Breastfeeding Reduces Childhood Obesity Risks

Abstract

Background:

The present study examined the effects of breastfeeding and its duration on the development of childhood obesity from 24 months through grade 6.

Methods:

U.S. longitudinal data collected from 1234 children were analyzed using logistic regression models and generalized estimating equation (GEE). Child height and weight were measured six times at ages of 24 months, 36 months, 54 months, grade 1, grade 3, and grade 6.

Results:

During the early 1990s, prevalence of breastfeeding was low in the United States, 60% and 48% at 1 and 6 months, respectively. Nonsmoking, white, married mothers with both parents in the household, and with income above the poverty line, were more likely to breastfeed at 1 month of age of their babies. Obesity rate of the children increased with age from 24 months to grade 6. Logistic regression showed that breastfeeding at month 1 was associated with 53% (odds ratio [OR]: 0.47, 95% confidence interval [CI]: 0.30–0.73) and 47% (OR: 0.53, 95% CI: 0.36–0.78) decreased risks for childhood obesity at grades 1 and 6, respectively. GEE analysis showed that breastfeeding at 1 month reduced risk for childhood obesity by 36% (95% CI: 0.47–0.88) from ages 24 months through grade 6. Regarding breastfeeding duration, more than 6 months (vs. never) was associated with a decreased risk for childhood obesity by 42% (OR: 0.58, 95% CI: 0.36–0.94).

Conclusions:

Breastfeeding at 1 month and more than 6 months reduced the risk of childhood obesity. Rate of breastfeeding was low in the United States in the 1990s, which may have had long-term implications on children.

Introduction

The prevalence of childhood obesity has increased substantially in the United States and worldwide during the past three decades.^1,2 In developing countries, between 1980 and 2013, the prevalence of obesity in children has increased from 8.1% to 12.9% for boys and 8.4% to 13.4% for girls.³ In developed countries, 23.8% of boys and 27.6% of girls were overweight or obese in 2013.³ In the United States, national data show that from 2011 to 2014, the prevalence of obese children was 8.9% for children aged 2–5, 17.5% for 6–11, and 20.5% for 12–19.⁴ Obese children are more likely to remain obese into adulthood and develop chronic diseases such as type 2 diabetes, heart disease, stroke, and some types of cancers.^5–7 It is imperative to study childhood obesity comprehensively, as well as to develop interventions to prevent obesity from occurring. Past research has examined causes of childhood obesity to include environmental factors, genetic factors, personal behaviors, and dietary habits.⁸ Some childhood obesity prevention programs have shown some success.^9,10

The association between breastfeeding and childhood obesity can be explained by looking at the biological mechanisms surrounding the process. Breast milk contains a lower amount of calories and nutrients such as sugar, water, protein, and fat, and includes bioactive substances, including leptin and ghrelin.⁸ Leptin is a hormone that helps control hunger.¹¹ The higher protein and fat levels in formula have been associated with higher adiposity levels.⁸ This association could be attributed to the increase of insulin, which encourages the storage of fat levels in the presence of high protein.¹² Formula-fed infants have higher levels of insulin and a longer insulin response than breastfed infants.¹²

Some research has reported a protective effect of breastfeeding in lowering childhood obesity risk. Breastfeeding is protective considering the lower caloric intake as well as lower fat and protein levels provided by breast milk rather than by formula.^8,13 Research supporting this theory has found an inverse relationship between duration of breastfeeding and childhood obesity.^7,8,14–17 A recent meta-analysis reviewing studies from 12 different countries concluded that breastfeeding was significantly associated with a reduced risk of childhood obesity.⁸ The results showed a 22% lower risk for those who have ever been breastfed compared to those who have never been breastfed, and those breastfed for more than 7 months had the most significant decrease in risk.⁸ A similar review investigated how breastfeeding affects the body composition of infants by using 15 studies in a systematic review and 11 studies for a meta-analysis.¹⁶ It was determined that, within the first year of life, formula can alter the normal trajectory of adipose.¹⁶

