Does Dairy Make PCOS Worse? The IGF-1 Question, Answered With Numbers

PCOS was renamed polyendocrine metabolic ovarian syndrome (PMOS) in 2026 by a global consensus of more than 50 organisations. Nothing about the dairy evidence below changed with it; this article uses PCOS because that is still what people search.

Does Removing Dairy Actually Change PCOS Markers?

The closest available evidence says yes — but only when dairy was removed alongside starch, not alone. Twenty-four overweight and obese women with PCOS followed an 8-week, ad libitum diet that removed both starchy foods and dairy products. By week 8, weight had fallen 8.61 kg (p < 0.001), fasting insulin had fallen 17.0 μg/mL, HOMA-IR had improved by 1.9 points, total testosterone had fallen 10.0 ng/dL (p = 0.008), free testosterone had fallen 1.8 pg/dL (p = 0.043), and Ferriman-Gallwey hirsutism score had dropped 2.1 points (p = 0.001) (Phy et al., 2015). A companion analysis of the same protocol in a smaller group found increased fat oxidation and reduced fasting insulin over the same 8 weeks (Pohlmeier et al., 2014).

Those are real, statistically significant numbers, and it is tempting to read them as “dairy makes PCOS worse.” That is a bigger claim than the trial can support.

What Does the Guideline Say About Dairy Specifically?

Nothing — and that absence is itself informative. The 2023 international PCOS guideline does not name dairy as a food to restrict, and a systematic review of dietary composition trials in PCOS found no diet varying by macronutrient or food-group composition outperforming any other for the condition’s core reproductive and metabolic outcomes (Teede et al., 2023; Moran et al., 2013). If an isolated, dose-controlled dairy-removal trial existed showing an independent effect, it would be the kind of finding these reviews are built to surface. It does not exist yet.

Is Cheese Good for PCOS, or Is It Different From Milk?

No trial has tested cheese separately from other dairy in PCOS, so any specific claim about cheese would be a guess dressed as a fact. What can be said honestly: the two PCOS intervention trials above removed dairy as a category — milk, cheese and yoghurt together — so they provide no evidence that solid, fermented dairy behaves differently from liquid milk in this population. Greek yoghurt and cottage cheese get a closer, separate look here, including what actually differs between dairy forms on protein and sugar content, which is a more answerable question than “is cheese good for PCOS” on its own.

Does Milk Specifically Cause PCOS Acne?

The strongest single piece of evidence linking dairy to a PCOS-relevant symptom is about acne, not about PCOS itself. A prospective cohort of 6,094 teenage girls found that higher milk intake — two or more servings a day versus less than one a week — was associated with more acne, and the association held for total milk (prevalence ratio 1.20, p < 0.001), whole milk (1.19), low-fat milk (1.17) and skim milk (1.19) (Adebamowo et al., 2006).

Two things make that finding worth taking seriously despite being outside a PCOS population. First, it is a dose-response relationship across nearly 6,100 people, not a single small trial. Second, skim milk showed essentially the same association as whole milk — which argues against a simple “it’s the milk fat” explanation, since skim milk has the fat removed but keeps the hormones, insulin-like growth factors and whey proteins naturally present in cow’s milk. That is the “IGF-1 question” in the title: milk (any fat content) reliably raises circulating IGF-1 and insulin more than an equivalent-calorie plant food, and both hormones are known drivers of the follicular overgrowth and oil production behind acne. That mechanism is plausible and consistent with the skim-milk finding — but it comes from cohort association, not a PCOS trial, and cohort data shows two things moving together, not proof that one causes the other. The fuller picture on PCOS acne, including what else drives it, is here.

Is Bloating After Dairy a PCOS Symptom or Something Else?

Digestive discomfort after dairy and hormonal PCOS symptoms are two different problems that frequently get filed under the same complaint. Lactose intolerance — reduced ability to digest the sugar in milk — causes bloating, gas and loose stool within a few hours of dairy, driven by digestion in the gut, not by androgens or insulin. Neither PCOS trial above measured or reported gastrointestinal symptoms; they measured weight, insulin and hormones over 8 weeks. If dairy causes you visible, same-day bloating, that is worth addressing on its own terms — the four common causes of PCOS bloating are broken down here — rather than folded into a testosterone-and-acne question it was never part of in the trial data.

Does Phenotype Change How Much This Matters?

Both PCOS dairy trials specifically recruited overweight and obese women, on the reasoning that hyperinsulinemia — common in that group — drives higher androgen production, and that dairy and starch both provoke a larger insulin response than non-starchy vegetables and protein (Phy et al., 2015). That rationale is specific to the insulin-resistant phenotype. If your PCOS presents with normal insulin sensitivity and a predominantly adrenal or inflammatory pattern, the mechanism these trials were built around may not be the one driving your symptoms, and there is no separate trial testing dairy removal in that group. Phenotype is also why a person with lean PCOS trying to replicate an 8.6 kg weight-loss result from a trial built around calorie deficit and weight loss is working from a mismatched starting point — there may be little or no weight to lose in the first place.

What Does “Dairy-Free for PCOS” Actually Mean in Practice?

If you remove dairy specifically to see whether your own acne, bloating or another symptom improves, the honest framing is a personal experiment, not a proven PCOS treatment. A few practical points the trials above don’t cover but matter for actually doing this:

• Replace what dairy was providing. Milk and yoghurt are common sources of calcium, iodine and protein. A dairy-free trial that also drops protein or calcium intake will not tell you anything useful about dairy — it will tell you about protein or calcium, the same confounding problem the trials above ran into.
• Give it real time before judging it. Skin takes weeks, not days, to show a visible change, since existing blemishes have to run their own course regardless of what you eat now; the full marker-by-marker diet timeline is here if you want a fuller sense of what changes when.
• Change one thing at a time if you actually want an answer. The two PCOS trials above are useful evidence precisely because they show what removing dairy and starch together can do — and precisely unhelpful for isolating dairy, because they changed two things at once. Removing dairy on its own, while holding everything else steady, is the only way a personal trial tells you anything about dairy specifically.
• Pick a genuine substitute, not just an absence. Fortified plant milks are formulated to match cow’s milk on calcium, and several also match its protein content — a direct comparison across cow, soy, oat and almond milk is here if you want to swap rather than simply cut.

Who This Will Not Help

If your PCOS symptoms are driven primarily by a phenotype where insulin resistance is not the main driver — an adrenal or inflammatory presentation, for instance — a dairy-focused change targets a pathway that may not be your dominant one, and the acne-specific evidence above comes from a general teenage population that was not selected for PCOS at all. If you already eat little dairy and have not seen the symptom improvement you were hoping for, more restriction is not obviously the next lever — the diets with the deepest PCOS trial evidence are built around what to add, not primarily around single-food removal, and are worth comparing before cutting dairy further.

It also will not help as a standalone fix for anyone expecting the acne cohort data above to behave like a PCOS trial result. That cohort measured a general population of teenage girls over three years of dietary recall — a real, dose-responsive signal, but one measured in people who were not selected for insulin resistance, high androgens or any PCOS diagnosis. Treating a 20% relative increase in acne prevalence in that population as equivalent to a guaranteed individual result overstates what a cohort association can promise any one reader.