Skip to content
Breastfeeding MomHealthy Babies. Confident Moms.

Nutrition & health

Does Breastfeeding Reduce Breast Cancer Risk? What the Numbers Really Mean

By Breastfeeding Mom Editorial TeamUpdated 10 min read

Soft pink satin ribbons

You may have seen breastfeeding described as cutting breast cancer risk by a quarter, or even by half. The honest answer is less dramatic and more useful. There is a real and repeated link, but on an absolute scale the average difference is modest. And nobody can say that breastfeeding did or didn’t decide what happens to one particular person.

For a shorter summary, see our overview of the benefits of breastfeeding. If you couldn’t breastfeed, or chose not to, none of this is a verdict on you.

The study behind the 4.3% figure

Almost every article on this topic traces back to one paper. In 2002 the Collaborative Group on Hormonal Factors in Breast Cancer published a reanalysis in The Lancet. It pooled individual data from 47 studies in 30 countries, covering 50,302 women with invasive breast cancer and 96,973 without it.

Among women who had given birth, those with breast cancer had breastfed less over their lifetimes: an average of 9.8 months, against 15.6 months for those without it. After accounting for age, number of births and age at first birth, relative risk fell 4.3% (95% CI 2.9 to 5.8) for every 12 months of breastfeeding. That came on top of a 7.0% fall for each birth. The size of the decline showed no statistically significant difference by age, menopausal status or ethnic origin.

Two details are easy to miss. The figure is about total lifetime breastfeeding, not one baby at a time. And 4.3% is a relative figure, so it does not mean 4.3 fewer women in every 100.

How much does breastfeeding lower breast cancer risk?

Probably by a modest amount. By our rough calculation, 12 months of breastfeeding would mean about 6 fewer diagnoses per 1,000 women over a lifetime, if the whole link is causal. This calculation is ours, not the study’s. The inputs are SEER’s estimate that about 13.0% of US women are diagnosed in their lifetime (2021 to 2023 data). We combined that with the Collaborative Group’s 4.3% per 12 months, compounded each year.

Total time breastfeeding Estimated lifetime risk Women diagnosed per 1,000 Difference per 1,000
None 13.0% 130 Starting point
12 months 12.4% 124 About 6 fewer
24 months 11.9% 119 About 11 fewer
36 months 11.4% 114 About 16 fewer

Using the study’s confidence interval (2.9% to 5.8%), the 12-month difference ranges from about 4 to 8 per 1,000.

A shorter time window shrinks the number further. The National Cancer Institute puts the 10-year risk from age 40 at 1.55%. Applying 4.3% to that gives a difference of about 0.7 per 1,000 women over those 10 years.

Please read this table as a rough illustration of scale, not a forecast. Applying a pooled relative figure to a lifetime risk is a simplification, and it has three weaknesses:

  • The starting point isn’t a true zero. The 13.0% figure includes women who breastfed and women who never gave birth. The 4.3% came from women who had given birth.
  • It assumes the whole link is causal. The next sections explain why that may overstate it.
  • It is an average. It doesn’t apply to someone with a BRCA variant or a strong family history.

Even on these generous assumptions, most women in every row are never diagnosed, and most of those who are diagnosed would have been diagnosed anyway.

Tracing the “cut in half” claim

The Lancet paper is also the source of the most dramatic version you may see. It estimated that breast cancer by age 70 in developed countries could fall by more than half, from 6.3 to 2.7 per 100 women. It also estimated that breastfeeding could account for almost two-thirds of that fall.

That scenario isn’t about one or two babies. It models women having the births and lifetime breastfeeding once typical of developing countries. Cancer Research UK’s summary describes that as six or seven children, breastfed for about two years each. By our arithmetic, that is roughly 12 to 14 years of breastfeeding in total, a pattern few US families would choose or could manage.

Headlines also travel further than methods. A 2026 meta-analysis from Universitas Sriwijaya in Indonesia reported an odds ratio of 3.59 for breastfeeding under 12 months compared with longer. All 23 of its studies were case-control, and its authors warn about selection and recall bias. We would not read it as “short breastfeeding more than triples your risk.” We also left out a 2022 meta-analysis that PubMed lists as retracted, so check for that flag if you see it quoted.

Confounding: why the true effect may be smaller

Nobody can randomly assign mothers to breastfeed for years, so this evidence is observational. Women who breastfeed longer may also differ in number of births, age at first birth, education, income, contraceptive use, alcohol use and more. A 2023 umbrella review from Southern Medical University in China found 15% lower risk with ever breastfeeding (17 meta-analyses across cancers). It warned that residual confounding could affect many of these links.