A World Health Organization report assessed 71 studies. Its pooled analysis shows that children who were breastfed had reduced odds of becoming overweight or obese, for those aged 1–9 years (odds ratio [OR] = 0.77) and aged 10–19 years (OR = 0.62).¹⁸ Another study examined the effect of breastfeeding on the weight status of 2–4-year olds in the United States and reported that those breastfed were more likely to have a healthy weight status at age 2 (68.2% vs. 59.9%) and 4 (69.5% vs. 65.4%).¹⁹ The Jiaxing Birth Cohort study in China concluded that those breastfed exclusively for 3–5 and ≥6 months had a lower risk of becoming overweight by 13% and 37%, respectively, compared to those breastfed for <1 month.⁷

On the contrary, some other studies have concluded that breastfeeding and childhood obesity have a small or insignificant relationship.^6,20,21 For example, one contradicting study conducted by the China Family Panel Studies found that there was no significant difference on the effect of breastfeeding for obesity or excessive weight.⁶

The differences in study findings on the relationship between breastfeeding and childhood obesity risks could be attributed to differences in study methodologies and analysis (e.g., adjustments for confounding factors such as socioeconomic status, maternal BMI,²² and maternal smoking^15,23,24) definitions, sample size, and not distinguishing between exclusive and partial breastfeeding.⁸ In addition, many previous studies defined obesity based on self-reported BMI, not on standard measured height and weight.^25,26 Moreover, a series of more appropriate methods for studying the relationship between breastfeeding and obesity have been established.²⁷

Therefore, further research based on accurate longitudinal data is needed to assess the causality. The present study examined the effects of breastfeeding and its duration on the development of childhood obesity from 24 months through grade 6 in the United States. We also assessed the breastfeeding practice.

Methods

Data Set and Study Sample

Data collected from a selected cohort from the National Institute of Child Health and Human Development (NICHD) Study of Early Child Care and Youth Development (SECCYD) were used. The main aim of SECCYD was to examine how early child care experience affects psychologic development from infancy to adolescence. The cohort consisted of 1364 infants and their families throughout the United States. Data collection occurred during four phases at 10 distinct sites: Phase I (1991–1995, when the children were 0–3 years), Phase II (1996–1999, through grade 1), Phase III (2000–2004, through grade 6), and Phase IV (2005–2007, through grade 9).

Families were enrolled in the study on successful completion of data collection at the 1-month interview. Recruitment was accomplished during the first 11 months of 1991 and 1364 families were enrolled. A more complete description of the data collection procedures can be found in the Manuals of Operation for the NICHD Study of Early Child Care (www.nichd.nih.gov/research/supported/seccyd/overview.cfm).

The present study restricted the analysis to children with measured height and weight data at least in one of the six time points of the follow-up surveys (i.e., age of 24 months, 36 months, 54 months, grades 1, 3, or 6) and to those with information on their breastfeeding history at 1 month after birth (n = 1234). This study was approved by the Institutional Review Board at East Tennessee State University. Study protocols for the original data collection were approved by the institutional review boards of the participating universities.

Study Variables

Outcome Variables: Childhood Overweight and Obesity

Standardized procedures were used to measure height and weight during the interviews. Height was measured with children standing without shoes, feet together, and their backs against a calibrated seven-foot measuring stick. Weight was measured using a physician's two-beam scale. Scales were calibrated monthly using certified calibration weights. Weight was measured with children in minimal clothing and recorded twice, each time to the nearest 0.25 pound (0.1 kg). BMI, defined as weight in kilograms divided by squared height in meters, was used to determine whether the child was overweight and obese. The CDC age- and gender-specific BMI growth charts were used to determine a percentile score.²⁸ Anthropometric measures, height and weight, for participants were documented during six distinct time periods: 24 months, 36 months, 54 months, grade 1, grade 3, and grade 6. Childhood obesity was determined if the participant's BMI (kg/m²) was ≥95th percentile in accordance with the 2000 CDC gender- and age-specific growth chart. Childhood overweight or obesity was determined if the participant's BMI (kg/m²) was ≥85th percentile.

Exposure Variable: Breastfeeding

Mothers were asked whether they were currently breastfeeding at the visit when the child was 1, 6, 15, and 24 months old. Only information on breastfeeding at 1 and 6 months was used due to a large amount of missing data on other time points. Breastfeeding was dichotomized into breastfeeding and not breastfeeding.

