by Marion Nestle
Mar 8 2011

Omega-3s in infant formula: time for a close look

The FDA just announced that it is planning to research health claims on infant formulas such as “supports brain and eye development.’’ The FDA wants to:

assess women’s understanding of and response to various statements on infant formula labels. The study results will be used to help the Agency to understand the role that certain types of statements on infant formula labels have in influencing formula choice….The study will focus on purchase choice, perceived similarity of the formula to breast milk, and perceived likelihood that the formula has certain health benefits.

Translation: the FDA thinks that claims for omega-3 fatty acids—DHA, in this case—on infant formulas mislead mothers into thinking the formulas are better than breast milk.

Despite lack of evidence for benefits, infant formula makers have been so successful in marketing the addition of these omega-3 fatty acids that you can hardly buy a formula without them.

I was in Ecuador a few months ago and saw this sign:

The idea?  Omega-3′s will make your kid smart.

My book, What to Eat, has a chapter on infant formula and baby food. Here are some relevant excerpts:

Infant formulas cause controversy and are endlessly contentious for three important reasons. Formulas are (1) largely unnecessary (most mothers can breast feed their infants), (2) not as perfect as breast milk for feeding babies, and (3) more expensive than breast feeding.

Breast milk is nutritionally superior to formula, but from a marketing standpoint it has one serious disadvantage: it is free. Beyond one-time purchases of breast pumps, storage bottles, or special clothing, nobody makes money from it.

For mothers who cannot, should not, or do not want to breast feed, formula is a socially and nutritionally acceptable substitute. But formula companies do not only promote formulas to mothers who must use formula. In subtle and not-so-subtle ways, they promote the use of formulas to all pregnant women and new mothers.

I go on to explain that because formula is the sole food for infants, its composition is highly regulated. Therefore, all infant formulas have the same composition, and all virtually indistinguishable.

Competition for market share explains why formula companies want to put distinctive nutrients in their formulas–especially nutrients considered “conditional.” A conditional nutrient is one that might have some benefits under some circumstances. Even if the health benefits are minimal or questionable, they can be used in advertising.

That is the principal reason why so many formulas now have fatty acids added—omega 6 arachidonic acid (ARA) and omega-3 docosahexaenoic acid —the same one that is in fish oil. These two fatty acids are normally present in breast milk, and there is some evidence, weak and questionable as it may be, that they support infant brain development and vision.

Formula makers got the FDA to agree that ARA and DHA are normal components of food (which they are) and, therefore, are Generally Recognized As Safe (GRAS). This means that companies could add ARA and DHA to infant formulas without having to prove that either of them really did anything useful or beneficial.

I then explain that the FDA apparently agreed to the GRAS petition with some reluctance, as indicated by its answer to the question, “What is the evidence that addition of DHA and ARA to infant formulas is beneficial?”

The scientific evidence is mixed. Some studies in infants suggest that including these fatty acids in infant formulas may have positive effects on visual function and neural development over the short term. Other studies in infants do not confirm these benefits. There are no currently available published reports from clinical studies that address whether any long-term beneficial effects exist.

My interpretation of the proposed research study is that the FDA thinks the addition of ARA and DHA may discourage mothers from breastfeeding and may unnecessarily cause them to buy more expensive formula.

If you agree, tell the FDA you think the study is a great idea, and the sooner it gets going, the better.

You can file comments at  Refer to Docket No. FDA–2011–N–0098]


I think this is an excellent idea – this is just another way in which consumers are being duped. I’ll be sure to file a comment.

  • Brandon
  • March 8, 2011
  • 5:17 pm

Arachidonic Acid is omega-6…

  • Cathy Richards
  • March 8, 2011
  • 7:19 pm

It’s a complex thing — the requirement for omega 3s has not been proven, partly because the amount in breastmilk varies based on mother’s diet (daily intake as well as any fatty acids released from fat stores from intake months ago!), so it’s hard to study.

To me, infant formula shouldn’t be allowed to make any health claims other than “meets federal requirements”. However, if I adopted a baby I would be shopping for a formula with omega 3…just in case! So it’s a complex issue.