A 2015 meta-analysis led from the Society for Applied Studies in New Delhi, with 98 estimates and a co-author from the World Health Organization, lets you watch that happen. The reading below is ours, from its subgroup table. It reported odds ratios, and lower means a stronger link:

How the studies were done Ever vs never breastfed: odds ratio
No adjustment for number of births 0.73
Crude adjustment for births 0.86
Fine adjustment for births 0.92
Fewer than 500 participants 0.50
1,500 or more participants 0.83
Case-control studies 0.77
Cohort studies, which record habits before diagnosis 0.85

Careful adjustment and larger studies bring the link closer to 1, which would mean no difference. The authors also found signs of publication bias. Treating these odds ratios roughly as risk ratios on a 13% baseline, an odds ratio of 0.73 would mean about 35 fewer diagnoses per 1,000, and 0.92 about 10 fewer. So the method can change the headline more than threefold. The link survives every row, which is why we say “probably.” Its size is much less certain.

Expert bodies read the same evidence with different levels of confidence:

Body What it says
National Cancer Institute PDQ “Solid evidence” of decreased risk; quotes 4.3% per 12 months
World Cancer Research Fund/AICR, 2018 Lactation “probably protects”; graded strong evidence, probable rather than convincing
American Cancer Society “May slightly lower” risk, especially after a year or more; hard to study in the US, where that duration is uncommon
American Academy of Pediatrics, 2022 Long-term breastfeeding “is associated with” protection against breast and ovarian cancers

Which breast cancers? The subtype picture

Breast cancer isn’t one disease. The most consistent finding is for hormone receptor-negative tumors, which are less common and often harder to treat. The picture for receptor-positive tumors, the most common kind, is unsettled:

Study Who Receptor-negative Receptor-positive
Nurses’ Health Studies, Harvard and German Cancer Research Center, 2019 199,514 women followed over time, 12,452 cancers ER-negative: hazard ratio 0.82 ER-positive: 0.99, no link
American Cancer Society-led review, 2015 27 studies, 36,881 cases ER/PR-negative: odds ratio 0.90; triple-negative: 0.78 (3 studies) No significant link overall in cohort studies
AMBER Consortium, Boston University and others, 2014 African American women: 3,698 ER-classified cases, 14,180 controls ER-negative: odds ratio 0.81 No reduced risk
University of Kentucky and Duke review, 2023 30 studies in its meta-analysis Triple-negative: relative risk 0.66 Luminal A 0.72; luminal B 0.71

Duration is a further loose end. A 2024 review from the Royal College of Surgeons in Ireland-Bahrain and the University of Pennsylvania found triple-negative odds 20% lower with ever breastfeeding (OR 0.80, 12 studies). Yet breastfeeding 12 months or more versus never showed no clear difference (OR 1.02, 95% CI 0.59 to 1.76, 7 studies), and the authors noted high inconsistency between studies. So “longer is better” is not established for this subtype.

Childbirth itself adds another wrinkle. A 2019 pooled analysis led from the University of North Carolina covered 15 cohorts, with 18,826 cases in women under 55. Compared with women who had never given birth, risk peaked about 5 years after a birth (hazard ratio 1.80). It crossed over to lower risk about 24 years after, and breastfeeding did not modify the overall pattern. That is not a contradiction. It compares mothers with non-mothers, while the Lancet figure compares mothers with each other. It’s also a relative number on a low baseline at young ages.

If you carry a BRCA variant

This is where studies disagree most, so ask a genetic counselor rather than relying on a single number.

Whatever the true answer, breastfeeding isn’t a substitute for the screening or risk-reducing options your care team discusses with you.

Why it might work

The mechanisms are still hypotheses. Researchers propose fewer lifetime menstrual cycles, lower exposure to ovarian hormones, and maturing of milk-making cells. A 2019 study from The Ohio State University found that abrupt weaning in mice led to more inflammation and abnormal cell growth than gradual weaning, but that’s a mouse model. It doesn’t show that weaning quickly raises cancer risk in people. If you do wean, our guide on how to stop breastfeeding covers comfort and timing.

If you can’t or don’t breastfeed

Please hold on to this part if the numbers above stung. Low supply, medicines, illness, work, past trauma and plain personal choice are all real reasons, and none of them is a moral failing. Here’s how we’d read the evidence:

And if you want to keep nursing but it’s hard, that’s a support problem rather than a personal one. Our breastfeeding support page lists where to find help.