Maternal Covariates

Maternal characteristics collected of interest were poverty level, marital status, and prenatal smoking status. Family income was reported by mothers and determined by an income-to-needs ratio (total family income relative to the federal poverty line for a family of a particular size). If the ratio of income to needs was less than 1, households were considered below poverty level, while households with 1 and above were considered at or above poverty level. Maternal marital status included married and living together, partnered and living together, separated and not living together, divorced and not living together, widowed, never married, and others. Due to the small sample size for subgroups, it was dichotomized into married and living together and other. The prenatal smoking variable was assessed by questions that addressed “The Year before My Child Was Born.” Smoking was dichotomized into never smoked and ever smoked.

Child Covariates

Child's variables were sex, race, and birthweight (grams). Birthweight was obtained from medical chart review. Child sex and race were determined at the interview when the child was 1 month of age. Due to the small sample size, only three race categories were used: whites, blacks, and others.

Statistical Analysis

First, we conducted descriptive analysis examining the breastfeeding status among U.S. children in comparison to the child and mothers' characteristics. Then using chi-square and Student's t tests, we tested if these factors were associated with breastfeeding status. Simple descriptive statistics, including means, standard deviations, and proportions, were used to describe childhood overweight and obesity by breastfeeding status at different ages of children. Univariate and multiple logistic regressions (model 1) were used to estimate the effects of breastfeeding at 1 month on the risk of childhood overweight and obesity, respectively, at six different assessment time points as the children grew older (i.e., at 24 months, 36 months, 54 months, grade 1, grade 3, and grade 6).

To account for the within-subject correlation of response on dependent variables of different distributions,²⁹ we used the generalized estimating equation (GEE) method to estimate the parameters, because in comparison with traditional regression analysis at one time point (cross sectional) or analysis of variance to compare a certain outcome variable between two different time points, longitudinal GEE considers all repeated measurements of the outcome accounting for their dependency.³⁰ GEE models accounted for within-subject correlation across the six measurements for weight status between ages of 24 months through grade 6. The working correlation matrix is usually unknown and must be specified in the analysis model. The auto-regressive with first-order working correlation matrix (decoded as AR(1)) has been widely used to specify the within-subject correlation structure in many repeated assessment settings. The correlation matrix can efficiently and also parsimoniously parameterize the within-subject correlations decreasing exponentially in time or space. Moreover, robust estimation of parameter estimates was adopted since the robust estimation produces consistent point estimates and standard errors even if the working correlation matrix was misspecified.^31–33 In this study, AR(1) was used as the working correlation structure in the analysis.

To examine the effect of breastfeeding duration on obesity risk, we restricted the analysis to the 545 children who had complete data on breastfeeding at 1 and 6 months. Breastfeeding duration was categorized as ≥6 months, <6 months, and never breastfed.

The effects of breastfeeding at 1 and 6 months duration on overweight and obesity risks from 24 months through grade 6 were examined using GEE, respectively (models 2 and 3). All adjusted models controlled for maternal characteristics (poverty level, marital status, and prenatal smoking status) and child characteristics (sex, race, and birthweight). The odds ratios (ORs) and their 95% confidence intervals (CIs) were reported. All data analyses were performed using SAS version 9.2 statistical software (SAS Institute, Inc., Cary, NC).

Results

Table 1 shows characteristics of the participants by breastfeeding status. In total, 60% of mothers breastfed their children at month 1 after birth. Mothers who were at or above the poverty line (vs. below poverty line), having a marital status of married and living together (vs. other), and never smoked (vs. ever smoked) had a higher prevalence for breastfeeding their children (67% vs. 37%, 67% vs. 32%, and 64% vs. 48%, respectively, all p < 0.0001). The prevalence of breastfeeding was highest among white children (66%) and lowest among black children (20%). Moreover, the mean birthweight in grams for children who were breastfed was found to be higher than that of those who were not.

Table 1.

