Posted:06/24/2004

Does Breastfeeding Protect Against Pediatric Overweight?

Analysis of Longitudinal Data From the Centers for Disease Control and Prevention Pediatric Nutrition Surveillance System

Grummer-Strawn LM, Mei Z. CDC Pediatric Nutrition Surveillance System
Pediatrics. 2004;113(2):e81-86
Pediatrics -February 2004 (Volume 113, Number 2)

The authors reviewed the contradictory findings in past research to determine whether breastfeeding prevents childhood overweight. Some studies have found that breastfeeding is associated with overweight, whereas others have shown that breastfeeding is associated with lower risk of childhood overweight. Many of the studies have small sample sizes, however, and fewer still generated population-based data.

This study evaluated the associations between duration of breastfeeding and childhood weight, with a particular interest in ethnic subgroups. The investigators employed data from the CDC's Pediatric Nutrition Surveillance System, a national data set with measures from children followed by public health clinics in the United States. The authors evaluated children born between 1988 and 1992, and constructed their histories until age 5. Only children with at least 1 visit in each year of life were included.

More than 177,000 children were included in the analysis data set. In a subset of the data, 7 states provided data on maternal pregnancy, thus allowing the authors to control for potential maternal and pregnancy effects on childhood weight. They divided duration of breastfeeding into 6 groupings: "never," "less than 1 month," "1-2 months," "3-5 months," " 6-11 months," and "12 months or greater." The outcome variable of interest was whether the child was overweight (body mass index (BMI)-for-age) > 95th percentile) at a visit after 4 years of age but before turning 5. The authors controlled for gender, ethnicity, and birth weight. In the subset of children with linked maternal pregnancy data, the authors also controlled for multiple factors that might influence weight, including maternal pre-pregnancy BMI and educational level.

The sample was 52% white, non-Hispanic; 29% black, non-Hispanic; and 14% Hispanic. The sample was comprised of mostly low-income families due to the use of public health clinic data. Seventy-one percent of the children were never breastfed. Roughly 10% were breastfed for < 1 month. The percentages continued to decline with duration, with only 4% breastfed for 6-11 months and only 2.4% reaching 12 months or longer.

At 4 years of age, 13.3% of the children were overweight (16.3% in the sample with linked maternal data). The percentage of children who were overweight declined with each incremental increase in duration of breastfeeding. For example, for children never breastfed, 13.6% were overweight. This percentage declined to 12.4% in children breastfed 3-6 months, 11.3% in children breastfed 6-11 months, and 11% in children breastfed for at least 12 months. For children without maternal data, this association held up even after logistic regression controlling for gender, ,ethnicity, and birth weight. When maternal data were added, however, the associations were not significant except for white children in whom breastfeeding at least 12 months was associated with lower risk (by half) of overweight. Interestingly, the potential effects of breastfeeding did not alter the mean BMI, only the percentage of those who were overweight.

Reviewer comment: I like this study because of the size of the cohort and because it followed children until they were age 4. However, the limitations of the study, all noted by the authors, give me a little more pause than the authors suggest it should. The fact that the effects noted on the entire sample were reduced to only 1 subgroup (non-Hispanic whites) when controlling for maternal factors causes me concern that being able to control for a few other very important factors (ie, childhood diet) might attenuate the findings even more. The study is intriguing for what it suggests—that breastfeeding is associated with lower risk for overweight, but the limitations of the study make it difficult to call the findings conclusive. Breastfeeding is good for infants for many reasons, but it may not be a method to prevent childhood overweight.