Skip to content
Breastfeeding MomHealthy Babies. Confident Moms.

Breastfeeding basics

Breast Milk Antibodies: What Science Shows When Baby Is Sick

By Breastfeeding Mom Editorial TeamUpdated 10 min read

You may have seen the claim that when your baby gets sick, their saliva tells your breast to make custom antibodies within hours. It’s a lovely idea, and part of it has a basis in real research. But when you go back to the original studies, they show something more modest and in some ways more interesting.

This guide separates what has been measured from what has been assumed, so you know what your milk does and doesn’t do for a sick baby. For the bigger picture on what’s in milk, see our guide to what breast milk is made of.

How antibodies get into your milk

The main antibody in milk is secretory IgA (sIgA). In a 1982 study in the Journal of Pediatrics, about 90% of the total IgA in human milk over the first year was in secretory form. IgG and IgM are also present, in much lower amounts.

A 2010 review from the University of Oslo explains the route. IgA in milk reflects what your gut and airway lining have encountered, so milk antibodies are aimed at germs in your own environment, which are also the ones your baby is likely to meet.

A 2021 review in Cell from the Ragon Institute of MGH, MIT and Harvard notes that babies receive maternal antibodies both ways: IgG through the placenta, and antibodies through milk. It says the role of placental IgG is established, while less is known about how milk antibodies protect. As described in a 2015 Harvard and Yale paper, sIgA works in the baby’s gut in several ways, including keeping germs from binding to the lining and clumping bacteria and viruses together.

Does breast milk change when your baby is sick?

Three small studies we read looked at this directly, and they measured different things. This table and the “our read” notes are our own comparison of their published data.

Study Who Milk cells sIgA
Riskin, Bnai Zion Medical Center, Israel, 2012 Mothers of 31 babies up to 3 months old in the hospital with fever, plus 20 healthy controls White blood cells fell from a median of 5,655 per ml in the acute illness to 2,122 after recovery No significant change
Bryan, Flinders University, Australia, 2007 36 mothers of babies hospitalized with bronchiolitis, 63 healthy controls Viable cells 1.3 million vs 0.3 million per ml Not reported in the abstract, which covered cells and cytokines
Hassiotou, University of Western Australia, 2013 21 mother-baby pairs and 63 samples, with infections in mother or baby White blood cells 0 to 2% of milk cells when healthy, higher with infection Modest rise with infection in mother or baby combined; not significant when only the baby was sick

Our read: Milk cells respond more clearly than antibodies do. In Riskin’s data, by our calculation, white cells were about 2.7 times higher during illness than after recovery, and macrophages about 4 times higher (1,220 vs 300 per ml). But sIgA, the antibody people usually mean, did not change. These babies were sick enough to be hospitalized, so we don’t know how much a common cold would move these numbers.

The fourth study often cited, from Harvard and Yale researchers in 2015, tracked 30 mother-baby pairs among the Toba people of Argentina, with 110 milk samples. Higher lactoferrin (a protective protein) went with more infant symptoms, suggesting a response to illness. Higher sIgA went with fewer symptoms, suggesting prevention rather than response. Neither was related to symptoms at the time of the visit. The authors themselves caution that monthly sampling and mothers’ recall limit any conclusions about timing, and that some babies may simply be sick more often than others.

Claim 1: “White blood cells jump to 94% when baby is sick.” That number is real but gets misapplied. In Hassiotou’s paper, the highest figure (93.6%) came from a mother with mastitis, an infection of the breast itself. When only the baby was sick, the authors describe a “small but significant” rise in white cells (P = 0.046) and call it the weakest response of any infection type. The healthy baseline was 0 to 2%.

Claim 2: “Your baby’s saliva tells your body what antibody to make.” This comes from a possibility, not a finding. In the discussion section of the same paper, the authors suggest that as milk flows backward in the ducts after a let-down, saliva and germs from the baby could reach the breast and trigger a local response. They didn’t test this, and they didn’t measure any germ-specific antibodies. Later work adds indirect support for a backward flow: a 2019 University of Manitoba analysis of 393 mother-baby pairs found its milk bacteria data supported the idea that a baby’s mouth influences milk’s bacteria. That is about bacteria, not made-to-order antibodies. We found no study showing that a mother’s milk made an antibody matched to her baby’s infection because of a saliva signal. What would change this conclusion is a study testing milk for germ-specific antibodies before and during a baby’s illness, in a mother who hadn’t been exposed herself.

One milk and saliva interaction has been shown in the lab, but it works differently. A 2015 University of Queensland study found that mixing newborn saliva with milk generated hydrogen peroxide at levels that slowed the growth of Staphylococcus aureus and Salmonella in the lab.