Breastfeeding Status (%) among U.S. Children, by Children and Their Mothers' Characteristics: NICHD SECCYD^a (n = 1234)

	Not breastfeeding	Breastfeeding	p
Overall, n (%)	499 (40.4)	735 (59.6)
Maternal characteristics
Poverty level^b			<0.0001
<Poverty line (%)	141 (62.7)	84 (37.3)
≥Poverty line (%)	310 (33.4)	619 (66.6)
Marital status^b			<0.0001
Married and living together (%)	321 (33.1)	648 (66.9%)
Other (%)	173 (67.8)	82 (32.2%)
Smoking status 1 year before birth^b			<0.0001
Never smoked (%)	327 (36.1)	578 (63.9)
Ever smoked (%)	143 (52.4)	130 (47.6)
Offspring characteristics
Sex^b			0.5745
Male (%)	260 (41.2)	371 (58.8)
Female (%)	239 (39.6)	364 (60.4)
Ethnicity^b			<0.0001
Whites (%)	346 (34.4)	661 (65.6)
Blacks (%)	117 (79.6)	30 (20.4)
Others (%)	36 (45.0)	44 (55.0)
Birthweight^c
Mean birthweight, g (SD)	3446.3 (509.8)	3530.5 (501.2)	0.0041

Number of missing observations: poverty = 42; smoking status = 15.

Mothers were interviewed at children 24 months after birth for their smoking status within 1 year before the birth.

Chi-square (χ²) test was used, the numbers indicate percentage values.

Student^'s t-test was used, the numbers indicate mean ± SD.

NICHD SECCYD, National Institute of Child Health and Human Development, the Study of Early Child Care and Youth Development; SD, standard deviation.

Table 2 describes the distribution of mean BMI percentile (BMI-P) and proportions of overweight and obese children by breastfeeding status at the time of assessment. In general, mean BMI-P or the percentage of children overweight and obese increased as child's age increased from 24 months through grade 6, overall or stratified by breastfeeding status, suggesting a positive relationship between age and being overweight or obese. In addition, breastfed children appeared to be less likely to be overweight and obese than those who were not breastfed, and on average, had lower BMI-Ps.

Table 2.

Distribution of Mean BMI-P^a and Proportion (%) of Overweight and Obese Children^b by Breastfeeding Status and Time of Assessment

	Not breastfeeding				Breastfeeding				Total
Time/child age	n	BMI-P^a	Overweight^b (%)	Obesity^b (%)	n	BMI-P^a	Overweight^b (%)	Obesity^b (%)	n	BMI-P^a	Overweight^b (%)	Obesity^b (%)
24 Months	386	57.1	18.4	6.7	605	54.6	13.9	4.8	991	55.0	15.6	5.6
36 Months	429	56.6	20.5	7.0	661	53.7	17.4	5.9	1090	54.8	18.6	6.3
54 Months	383	63.5	27.9	11.2	648	59.4	22.8	8.2	1031	60.9	24.7	9.3
Grade 1	372	65.6	29.8	15.6	619	61.2	22.6	9.4	991	62.8	25.3	11.7
Grade 3	358	67.8	34.9	20.4	580	62.9	29.0	14.5	938	64.7	31.2	16.7
Grade 6	348	68.6	41.7	24.1	569	61.3	29.4	15.1	917	64.1	34.0	18.5

A standardized protocol to measure child's weight and height was used at all six times (24 months, 36 months, 54 months, grade1, grade 3, and grade 6). BMI was calculated by [BMI = weight (kg)/height (m)²].

Overweight and obesity were defined as a BMI-for-age above the 85th and 95th percentile, respectively, of the U.S. Centers for Disease Control and Prevention sex-specific BMI-for-age growth charts. Overweight (BMI-P ≥ 85) in this study included obesity.

BMI-P, BMI percentile.

Table 3 examines distribution of childhood overweight and obesity by breastfeeding status and time of assessment. With regard to breastfeeding duration, it appears that the lowest prevalence of obesity was found in children who were breastfed for 6 months compared to those who were breastfed less than 6 months or never breastfed. In addition, the differences in the prevalence of obesity in these groups became larger with increasing age, for example, 5.0% vs. 9.7% at 24 months, while 14.4% vs. 22.4% in grade 6.

Table 3.

Distribution of Childhood Overweight and Obesity by Breastfeeding Status and Time of Assessment

	Overweight (BMI ≥85th percentile; n, %)			Obesity (BMI ≥95th percentile; n, %)
Time of assessment	BF for ≥6 months	BF <6 months	Never BF	BF for ≥6 months	BF <6 months	Never BF
24 Months	27 (16.9)	22 (12.8)	24 (21.2)	8 (5.0)	9 (5.2)	11 (9.7)
36 Months	34 (19.8)	31 (16.9)	26 (21.0)	6 (3.5)	15 (8.1)	11 (8.8)
54 Months	40 (24.2)	40 (22.4)	30 (25.0)	13 (7.9)	16 (8.9)	13 (10.8)
Grade 1	37 (23.4)	35 (21.0)	30 (26.1)	12 (7.6)	13 (7.8)	14 (12.2)
Grade 3	44 (29.7)	48 (30.0)	39 (35.1)	24 (16.2)	21 (13.1)	24 (21.6)
Grade 6	40 (27.4)	49 (32.5)	39 (36.5)	21 (14.4)	22 (14.6)	24 (22.4)

BF, breastfeeding.

