If breastfeeding reduces obesity why did obesity soar as breastfeeding rates rose?


There’s a new paper about breastfeeding and obesity, Infant Feeding and Weight Gain: Separating Breast Milk From Breastfeeding and Formula From Food, being highlighted in the mainstream press. It claims to show that breastfeeding reduces obesity.

Really? Then why did obesity soar as breastfeeding rates rose?

[pullquote align=”right” cite=”” link=”” color=”” class=”” size=””]It’s almost like breastfeeding has no relationship with obesity at all![/pullquote]

Because breastfeeding does not reduce obesity. The claim is a product of the motivated reasoning that characterizes nearly all breastfeeding research.

Let’s take a look at what has been happening with obesity rates in the US:


As this graph from the CDC shows, US adult obesity has risen at an astounding rate. The prevalence of obesity has more than tripled from 10.7% in 1960 to 35.9% in 2010.

What was happening to breastfeeding rates at the same time?


I added the green line for breastfeeding rates. You can see that the rate of breastfeeding rose dramatically during that time. That DOESN’T mean that breastfeeding causes obesity, but it does show that it doesn’t prevent it.

What about childhood obesity? Perhaps there’s a benefit in children that we haven’t seen yet in adults.


Obesity has risen dramatically in every age group from 2 to 19 years even though breastfeeding rates were soaring.

How does childhood obesity vary by race?


That’s curious. Though Hispanics have breastfeeding rates comparable to or higher than white babies in the US, they also have the highest rate of childhood obesity.

It’s almost like breastfeeding has no relationship with obesity at all!

So how can breastfeeding researchers claim with a straight face that breastfeeding reduces obesity?

The authors of the new paper acknowledge that the connection between breastfeeding and obesity is tenuous at best and possibily non-existent:

Breastfeeding has many established benefits for maternal and child health, but its impact on obesity is unclear. In a meta-analysis of 113 studies, it was found that breastfed infants have a 26% reduced risk of obesity later in life; however, considerable heterogeneity was observed, and the association was attenuated among high-quality studies.

No matter. Breastfeeding researchers really, really want to claim that breastfeeding prevents obesity so they’ve undertaken a series of mental gymnastics to reach that conclusion. It involves creating the new diagnosis of infant “obesity” and recalibrating infant growth curves downward. I wrote about those mental gymnastics when critiquing a previous paper published by members of the same group that published this one.

Infant “obesity” was defined as more than 2 standard deviations from the mean of infant weight for length.

The Obesity Medical Association defines obesity as:

a chronic … disease, wherein an increase in body fat promotes adipose tissue dysfunction … resulting in adverse metabolic, biomechanical, and psychosocial health consequences.

As far as I can determine, there’s no evidence that infants whose weight for length is more than 2 standard deviations from the mean are suffering from any adverse metabolic or biomechanical consequences. So infant “obesity” may not be obesity at all.

Furthermore, this “diagnosis” is only made possible by a change in infant growth charts.

Why were infant growth charts changed? Previous standards evaluated growth based predominantly on formula fed infants since most infants were formula fed at the time they were developed. Many breastfed infants were diagnosed as underweight using these charts. Breastfeeding advocates claimed that it was wrong to evaluate breastfed infants using formula fed infants as the standard.

They had a point, but it’s not clear that it was a valid one. It’s based on the assumption that every breastfed infant is fully fed when the reality is that breastfeeding has a significant failure rate and some breastfed babies are actually underfed. Far fewer babies receiving formula are underfed since they can eat until satiety instead of merely until the milk runs out.

The WHO charts purportedly show “how infants and children should grow rather than simply how they do grow.” But they don’t measure how infants “should” grow, they measure how breastfed infants, including underfed infants, grow. It’s a classic example of the naturalistic fallacy: if something is a certain way in nature, that’s how it ought to be. But that’s makes as much sense as constructing a child growth chart including those with rickets to evaluate contemporary children who have easy access to calcium and vitamin D.

The conclusions of the new paper must be considered in light of this information.

The authors report:

Among 2553 mother-infant dyads, 97% initiated breastfeeding, and the median breastfeeding duration was 11.0 months. Most infants (74%) received solids before 6 months. Among “exclusively breastfed” infants, 55% received some expressed breast milk, and 27% briefly received formula in hospital. Compared with exclusive direct breastfeeding at 3 months, all other feeding styles were associated with higher BMIzs: adjusted β: +.12 (95% confidence interval [CI]: .01 to .23) for some expressed milk, +.28 (95% CI: .16 to .39) for partial breastfeeding, and +.45 (95% CI: .30 to .59) for exclusive formula feeding. Brief formula supplementation in hospital did not alter these associations so long as exclusive breastfeeding was established and sustained for at least 3 months. Formula supplementation by 6 months was associated with higher BMIzs (adjusted β: +.25; 95% CI: .13 to .38), whereas supplementation with solid foods was not. Results were similar for weight gain velocity.

CONCLUSIONS: Breastfeeding is inversely associated with weight gain velocity and BMI. These associations are dose dependent, partially diminished when breast milk is fed from a bottle, and substantially weakened by formula supplementation after the neonatal period.

The implication is that babies who receive anything other than breastmilk directly from the breast are “overfed,” because the authors assume that all breastfed babies are fully fed.

But the reality that a substantial proportion of women (up to 15% of first time mothers) will be unable to produce enough breastmilk to fully nourish their babies. Therefore some breastfed babies are almost certainly being underfed. Those babies are more likely to need supplementation and early introduction of solids in order to thrive. In other words, the supplemented babies may be the normal ones, and the breastfed babies may be underweight.

The authors insist:

This study confirms that sustained and exclusive breastfeeding is associated with favorable anthropometric outcomes during infancy…

But that’s not what the study shows at all. It merely shows that sustained and exclusive breastfeeding is associated with LOWER anthropometric outcomes; the authors assume lower is better but it may actually be worse.

The bottom line is that if breastfeeding doesn’t reduce obesity in adulthood and it doesn’t reduce obesity in childhood, there is no practical import to reducing “obesity” in infancy.

Let me amend that: there’s no practical benefits to babies, children and adults, but there is a practical benefit for breastfeeding researchers. By making up “benefits” of breastfeeding, they justify their cult-like belief that breast must be best.