Claim 3: “Milk has more antibodies when you are sick, so a sick mother’s milk is stronger.” There is a measured effect, but it is small relative to natural variation. In Hassiotou’s statistical model, mature milk from healthy pairs averaged 860.7 µg/ml sIgA, and infection in either mother or baby added 169.9 µg/ml. By our calculation, that is about a 20% rise. Split by type of infection, the difference when only the baby was sick was 88 µg/ml and not statistically significant (P = 0.63). In the same combined model, colostrum was 1,288.5 µg/ml above that baseline, or about 2.5 times the level in mature milk. In a Gambian study of 1,590 samples from University of Glasgow researchers, median IgA differed more than that between seasons alone (0.853 vs 0.518 g/L). The Hassiotou study was also funded in part by a breast pump maker, Medela, which the paper discloses.

Concentration isn’t dose. By our calculation from those averages, a 5 ml (1 teaspoon) feed of colostrum holds about 11 mg of sIgA, while 1 oz (30 ml) of mature milk holds about 26 mg. Colostrum is concentrated, and colostrum’s other components matter too, but later milk keeps delivering antibodies through sheer volume.

What vaccine studies add

Vaccination gives researchers something infections rarely do: a known start date. That is why these studies are cleaner.

  • COVID-19, Israel, 2021. Perl and colleagues at Shamir Medical Center followed 84 women and 504 milk samples after two mRNA doses. 61.8% of samples had specific IgA at 2 weeks, 86.1% at week 4 and 65.7% at week 6. Specific IgG reached 97% of samples at weeks 5 and 6.
  • Infection vs vaccination, US, 2022. At the University of Rochester, 77 parents gave milk samples. Infection brought a variable, mostly IgA response (12 of 45 had low or no response) that held out to 90 days. Vaccination brought a more uniform, mostly IgG response that began to decline by 90 days after the second dose. Milk from both groups neutralized live virus in the lab.
  • Reaching the baby, US, 2022. Among 30 women studied by UMass Amherst and Baystate Medical Center, specific IgG and IgA turned up in 33% and 30% of infant stool samples.
  • Pooled data, 2022. A Texas Tech review of 18 cohort studies found 70% of samples positive for IgA and 91% for IgG after a second dose, and called for research on actual infant protection.
  • Flu, Bangladesh. In a randomized trial with Cincinnati Children’s researchers, 340 pregnant mothers were assigned flu or pneumococcal vaccine, and milk from 57 was tested. Flu-specific IgA stayed higher in the flu vaccine group for at least 6 months.

What these studies do not show: finding antibodies in milk or in a baby’s stool doesn’t prove fewer infections. None of the COVID-19 papers above measured whether breastfed babies got sick less because of them. The CDC’s wording reflects this: milk antibodies could help protect babies. Vaccine recommendations get updated, so ask your doctor what applies to you now.

How long does the protection last?

Two separate questions are often blended here.

How long does milk keep making antibodies? As long as you breastfeed, as far as studies have looked. IgA levels fall in the first weeks and then stabilize. Hassiotou’s data show sIgA leveling off and then rising in later lactation. In a North Carolina State University study of 19 women, IgA rose between 11 and 17 months postpartum.

How long does your baby stay protected? The data are less clear, and they don’t fully agree.

Study Design Finding
PROBIT, Belarus, led from McGill University, 2001 Cluster-randomized, 17,046 pairs First-year gut infections 9.1% vs 13.2% (adjusted odds ratio 0.60). Respiratory infections no different (39.2% vs 39.4%)
Generation R, Netherlands, 2010 Cohort, 4,164 babies Exclusive for 4 months, then partial: lower risk of upper respiratory, lower respiratory and gut infections to 6 months, and lower respiratory infections at 7 to 12 months. Partial breastfeeding alone: no significant reduction
Dundee, Scotland, 1990 Cohort, 674 babies Breastfed 13 weeks or more: less gut illness through 52 weeks, including beyond the end of breastfeeding. Under 13 weeks: similar to bottle-fed babies

Our read: Protection against gut infections has the most support, and PROBIT matters because it is randomized. By our calculation, the 4.1-percentage-point gap means about 1 fewer baby with a gut infection for every 24 babies at the sites promoting breastfeeding. The respiratory results conflict: PROBIT saw no difference, while Generation R did. PROBIT compared hospitals promoting breastfeeding, not breastfed and non-breastfed babies, and cohort studies can’t fully rule out differences between families. These studies measure breastfeeding overall, not antibodies, and milk has many other protective parts. So they can’t tell you how much of the benefit is due to sIgA.

Mouse studies suggest early IgA could have long-term effects on the gut, and a 2019 University of Pittsburgh study tied maternal IgA to protection from necrotizing enterocolitis in preterm babies and confirmed the role in mice. Those findings are promising, but they don’t show lifelong protection in people.

What this means when your baby is sick

This is our practical summary of the evidence above.

  • Keep feeding as usual if your baby will. The CDC says sick babies need fluids and breast milk is the best option. Hydration and comfort are the benefits you can count on.
  • You don’t need to do anything special to “trigger” antibodies. We found no evidence that extra skin contact or added feeds raise antibodies against your baby’s germ.
  • If you’re sick too, keep going with hygiene. See our guide to breastfeeding while sick.
  • If you can’t nurse, express. Expressed milk can be given by cup, syringe or bottle.
  • If you can’t breastfeed at all, or you have stopped, you haven’t caused your baby’s illness. Babies catch infections whatever they’re fed. All 30 breastfed babies in the Argentine study had illness during follow-up. For the wider research on the benefits of breastfeeding, and how to weigh them, our breastfeeding basics section can help.