After adjusting for maternal characteristics (poverty status, marital status, and smoking status) and child characteristics (sex, race, and birthweight), those children breastfed at month 1, compared to those not breastfed, had a 29%, 28%, and 39% lower risk of being overweight or obese at ages of 54 months, grade 1, and grade 6, respectively. Compared to children not breastfed, those breastfed at 1-month old had a lower obesity risk at grade 1 and grade 6, by 53% and 47%, respectively (see Model 1 in Table 4). GEE analysis showed that breastfeeding at 1 month of age was negatively associated with BMI-P and reduced risk for overweight and obesity by 24% and 36% from ages of 24 months through grade 6, respectively (see Model 2 in Table 4). Regarding the effect of breastfeeding duration, more than 6 months (vs. never) was associated with a decreased risk for childhood obesity (OR: 0.58, 95% CI: 0.36–0.94) (see Model 3 in Table 4).

Table 4.

Longitudinal Analysis Using Generalized Estimating Equation for the Effects of Breastfeeding and its Duration on BMI Percentile (beta, 95% CI), Overweight and Obesity (OR, 95% CI) from 24 Months through Grade 6

	24–Months	36 Months	54 Months	Grade 1	Grade 3	Grade 6
Model 1^a
BF at 1 month
1. Overweight
Crude OR	0.72 (0.51–1.01)	0.82 (0.60–1.11)	0.76 (0.57–1.02)	0.69 (0.51–0.92)^*	0.76 (0.57–1.01)	0.58 (0.44–0.77)^***
Adjusted OR^d	0.85 (0.57–1.26)	0.85 (0.59–1.22)	0.71 (0.51–0.99)^*	0.72 (0.51–1.00)^*	0.87 (0.63–1.20)	0.61 (0.45–0.84)^**
2. Obesity
Crude OR	0.70 (0.40–1.20)	0.83 (0.51–1.36)	0.70 (0.46–1.08)	0.56 (0.38–0.83)^**	0.66 (0.47–0.93)	0.56 (0.40–0.78)^***
Adjusted OR^d	0.70 (0.37–1.31)	0.88 (0.49–1.58)	0.75 (0.47–1.21)	0.47 (0.30–0.73)^***	0.70 (0.47–1.04)	0.53 (0.36–0.78)^**

Model 2^b
	BMI-P^b		Overweight (BMI-P ≥ 85)^c		Obesity (BMI-P ≥ 95)^c
	Crude β (SE)	Adjusted β (SE)^d	Crude OR (95% CI)	Adjusted OR (95% CI)^d	Crude OR (95% CI)	Adjusted OR (95% CI)^d
BF at 1 month
Not BF (reference)
BF	−4.26 (1.41)^**	−3.71 (1.5)^*	0.72 (0.59–0.88)^**	0.76 (0.60–0.96)^*	0.65 (0.50–0.85)^**	0.64 (0.47–0.88)^**

Model 3^b,e
BF duration (months)	BMI-P^b		Overweight (BMI-P ≥ 85)^c		Obesity (BMI-P ≥ 95)^c
Never BF (reference)	Crude β (SE)	Adjusted β (SE)^d	Crude OR (95% CI)	Adjusted OR (95% CI)^d	Crude OR (95% CI)	Adjusted OR (95% CI)^d
BF for ≥6 months	−3.02 (2.71)	−1.36 (2.83)	0.78 (0.53–1.14)	0.84 (0.53–1.32)	0.58 (0.36–0.94)^*	0.58 (0.32–1.04)
BF <6 months	−5.36 (2.69)^*	−3.13 (2.77)	0.75 (0.51–1.09)	0.86 (0.56–1.32)	0.63 (0.39–1.03)	0.62 (0.35–1.08)

A standardized protocol to measure child's weight and height was used at all six times (from 24 months through grade 6). BMI was calculated by [BMI = weight (kg)/height (m)²].

