The studies for the "50 ml breastmilk per day is enough to show benefits" claim?

[jesuslovesme527]

I am 4 months postpartum and have been exclusively pumping during that time. At the many various moments I've considered stopping, I've tried to figure out how long I can stretch my freezer stash to still give my baby the much touted benefits of breastmilk, and I've seen the line that while breastmilk's benefits are dose dependent, as little as 50 ml of breastmilk might be enough (i.e. Kellymom). I had never seen any actual studies to support that claim, but I think I might have found at least some of the research that generated it, so I thought I would share here.

Furman L, Taylor G, Minich N, Hack M. The effect of maternal milk on neonatal morbidity of very low-birth-weight infants. Arch Pediatr Adolesc Med. 2003 Jan;157(1):66-71. doi: 10.1001/archpedi.157.1.66. PMID: 12517197.

Schanler RJ, Lau C, Hurst NM, Smith EO. Randomized trial of donor human milk versus preterm formula as substitutes for mothers' own milk in the feeding of extremely premature infants. Pediatrics. 2005 Aug;116(2):400-6. doi: 10.1542/peds.2004-1974. PMID: 16061595.

The TLDR: These are two studies on premature infants that concluded at least 50 mL per kg per day were needed to show a decrease in the rate of sepsis (and NEC) in very low birth weight infants.

So, as much as I wish it were not so, I do not think one can extrapolate at least from these studies that 50 mL of breastmilk is enough to provide whatever unique benefits that breastmilk may provide (note: any amount of breastmilk is of course still nutritious, but I think most people want to provide breastmilk for whatever immune support benefits formula cannot provide). For starters, it's 50 mL per kg, meaning an 11 lbs/5kg baby would need around 250 mL/8 oz breastmilk per day. Second, these studies looked only at a very specific population (low birth premature infants) and for a specific outcome (sepsis/NEC), so it's hard to generalize from this that the same amount would be enough or have other benefits for an older, healthy infant. Moreover, at birth and in the weeks after birth, breastmilk is first colostrum and then transitional milk, which both differ from mature milk in immunoglobulin makeup, for example.

That said, I have no medical or medical research background, so would be interested in hearing others' thoughts!

 

[gizmorazaar]

@jesuslovesme527 I don’t think you will find what you are looking for.

Research studies need a clear signal to measure, but for healthy full term babies it has long been difficult to demonstrate a clear breastmilk benefit at all. There is very little measurable difference in outcome between EBF and fully formula fed infants. So if the maximum difference is difficult to confirm, then an effect of 50 mls per kg is going to be a whole lot harder to see.

Researchers turn to edge cases when they need to maximize the signal. And in this case they were able to demonstrate that a benefit exists in sensitive populations. Thats sufficient to show that a benefit exists. Extrapolating beyond study parameters isn’t valid, but it’s likely that the benefit extends beyond what they can detect.

 

[jesuslovesme527]

@gizmorazaar This makes a lot of sense, thanks! I do wish there were studies that looked at outcomes for combination fed infants at various rates (I.e., 70% breast milk intake, vs. 50% vs. 30%, etc. And as compared with fully formula-fed and EBF) but you’re right that it’s hard enough to find well designed studies that look just at EBF vs fully formula-fed.

 

[gizmorazaar]

@jesuslovesme527 It’s not that the studies aren’t well designed. It’s that the signal is already too weak to slice into even smaller pieces.

Long term, there’s no confirmed difference at all. Short term it’s on the order of maybe one less illness, on average. So the outcome is statistical.

Think about how this study would need to be carried out. To compare 30% vs 70% breastmilk you need a population of moms who agree to measure every single feeding and stop/switch when they hit the limit. Yeah, that’s not happening.

Let’s say it did happen, you found a hundred unusually compliant and analytically minded moms. 50 of the babies had an average of 3.7 colds their first year, the other 50 had an average of 4.1 colds. That’s a pretty big difference but is it significant? Did they all have the same number of siblings in preschool? Was the signal even in the right direction? A few random extra colds could flip it. The conclusion of my fully made up scenario is almost certainly going to be no demonstrated effect.

 

[chriscrossan]

@jesuslovesme527 This is a fantastic post. Thank you for doing this research and sharing it with everyone so that mothers can make informed decisions.

 

[hman]

@chriscrossan Agree. There are so many myths about the magical benefits of breastfeeding. If it works for you, great. If not, there’s essentially no evidence that you’re not giving your child the “best”.

 

[jaenalyn]

@jesuslovesme527 I agree that the 50mls number is one that seemingly comes from thin air. Maybe intended to encourage low supply lactating parents, maybe intended to coerce parents who wished to cease lactation. Idk.

The best designed study of breastfeeding that I know of is the 2001 PROBIT cohort.

TL;DR the children who were in the promotion of breastfeeding arm had maybe one fewer GI viral illness and slightly lower rates of atopic dermatitis at 1 year. At 6.5yrs kids who were breastfed had maybe 3 points higher IQ scores.

Respiratory illnesses, Asthma, allergies, obesity, height, cavities, behavior issues, all the same.

Take a look at it, it has good study design, looks at a European country, with access to healthcare, maternity leave, clean water, and formula. I’d say it’s approachable and applicable for average western parents to use in decision making.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4052126/

If your reason to continue breastfeeding is a hope to prevent respiratory viral illnesses, it’s probably not gonna help.