  • megan
  • March 8, 2011
  • 7:41 pm

I’m writing my senior thesis on the marketing of functional foods, and I was wondering if you knew when exactly the “Formula makers got the FDA to agree that ARA and DHA are normal components of food.”

  • Brad
  • March 8, 2011
  • 9:04 pm

There is some good report on this by the Cornucopia Institute…..see this link:

  • Brad
  • March 8, 2011
  • 9:05 pm

I mean a good report….not “some good report”


This is great news, ’bout time FDA did something about this! USDA should step up to the plate too, WIC is wasting millions on this stuff.

See an excellent analysis of this issue by Zoe Neuberger at the Center on Budget & Policy Priorities:

This is great news, ’bout time FDA did something about this! USDA should step up to the plate too; WIC is wasting millions on this stuff.

See an excellent analysis of this issue by Zoe Neuberger at the Center on Budget & Policy Priorities:

  • Sarah
  • March 9, 2011
  • 3:47 am

Shouldn’t we be trying to make formula as close to breast milk as possible? I don’t think anyone chooses formula because they think it’s better for their baby. They choose formula because they can’t breastfeed, have trouble breastfeeding, for convenience, or because they want to keep smoking (not saying it’s right, but it’s a pretty big reason).

I think anything that is in natural breastmilk should be included in formula – but I’m not sure the companies involved should be able to make health claims about it unless backed up by rigorous science.

It’s funny but if I’d seen that sign in Ecuador I probably would have thought it was a special formula for over 3s.

  • Anthro
  • March 9, 2011
  • 9:11 am

I will register a comment.

It isn’t just formula. They are putting all sorts of things in dog food now–it’s getting difficult to buy anything for people or pets without all these very questionable additives. Time to pull out your newest book and start making my own dog food again.

The thing is, that if you EAT foods with Omega-3′s (easy enough to do), you don’t need all this crap–same goes for the dog!

The billboard seems to be suggesting that your child will be as cute and bright as the one in the picture if you use the formula. There might be a case for that if the breast feeding mother is malnourished, but wouldn’t it be cheaper to feed the mother adequately than to use expensive formula?

  • Sheila
  • March 9, 2011
  • 2:33 pm

Wonderful news indeed. Formula companies have done an amazing job of making the discussion about which formula is best, instead of comparing formula to breast milk. I’ve been researching the composition of major infant formula brands and there really isn’t much variation among the products!

Marion Nestle
  • Marion
  • March 9, 2011
  • 2:44 pm

@Brandon: Thanks very much for the correction–the first error I’ve found in What to Eat since it came out in 2006. I see what happened. I confused ALA (omega-3 alpha linolenic acid) with ARA (omega-6 arachidonic acid). It’s correct in other places in the book, but not this one. Apologies and thanks again.

  • Cathy Richards
  • March 9, 2011
  • 5:27 pm

Re: if it’s in breastmilk, it should be in formula — on the surface that sounds correct, however in practice it is difficult to do (a few reasons to follow) and not the actual goal.

The actual goal is to mimic the effect of breastmilk. Copying “ingredients” won’t work because so far we haven’t been able to copy many of the “ingredients” of breastmilk that affect how it is digested and how the nutrients are utilized. Eg. Lipases that affect fat digestion (a key issue with omega 3s), immunoglobulins dependent on the mother’s environmental exposure to pathogens, lactoferrin that affects gut flora and thus the absorption and in vitro production of fatty acids, etc etc etc.

Why is it difficult to replicate breastmilk? Just using omega 3′s as an example — breastmilk content of omega 3 reflects the mother’s daily intake of all fats, her fat store mobilization (historical fat intake), variations in fat content based on age of child (composition is not the same on Day 3 as on month 15), time of day, temperature (in hot weather, breastmilk is automatically higher in water content) and duration of breastfeeding session (foremilk vs hind milk).

Infant formula is basically milk, sugar, oil, and vitamins. It’s “Ensure” for babies. That will never be able to replace the phenomenal individualization of the breastfeeding mother and baby.

