PCOS and Milk Supply: What the Breastfeeding Research Actually Shows

Does PCOS cause low milk supply?

A single 2000 case series of three mothers is the origin of the widely repeated claim that PCOS causes low milk supply, and twenty-five years of follow-up research since has produced a far more mixed picture than that certainty suggests. Exclusive breastfeeding rates in PCOS, across the studies that followed, run from meaningfully lower than the general population to statistically identical to it, depending on the study and, more than anything else, on body weight. If you have searched this while exhausted and worried you are failing at feeding a baby the same way you struggled to conceive one, the honest evidence is more forgiving than most of what ranks for this question.

What the actual studies found, side by side

Four studies, spanning 1993 to 2026, measured different things in different populations, and no two agree closely enough to produce one clean percentage.

Read across the row, and the pattern is not “PCOS lowers milk supply.” It is closer to: a gap shows up early and inconsistently, it is frequently gone by three to six months, and the largest and most recent study found it concentrated almost entirely in women carrying a higher BMI, not in PCOS on its own. That last point matters enough to repeat: in the 2026 cohort, normal-weight women with PCOS breastfed exclusively at essentially the same rate as women without PCOS. This sits alongside the rest of our PCOS pregnancy and fertility coverage, where the same pattern — PCOS raising a risk on average while leaving many individuals unaffected — shows up repeatedly.

The insulin theory: real mechanism, wrong population to prove it in

A 1993 study of 33 women with insulin-dependent (type 1) diabetes against 33 controls found clearly delayed lactogenesis and measurably lower milk intake in their infants during the first two weeks — the clearest human evidence that insulin’s role in lactation is real, not just theoretical, and it was worse in the women with poorer metabolic control. Insulin is one of a small handful of hormones the mammary gland needs to complete lactogenesis — the transition from a pregnant breast to one actively making milk — alongside progesterone withdrawal and sustained prolactin. That sequencing was worked out in the physiology literature, not in PCOS specifically. The diabetes study above is a different population and a different disease, and it is fair to say so plainly: nobody has run the equivalent study in PCOS. The insulin-resistance theory in PCOS is a reasonable extrapolation from a related pathway, not a result measured in PCOS itself. The pregnancy risks PCOS actually raises covers the same insulin-resistance mechanism working on gestational diabetes, where the trial evidence is larger.

The “insufficient glandular tissue” theory

The theory with a name attached is this one, first proposed from three case reports in 2000: that PCOS can interfere with mammogenesis — the growth of milk-making glandular tissue during pregnancy — leaving too little tissue to produce a full supply regardless of how well the baby feeds or how often. It is the theory most repeated online, and it is also the one with the most direct, if still limited, data behind it. A follow-up study of 186 women with PCOS measured actual bra-size change across pregnancy and found it mattered: women whose breasts did not visibly enlarge during pregnancy breastfed for a shorter time, and — this is the part that ties the theories together — they also had higher fasting insulin, higher BMI, and higher triglycerides than women whose breasts did grow. Androgen levels and whether the mother had taken metformin in pregnancy showed no relationship to breast growth or breastfeeding duration in the same data.

Progesterone, androgens, and where the evidence actually stops

The same 2008 study that found the early breastfeeding gap also measured gestational androgens directly, and the result complicates the tidy “high androgens cause low supply” version of this story: dehydroepiandrosterone-sulphate (DHEA-S) showed a weak negative association with breastfeeding rate, but testosterone, androstenedione, sex-hormone binding globulin, and free testosterone index showed no association at all. One androgen, weakly, is not the same as androgens broadly. Progesterone withdrawal after delivery is a documented, necessary trigger for lactogenesis in every studied population, PCOS or not — but no study has measured whether PCOS pregnancies withdraw progesterone differently. The 2023 international guideline for PCOS, which this site checks its clinical framing against, does not include a recommendation on breastfeeding or milk supply at all. That is not a criticism of the guideline; it is a fair description of how early this specific question still is.

Does PCOS cause oversupply, too?

Zero published studies have measured oversupply in PCOS specifically, even though it is the direction current search results almost never mention: some women with PCOS report the opposite problem to undersupply — an oversupply of milk, sometimes alongside forceful let-down and an overfull, uncomfortable breast. Lactation consultants describe seeing this pattern in PCOS clinically, and it is mechanistically plausible for the same reason undersupply is: if PCOS can disrupt the hormonal signaling that governs milk production in one direction, disruption in the other direction is not automatically ruled out. That research gap is itself informative. Every study in Table 1 above set out to test the undersupply hypothesis, which means oversupply has had essentially no chance to be studied, confirmed, or ruled out.

Who this genuinely does not seem to affect

The largest and most recent dataset — 653 women with PCOS against a comparison cohort of 63,927 — found no reduction in exclusive breastfeeding at all for women whose BMI sat in the general “normal” range. The earliest and smallest study to find a real gap found the same thing a different way: that gap was gone by three months. Neither of those facts guarantees your experience, but both cut directly against the framing that a PCOS diagnosis alone puts your supply at risk. What the research keeps landing on instead is metabolic status — insulin, triglycerides, BMI — as the more consistent thread than PCOS as a label. What changes for PCOS after pregnancy covers the wider postpartum picture, including a separate, larger dataset that found no PCOS-specific breastfeeding gap once education and prior births were accounted for.

If you are struggling right now: what actually helps

None of the four mechanisms in Table 2 change what the next practical step looks like, and none of them are things to try to fix yourself with a supplement or a specific dose — this article will not give you one, because reporting what has been studied is different from telling you what to take. An IBCLC (International Board Certified Lactation Consultant) is trained specifically to assess positioning, transfer, and true supply versus a baby who is feeding inefficiently, which look identical from the outside and require different fixes. A midwife, health visitor, or your baby’s pediatrician can usually refer you to one, and can also weigh the baby to see whether any of the red flags at the top of this page are actually present, rather than assumed. One RCT-embedded study did test whether a specific medication changed breastfeeding outcomes in PCOS: metformin taken through pregnancy showed no measurable effect on breast growth or breastfeeding duration compared with placebo. What metformin actually does in PCOS covers that trial and what the drug is and is not shown to help with, including in a future pregnancy — a question covered directly on our second-pregnancy planning page for anyone already thinking that far ahead.

You may see PCOS written as polyendocrine metabolic ovarian syndrome (PMOS), after a 2026 global consensus of more than 50 organisations renamed it. Nothing in the research above changes under either name — this article uses PCOS because that is still what most people search.