Vaccines, masks, hepa filters, avoiding sick people, washing hands and avoiding daycare (sorry) is your best bets.

 

[javier_is_life]

@jaenalyn Also worthwhile pointing out that in the PROBIT study, to quote - "at 6 months the rates of both exclusive (7.9% vs. 0.6%) and predominant (10.6% vs. 1.6%) breastfeeding were low."

 

[teq]

@jaenalyn Yup I just had this convo with my baby’s doctor. I asked if breastfeeding would help prevent respiratory illnesses in my newborn since my older child attends daycare, and she said no breastfeeding wouldn’t help with that.

 

[gizmorazaar]

@jaenalyn I would personally consider a 3 point IQ difference to be sufficient reason to breastfeed. For a scale with a mean of 100, 3 points is a large shift.

However I don’t think it’s that simple or that large when other variables are factored in. We already know that in most studies of breastfeeding, other factors dwarf the effect of the milk itself. (Income, employment, culture, family support, etc). There have been subsequent studies that show that the effect gets smaller (though still positive) when maternal IQ is input, so part of that is genetic. I don’t think I’ve seen a study where dad’s IQ was also factored in.

There used to be a reported difference in obesity rates, but that was chipped away as other correlating factors were found to be more significant, and it’s now considered to be secondary. IQ may or may not hold up. My guess (just a guess, and I have my own biases) is that it is real and worthwhile but not as large as that study would suggest.

 

[clemi]

@gizmorazaar 3-5 standardized pts in an IQ test (for most tests like the WAIS which was used in that study) MAY move you into a new bracket (from average to above average), but I wouldn’t say that this study shows causation since it didn’t look like this analysis controlled for wealth, parent education, parent IQ, etc. I could be mistaken and missed that since the paper didn’t talk much about their analysis plan for that particular variable. IQ is very complex and it’s unlikely that something like breastfeeding could, by itself, impact IQ that much.

 

[gizmorazaar]

@clemi My point on the large shift was based on the principle that a small difference in the mean of a bell curve produces a large shift in the extremes. It’s not going to send little Janey to the top of the class but population wise the effect could be significant. I agree that the other variables are likely larger and may account for the entire difference.

I also recall seeing a report where the means were not statistically different but there was a significant difference in the number of low IQ children in the bf population. So that might suggest greater benefit in vulnerable or otherwise sensitized populations.

 

[clemi]

@gizmorazaar Check something called the Flynn effect of IQ, typically IQ scores increase over generations but the distribution of scores tends to remain the same. It’s a little unrealistic to think that breastfeeding is going to reduce rates of intellectual disability in our population. It also makes sense that less breastfed babies have low IQ, as IQ is correlated with many other developmental delays that can impact the success of breastfeeding.

 

[jaenalyn]

@clemi The first probit trial publication speaks better to the study design.

Two randomized groups of expectant mothers are enrolled, one is given breastfeeding classes before birth and breastfeeding consultant after birth. The longitudinal cohort data looks at the entire arm.

As @javier_is_life notes, even the intervention arm had relatively low rates of any breastfeeding at 6mo.

The hope of the study is to remove factors such as SES, maternal education, ability to lactate and so on.

 

[clemi]

@jaenalyn Random assignment is great but you need to check dropout— with such a small intervention retention I would imagine that perhaps the only women who were predominantly or exclusively breastfeeding by 6 months may have been similar (I.e., mostly SAHM from upper middle class families). I would say the authors are really stretching their findings here…

 

[lululala]

@gizmorazaar Also IQ is white supremacist pseudo-science and should not be a measure for anything.

 

[gizmorazaar]

@lululala It correlates with intelligence. g is better but has its own limitations. We can only study what we can measure, so we try to work around the flaws as best we can. Or else walk away. Some prefer ignoring anything we cannot accurately measure, but that is not my perspective.

 

[rayn91]

@jesuslovesme527 This study has lots going on, but in one part of it, they compare outcomes for infants never breastfed, partially, and exclusively. They don't seem to list an amount for the partially breastfed infants. The study talks a lot about breastfeeding influencing the microbiome. It also talks about delivery (breastfeeding vs drinking pumped milk).

https://www.sciencedirect.com/science/article/pii/S1931312820303504?via=ihub

*********** Excerpt:
For instance, R. mucilaginosa was present in 0% of never-breastfed infants compared with 7.3%, 14.4%, and 17.1% of infants that no longer, partially, and exclusively breastmilk fed, respectively (Figure 5C). Overall, it appears that breastmilk feeding mode primarily influenced the commonly shared bacteria in milk, whereas breastfeeding exclusivity primarily influenced those in the infant gut.

 

[jesuslovesme527]

@rayn91 I’ve seen this study and really appreciate how it also looks at mode of breastfeeding, which few seem to! But what I don’t understand (I am far from knowledgeable about the microbiome!) is how much this all matters— can you explain in laymen’s terms what is the effect on health outcomes caused by having 17.1% vs. 14.4% vs. 0% R. mucilaginosa, for instance?

 

[rayn91]

@jesuslovesme527 I don't know that we know yet. I've heard that there's some research that your microbiome can affect your immune system. I've also heard some stat about your microbiome being in greatest development until age 3. But I don't think we know long-term what that means, if anything. I don't think the research is there yet.