I disagree that infant formula is highly regulated. The GRAS system was canabalized in the late 1990′s. It has become a notification system only. The FDA refused GRAS for DHA and ARA made by Martek Bioscience in the late 1990′s because “independent studies” showed areas of concern (liver and spleen weights in rats higher, blood abnormalities) The GRAS system was an approval process and it became a notification system only. As a notification system, Martek stated its belief that their oils were GRAS based on Martek studies and the FDA had no questions. Thus, the whole GRAS process does not reflect independent study nor approval by the FDA. I believe Martek oils would not have been approved, if the FDA GRAS system had not changed.

Most of the components in infant formula are genetically engineered. Since the FDA believes that that which is genetically engineered is equivalent to the natural components, there is no labeling. Lactoferrin is a prevalent component of human milk, cow’s have virtually no lactoferrin. Human lactoferrin is being made through recombinant processes and transgenics of cows. Regulation is after-the-fact based on faith that the industry keeps records of problems. Thus making infants guinea pigs to industry. I believe all lot more should and could be done.

  • James Akre
  • March 10, 2011
  • 4:34 pm

I am grateful for this post and the regular contributions you make for the improved health and nutrition of the young of our species, who continue to be routinely fed on formula that is based on the milk of an alien species. But please don’t say that breast milk is free.

While we occasionally speak about the money breastfeeding saves, we mostly ignore what breastfeeding costs. Breast milk is most assuredly not free. I would start by describing it as priceless, even as breastfeeding itself has at least three price tags directly attached: a mother’s time (which far too many people erroneously consider to be on the house), the energy cost of producing milk (up to 500 kcal a day that need to come from somewhere) and the opportunity cost.

There is no difficulty recognizing the first two tags, which are an altogether spectacular bargain when we consider the payback in terms of positive lifelong consequences for children, mothers and thus the entire society. But the third one may not be so familiar. I’m borrowing from economic theory where “opportunity cost” refers to the cost of something in terms of an opportunity forgone – for example mothers who must choose between staying at home with their children and returning to paid employment outside the home to meet their families’ financial needs. As we all know from personal experience, there really is no such thing as a free lunch.

James Akre
Geneva, Switzerland

[...] There is much research being done about omegas and the way it helps the brain develop for example. FoodPolitics has a good example article on how omegas can help. This is just one example of many that constitute [...]

The Department of Labor will also be a player on issues related to breastfeeding. The Department created a page with information about breastfeeding at work ( and requested information from the public in December ( about how to implement a requirement under the Affordable Care Act related to providing reasonable space and time for certain nursing mothers to express breast milk. My colleagues at the Public Health Law Network recently produced a fact sheet explaining how health care reform is expected to make it easier for nursing moms to pump at work. ( This should help provide more women with the conditions that would enable them to choose to breastfeed.

[...] Read article by Dr. Marion Nestle from her blog here [...]

  • Mike Pescatore
  • April 21, 2011
  • 12:02 pm

GRAS ingredients should not be allowed to be added to infant formulas. I have to make a few corrections to the article above. First off, DHA and ARA found in human milk are identical to the molecules of DHA/ARA found in DHASCO/ARASCO. The difference among the two sources is the triacylglycerol structure. DHASCO/ARASCO molecules are attached in all three positions among the glycerol backbone, in some cases occupying two or three of the available positions. If we look at palmitic acid in human milk, we find it predominantly occupies the middle position(SN-2) of the triacylglycerol. After digestive processes, this specific position provides optimal absorption. If we place the palmitic acid in the outer positions(SN-!,SN-3), palmitic may form insoluble soaps with calcium and magnesium. This may result in constipation and malabsorption issues as well as loss of calcium retention. As you may see, the position of the fatty acid molecule will affect the metabolism and functionality. Right away, we can question why such sources of fats(palm olein) would be allowed to be added to the sole source of infant nutrition. DHA and ARA are very powerful bioactive fatty acids. Researchers are aggressively trying to understand how they exert numerous cellular effects. They play a role in cholesterol homeostasis, immune function, protein signaling and so on. The point is, how can one say with any certainty that altering their triacyglycerol structure will not exert adverse effects. This is exactly what science predicts-Manufacturers were allowed to structurally alter triacyglycerols of DHA and ARA and supplement into infant formulas. Today’s infants are part of an experiment and the manufacturers are solely responsible for the safety.

Leave a comment