Logistic regression analysis for the effect of BF at 1 month on the risk for overweight and obesity at multiple time points from 24 months through grade 6.

Models 2 and 3 used GEE for the analysis. Analysis of the GEE parameter estimates was based on the use of first-order auto-regressive working correlation [AR(1)] structure, with BMI percentile as continuous outcome and overweight and obesity as dichotomous outcomes, respectively.

Overweight and obesity were defined as a BMI-for-age above the 85th and 95th percentile, respectively, of the Centers for Disease Control and Prevention sex-specific BMI-for-age growth charts.

After adjusting for maternal characteristics (poverty status, marital status, prenatal smoking status) and child characteristics (sex, race, and birthweight).

Model 3 was based on the sample size of 545 subjects.

< 0.05, ^** < 0.01, ^*** < 0.001.

CI, confidence interval; GEE, generalized estimating equations; OR, odds ratio; SE, standard error.

Discussion

Past research shows an inverse association between breastfeeding and the risk of childhood obesity. However, the amount and duration of decreased risk vary in different studies.^7,14,15,17 A small study in Western Australia determined that children breastfed for ≤4 months had a higher probability of exceeding the 95th percentile than those breastfed for ≥12 months.¹⁷ They also concluded that the 4-month cutoff point was more associated with higher BMI in those from birth to 14 years of age.¹⁷ A similar study in the Brazilian semiarid region concluded that exclusive breastfeeding ≥6 months served as a protective factor against obesity for children as early as the age of 2.¹⁴ Our study examined the effects of breastfeeding and its duration on the risk of childhood overweight and obesity using a national longitudinal data and confirmed the inverse association. Nonsmoking, white, married mothers with both parents in the household, and with income above the poverty line, were more likely to breastfeed at 1 month of age of their babies. Obesity rate of the children increased with age from 24 months to grade 6. Breastfeeding at month 1 was associated with a lower risk for childhood obesity at 54 months, grade 1, and grade 6. The longer duration of breastfeeding was correlated with a lower prevalence of childhood obesity. In addition, breastfeeding at month 1 was associated with a 36% reduced risk for obesity from 24 months through grade 6 using GEE.

Certain studies refute the relationship between childhood obesity and breastfeeding. A Belarusian randomized control trial observed breastfeeding practices in both the control and the intervention group and determined that the intervention had an effect on duration and exclusivity of breastfeeding, but did not influence childhood obesity.²¹ Another study using data from the Child Development Supplement of the Panel Study of Income Dynamics found no causal link between breastfeeding and childhood obesity after adjusting for confounders.²⁰ These results could be due to the retrospective reporting of breastfeeding duration, the inclusion of infants who were preterm or small for gestational age, or not adjusting for maternal BMI.²⁰

As stated in the introduction, the association between breastfeeding and childhood obesity can be explained by observing biological mechanisms surrounding the process. By breastfeeding, the infant is allowed to control how much milk he/she ingests. This allows the infant to learn internal cues of when they are hungry or not, allowing for self-regulation that carries over into healthy eating habits reducing the likelihood of childhood obesity.¹¹ Breastfeeding also provides enough nutrition for the infant so that solid foods do not need to be introduced until he/she is 6 months old. The early introduction of solid foods has also been linked to higher risks of obesity.¹¹

Although a longitudinal study has its strengths, in this case, however, the initial analyses were based on data collected decades ago. The practice of breastfeeding and duration may have changed, possibly altering our results. Selection bias may have also occurred due to lose of follow-up and incomplete data (n = 130). The sample analyzed (n = 1234) had more participants who were at or above the poverty line. Since lower socioeconomic status is associated with higher likelihood of a child becoming obese, this may weaken the strength of our results.³⁴ Known risk factors for childhood obesity were included such as socioeconomic factors, maternal obesity, excessive gestational weight, smoking during pregnancy, and daily activity level.³⁵ Accounting for these factors and adjusting for potential confounders assist in the accuracy of the study. Our data set was also lacking in data such as prenatal maternal BMI and dietary intake for the children, which could have changed the results from adjusting to potential confounding. However, using the NICHD SECCYD data set allowed for strengths in our study. The sample size is relatively large with recruitment of mothers from all over the United States allowing for external validity. SECCYD assessed children's weight status and their breastfeeding status at multiple times, and includes a broad range of family, social, and health-related variables that allowed for the adjustment of potential confounders. Standardized procedures were used to measure children's weight and height directly, which ensure data accuracy.^25,26