When to call your doctor

At best, breastfeeding lowers average risk. It doesn’t rule out cancer, and milk-related lumps can make a new lump easy to dismiss. Contact your doctor if you notice:

  • A new lump or mass that doesn’t go away. The American Cancer Society notes most lumps aren’t cancer, but cancers can be hard or soft, painless or painful
  • Swelling, skin dimpling, or red, flaking or thickened skin on the breast or nipple
  • A nipple turning inward, or discharge that isn’t milk
  • Redness or a sore area that doesn’t clear with treatment for mastitis

Also ask for a referral to discuss your risk if close relatives had breast or ovarian cancer, or if you know of a BRCA variant in your family. You can have breast symptoms checked while breastfeeding, so don’t wait until you wean.

Frequently asked questions

How long do you have to breastfeed to lower breast cancer risk?

The studies we read don't show a clear cutoff. A 2015 review compared each duration with no breastfeeding. Odds were 7% lower with under 6 months, 9% lower with 6 to 12 months and 26% lower with more than 12 months. The largest links tend to appear with longer total time, and those figures are associations.

Does pumping count the same as nursing?

We don't know. The studies we read recorded breastfeeding by duration, and none of them separated feeding at the breast from pumping. So we can't tell you that pumped milk gives the same result, or that it doesn't.

If I breastfed, can I skip mammograms?

No. The US Preventive Services Task Force recommends screening every other year for women aged 40 to 74, without any exception for having breastfed. Ask your doctor whether your family history or genetics call for a different plan.

Does breastfeeding lower ovarian cancer risk too?

It may, but expert panels are more cautious here. A 2015 review linked more than 12 months of breastfeeding with 37% lower odds of ovarian cancer, while the World Cancer Research Fund grades the evidence as limited but suggestive.

Do these numbers tell me my own risk?

No. They are averages across large groups. Your own risk depends on age, family history, genetics, breast density and other factors, so a doctor can give you a better estimate than any population figure.

Sources

  1. Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and breastfeeding: collaborative reanalysis of 47 studies. The Lancet, 2002https://pubmed.ncbi.nlm.nih.gov/12133652/
  2. Chowdhury R et al. Breastfeeding and maternal health outcomes: a systematic review and meta-analysis. Acta Paediatrica, 2015https://pmc.ncbi.nlm.nih.gov/articles/PMC4670483/
  3. National Cancer Institute SEER: Cancer Stat Facts, Female Breast Cancerhttps://seer.cancer.gov/statfacts/html/breast.html
  4. National Cancer Institute: Breast Cancer Risk in American Womenhttps://www.cancer.gov/types/breast/risk-fact-sheet
  5. National Cancer Institute: Breast Cancer Prevention (PDQ), health professional versionhttps://www.cancer.gov/types/breast/hp/breast-prevention-pdq
  6. World Cancer Research Fund/AICR: Lactation and the risk of cancer (2018)https://www.wcrf.org/wp-content/uploads/2024/10/Lactation.pdf
  7. American Cancer Society: Lifestyle-related Breast Cancer Risk Factorshttps://www.cancer.org/cancer/types/breast-cancer/risk-and-prevention/lifestyle-related-breast-cancer-risk-factors.html
  8. Islami F et al. Breastfeeding and breast cancer risk by receptor status. Annals of Oncology, 2015https://pubmed.ncbi.nlm.nih.gov/26504151/
  9. Fortner RT et al. Parity, breastfeeding, and breast cancer risk by hormone receptor status: the Nurses' Health Studies. Breast Cancer Research, 2019https://pmc.ncbi.nlm.nih.gov/articles/PMC6416887/
  10. Kotsopoulos J et al. Breastfeeding and the risk of breast cancer in BRCA1 and BRCA2 mutation carriers. Breast Cancer Research, 2012https://pmc.ncbi.nlm.nih.gov/articles/PMC3446376/
  11. Nichols HB et al. Breast cancer risk after recent childbirth: a pooled analysis of 15 prospective studies. Annals of Internal Medicine, 2019https://pubmed.ncbi.nlm.nih.gov/30534999/
  12. American Cancer Society: Breast Cancer Signs and Symptomshttps://www.cancer.org/cancer/types/breast-cancer/screening-tests-and-early-detection/breast-cancer-signs-and-symptoms.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