When to call your pediatrician

Breast milk doesn’t replace medical care. The AAP says to call if your baby is younger than 3 months with a temperature of 100.4°F (38.0°C) or higher, or if a child seems to be getting worse or still acts sick after a fever comes down. Also call if your baby:

  • Is feeding much less than usual or has fewer wet diapers
  • Is unusually sleepy, floppy or hard to wake
  • Is breathing fast or with effort, or has repeated vomiting or diarrhea

If you have a fever with a red, painful area on your breast, call your own doctor, since it may be mastitis. A lactation consultant (IBCLC) can help you protect your supply if your baby is too ill to nurse well.

Frequently asked questions

Does pumped or frozen breast milk still have antibodies?

The CDC says expressed milk can be given to sick babies by cup, syringe or bottle. The studies we read did not test home freezing directly. A 2013 review from Cincinnati Children's says long-term storage and freeze-thaw cycles can degrade some milk components, and reports that pasteurizing donor milk significantly reduces secretory IgA.

Does formula contain antibodies?

Formula doesn't contain your antibodies. The AAP describes breast milk as delivering a mix of nutrients and antibodies that can't be found in any other source. Formula is still a safe way to feed a baby, and your pediatrician can help if you are combining the two.

Should I keep breastfeeding if I am the one who is sick?

In most cases, yes. The CDC says mothers with flu or COVID-19 can keep breastfeeding with precautions, that current evidence suggests the COVID-19 virus isn't spread through milk, and that milk contains antibodies and other factors that can help protect your baby. Wash your hands and consider a mask while feeding.

Does breast milk treat my baby's illness?

We found no study showing that breast milk treats an active infection. It offers fluids, nutrition and comfort, and it is linked to a lower risk of some infections. A sick baby may still need medical care, so don't delay calling your pediatrician.

Are there more antibodies in colostrum?

Yes, by concentration. A 2013 University of Western Australia study found secretory IgA was much higher in colostrum than in mature milk, though babies drink far less colostrum, so the amount they receive per feed is not as different as the concentration suggests.

Sources

  1. Hassiotou et al. 2013, Clinical & Translational Immunology: Maternal and Infant Infections Stimulate a Rapid Leukocyte Response in Breastmilkhttps://pmc.ncbi.nlm.nih.gov/articles/PMC4232055/
  2. Riskin et al. 2012, Pediatric Research: Changes in Immunomodulatory Constituents of Human Milk in Response to Active Infection in the Nursing Infanthttps://pubmed.ncbi.nlm.nih.gov/22258136/
  3. Breakey et al. 2015, Evolution, Medicine, and Public Health: Illness in Breastfeeding Infantshttps://pmc.ncbi.nlm.nih.gov/articles/PMC4334701/
  4. Perl et al. 2021, JAMA: SARS-CoV-2-Specific Antibodies in Breast Milk After COVID-19 Vaccinationhttps://pmc.ncbi.nlm.nih.gov/articles/PMC8042567/
  5. Young et al. 2022, JAMA Pediatrics: Human Milk Antibody Induction, Persistence, and Neutralizing Capacityhttps://pmc.ncbi.nlm.nih.gov/articles/PMC8581794/
  6. Narayanaswamy et al. 2022, Obstetrics & Gynecology: Neutralizing Antibodies and Cytokines in Breast Milkhttps://pmc.ncbi.nlm.nih.gov/articles/PMC8759542/
  7. Kramer et al. 2001, JAMA: Promotion of Breastfeeding Intervention Trial (PROBIT)https://pubmed.ncbi.nlm.nih.gov/11242425/
  8. Duijts et al. 2010, Pediatrics: Prolonged and Exclusive Breastfeeding Reduces the Risk of Infectious Diseases in Infancyhttps://pubmed.ncbi.nlm.nih.gov/20566605/
  9. Howie et al. 1990, BMJ: Protective Effect of Breast Feeding Against Infectionhttps://pmc.ncbi.nlm.nih.gov/articles/PMC1661904/
  10. Perrin et al. 2017, Maternal & Child Nutrition: Human Milk Composition in the Second Year Postpartumhttps://pmc.ncbi.nlm.nih.gov/articles/PMC6866067/
  11. CDC: Influenza (Flu) and Breastfeedinghttps://www.cdc.gov/breastfeeding-special-circumstances/hcp/illnesses-conditions/flu.html
  12. CDC: COVID-19 and Breastfeedinghttps://www.cdc.gov/breastfeeding-special-circumstances/hcp/illnesses-conditions/covid-19.html

This guide is general information, not medical advice. For your own situation, talk to your doctor, midwife, pharmacist or an IBCLC lactation consultant. Read our medical disclaimer and find breastfeeding support near you.

Related articles