Because the data set does allow for external validity in the United States, public health implications can be made. Hospitals and maternal health classes need to promote breastfeeding as a prevention strategy for childhood obesity. Intervention activities have further proof of what groups, such as mothers below the poverty line and those who smoke, they need to target to increase the prevalence of breastfeeding. Public health policy and formula manufacturers need to focus on creating safer alternatives for those mothers who cannot breastfeed their children.

Conclusions

We observed a positive relationship between age and weight status in our longitudinal sample. The lowest prevalence of obesity was found in children who were breastfed for 6 months compared to those who were breastfed less than 6 months or never breastfed. Breastfeeding reduced the risk of childhood obesity. Rate of breastfeeding was low in the United States in the 1990s, which may have some long-term effects on the affected children. Interventions to combat childhood obesity need to include the promotion of breastfeeding.

Footnotes

Acknowledgments

This study was conducted by the National Institute of Child Health and Human Development (NICHD) Early Child Care Research Network supported by NICHD through a cooperative agreement that calls for scientific collaboration between the grantees and the NICHD staff. East Tennessee State University (ETSU) has restricted data use agreements to analyze the SECCYD data. Part of Dr. Liang Wang's effort was supported by the Research Development Committee (RDC) of ETSU (grant no. 15-025M). Part of Dr. Youfa Wang's related effort was supported by research grants from the National Institute of Child Health and Human Development (NICHD, grant nos. U54HD070725, R01HD064685-01A1). Dr. Bin Xie was supported by 15Grant 25700346 from the American Heart Association. The funding agencies have no role in the conduct and writing of this research.

Author Disclosure Statement

No competing financial interests exist.

References

Wang

, Lobstein

. Worldwide trends in childhood obesity. Int J Pediatr Obes, 2006; 1:11–25.

, Fleming

, Robinson

, et al. Global, regional, and national prevalence of overweight and obesity in children and adults during 1980–2013: A systematic analysis for the Global Burden of Disease Study 2013. Lancet, 2014; 384:766–781.

The GBD 2013 Obesity Collaboration, Ng

, Fleming

, et al. Global, regional and national prevalence of overweight and obesity in children and adults 1980–2013: A systematic analysis. Lancet (London, England), 2014; 384:766–781.

Ogden

, Carroll

, Fryar

, Flegal

. Prevalence of obesity among adults and youth: United States, 2011–2014. NCHS Data Brief. 2016. Available at www.cdc.gov/nchs/data/databriefs/db219.pdf Last accessed August 11, 2016.

Wang

, Lim

. The global childhood obesity epidemic and the association between socio-economic status and childhood obesity. Int Rev Psychiatry, 2012; 24:176–188.

Jing

, Xu

, Wan

, et al. Studies, effect of breastfeeding on childhood BMI and obesity: The China Family Panel Studies. Medicine (Baltimore), 2014; 93:e55.

Zheng

, Liu

, Li

, et al. Exclusive breastfeeding is inversely associated with risk of childhood overweight in a large Chinese cohort. J Nutr, 2014; 144:1454–1459.

Yan

, Liu

, Zhu

, et al. The association between breastfeeding and childhood obesity: A meta-analysis. BMC Public Health, 2014; 14:1267.

Wang

, Wu

, Wilson

, et al. Childhood Obesity Prevention Programs: Comparative Effectiveness Review and Meta-Analysis [Internet]. Agency for Healthcare Research and Quality (US): Rockville, MD, 2013. Report No.: 13-EHC081-EF. AHRQ Comparative Effectiveness Reviews.

10.

Wang

, Cai

, Wu

, et al. What childhood obesity prevention programmes work? A systematic review and meta-analysis. Obes Rev, 2015; 16:547–565.

11.

Moreno

, Furtner

, Rivara

. Breastfeeding as obesity prevention. Arch Pediatr Adolesc Med, 2011; 165:772.

12.

Hediger

, Overpeck

, Kuczmarski

, Ruan

. Association between infant breastfeeding and overweight in young children. J Am Med Assoc, 2001; 285:2453–2460.

