian Nurs Res (Korean Soc Nurs Sci). 2021;15(3):210–214 [DOI] [PubMed] [Google Scholar]
- 32. Lind MV, Larnkjær A, Mølgaard C, Michaelsen KF. Breastfeeding, breast milk composition, and growth outcomes. Nestle Nutr Inst Workshop Ser. 2018;89:63–77 [DOI] [PubMed] [Google Scholar]
- 33. Yu X, Rong SS, Sun X, et al. Associations of breast milk adiponectin, leptin, insulin and ghrelin with maternal characteristics and early infant growth: a longitudinal study. Br J Nutr. 2018;120(12):1380–1387 [DOI] [PubMed] [Google Scholar]
- 34. Isesele P, Enstad S, Huong P, et al. Breast milk from non-obese women with a high omega-6 to omega-3 fatty acid ratio, but not from women with obesity, increases lipogenic gene expression in 3T3-L1 preadipocytes, suggesting adipocyte dysfunction. Biomedicines. 2022;10(5):1129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Voerman E, Santos S, Patro Golab B, et al. Maternal body mass index, gestational weight gain, and the risk of overweight and obesity across childhood: an individual participant data meta-analysis. PLoS Med. 2019;16(2):e1002744. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Kjaer TW, Faurholt-Jepsen D, Medrano R, et al. Higher birthweight and maternal pre-pregnancy BMI persist with obesity association at age 9 in high risk Latino children. J Immigr Minor Health. 2019;21(1):89–97 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Kvaavik E, Tell GS, Klepp K-I. Surveys of Norwegian youth indicated that breast feeding reduced subsequent risk of obesity. J Clin Epidemiol. 2005;58(8):849–855, e841 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.