13.

Stolzer

. Breastfeeding and obesity: A meta-analysis. Open J Prev Med, 2011; 1:88–93.

14.

Assunção

, Ferreira

, Coutinho

, et al. Protective effect of breastfeeding against overweight can be detected as early as the second year of life: A study of children from one of the most socially-deprived areas of Brazil. J Health Popul Nutr, 2015; 33:85–91.

15.

Bammann

, Peplies

, De Henauw

, et al. Early life course risk factors for childhood obesity: The IDEFICS case-control study. PLoS One, 2014; 9:e86914.

16.

Gale

, Logan

, Santhakumaran

, et al. Effect of breastfeeding compared with formula feeding on infant body composition: A systematic review and meta-analysis. Am J Clin Nutr, 2012; 95:656–669.

17.

Oddy

, Mori

, Huang

, et al. Early infant feeding and adiposity risk: From infancy to adulthood. Ann Nutr Metab, 2014; 64:262–270.

18.

Horta

, Victora

. Long-Term Effects of Breastfeeding: A Systematic Review. World Health Organization: Geneva, Switzerland, 2013.

19.

Moss

, Yeaton

. Early childhood healthy and obese weight status: Potentially protective benefits of breastfeeding and delaying solid foods. Matern Child Health J, 2013; 18:1224–1232.

20.

Jiang

, Foster

. Duration of breastfeeding and childhood obesity: A generalized propensity score approach. Health Serv Res, 2013; 48(2 Pt 1):628–651.

21.

Martin

, Patel

, Kramer

, et al. Effects of promoting increased duration and exclusivity of breastfeeding on adiposity and insulin-like growth factor-I at age 11.5 years: A randomized trial. JAMA, 2013; 309:1005–1013.

22.

Leng

, Li

, Zhang

, et al. GDM women's pre-pregnancy overweight/obesity and gestational weight gain on offspring overweight status. PLoS One, 2015; 10:e0129536.

23.

Wang

, Mamudu

, Alamian

, et al. Independent and joint effects of prenatal maternal smoking and maternal exposure to secondhand smoke on the development of adolescent obesity: A longitudinal study. J Paediatr Child Health, 2014; 50:908–915.

24.

Wang

, Mamudu

, Wu

. The impact of maternal prenatal smoking on the development of childhood overweight in school-aged children. Pediatr Obes, 2013; 8:178–188.

25.

Keith

, Fontaine

, Pajewski

, et al. Use of self-reported height and weight biases the body mass index-mortality association. Int J Obes, 2011; 35:401–408.

26.

Krebs

, Himes

, Jacobson

, et al. Assessment of child and adolescent overweight and obesity. Pediatrics, 2007; 120(suppl 4):S193–S228.

27.

Adair

. Methods appropriate for studying the relationship of breast-feeding to obesity. J Nutr, 2009; 139:408S–411S.

28.

Kuczmarski

, Ogden

, Guo

, et al. CDC growth charts for the United States: Methods and development. Vital Health Stat, 2000; 11:1–190.

29.

Zeger

, Liang

. Longitudinal data analysis for discrete and continuous outcomes. Biometrics, 1986; 42:121–130.

30.

Twisk

JWR

. Applied Longitudinal Data Analysis for Epidemiology: A Practical Guide. Cambridge University Press: Cambridge, 2003, pp. 120–144.

31.

Horton

NJ LS

. Review of software to fit generalized estimating equation regression models. Am Stat, 1999; 53:160–169.

32.

Liang

, Beaty

, Cohen

. Application of odds ratio regression models for assessing familial aggregation from case-control studies. Am J Epidemiol, 1986; 124:678–683.

33.

Zeger

, Liang

. An overview of methods for the analysis of longitudinal data. Stat Med, 1992; 11:1825–1839.

34.

Wang

, Beydoun

. The obesity epidemic in the United States- gender, age, socioeconomic, racial/ethnic, and geographic characteristics: A systematic review and meta-regression analysis. Epidemiol Rev, 2007; 29:6–28.

35.

Robinson

, Crozier

, Harvey

, et al. Modifiable early-life risk factors for childhood adiposity and overweight: An analysis of their combined impact and potential for prevention. Am J Nutr, 2015; 101:368–375.