The PCOS Acne Diet: What the Evidence Supports, and What It Does Not

PCOS was renamed polyendocrine metabolic ovarian syndrome (PMOS) in 2026 by a consensus of more than 50 organisations (Lancet 2026). Nothing about the diet-acne evidence below moved with the rename; this article uses PCOS because that is still what people search.

Does Diet Actually Cause PCOS Acne? What the Evidence Actually Tests

No study has put people with PCOS on a controlled diet and measured what happened to their acne — that specific trial does not exist. Everything below comes from two different kinds of evidence instead: large general-population cohort studies tracking dairy intake against acne over time, and two small randomised trials — one on glycemic load, one on chocolate — that happened to enrol only men. The full text of the 2023 international PCOS guideline mentions acne six times and diet seven times, and the two never appear together — acne is discussed only as a sign to assess when diagnosing hyperandrogenism, never as a target for dietary treatment (Teede et al., 2023). That gap between what gets marketed as “the PCOS acne diet” and what the guideline itself actually addresses is the honest starting point here.

That does not make the evidence worthless — insulin and androgen pathways are shared mechanisms, so a dietary effect found in a general population plausibly transfers to PCOS, where those same pathways are already running hotter. It does mean every number below needs its population stated plainly, not borrowed as if it were PCOS-specific.

Does Dairy Trigger PCOS Acne? What the Cohort Numbers Actually Show

Skim milk is the dairy category with the most consistent signal across the largest studies, and it is not the one most people blame first. A prospective cohort of 6,094 girls aged 9 to 15 found that comparing the highest milk intake (2 or more servings a day) against the lowest (under 1 a week), the prevalence ratio for acne was 1.20 for total milk, 1.19 for whole milk, 1.17 for low-fat milk, and 1.19 for skim milk — all statistically significant (Adebamowo et al., 2006). A separate cohort of 4,273 boys found a different pattern: only skim milk reached significance (prevalence ratio 1.19, p = 0.02), while total milk did not (p = 0.77) (Adebamowo et al., 2008) — a sex difference the authors could not fully explain, but one that argues against treating “dairy causes acne” as a single uniform rule.

A meta-analysis pooling the observational literature found the same skew toward skim and low-fat milk: total dairy carried an odds ratio of 2.61, total milk 1.48, low-fat milk 1.25 and skim milk 1.82, all statistically significant, while yogurt and cheese showed no significant association at all (Aghasi et al., 2019). That pattern — skim worse than whole, fermented dairy not implicated — points toward a specific bioactive component of liquid milk (leading theories involve whey proteins and hormone residues that survive pasteurisation) rather than dairy fat, since removing the fat did not remove the association.

The one large study in adults, rather than teenagers, found a much weaker signal. The NutriNet- Sante cohort followed 24,452 French adults with a mean age of 57 — a population you would not otherwise reach for PCOS-acne evidence — and found milk intake only barely reached statistical significance for current acne (adjusted odds ratio 1.12, 95% CI 1.00–1.25), while fatty and sugary food intake overall carried a stronger association (aOR 1.54) (Penso et al., 2020). Every one of these four studies is observational — they show milk intake and acne moving together, not that removing milk clears skin — and none was conducted in a population selected for PCOS or hyperandrogenism.

Does a Low-Glycemic Diet Actually Clear Acne? The One Real Randomised Trial

Yes, in the one trial that has actually tested it — with a confound worth naming. Forty-three male acne patients aged 15 to 25 were randomised to a 12-week low-glycemic-load diet (25% of energy from protein, 45% from low-glycemic-index carbohydrate) or a control diet emphasising carbohydrate without reference to glycemic index. At 12 weeks, total lesion counts fell by 23.5 in the low-glycemic-load group versus 12.0 in the control group (p = 0.03), and insulin sensitivity improved significantly more in the intervention arm (Smith et al., 2007).

The confound is real and the study reports it honestly: the low-glycemic-load group also lost more weight (2.9kg versus 0.5kg, p < 0.001) and more body mass index than the control group. Weight loss itself can improve acne through the same insulin and androgen pathways a lower glycemic load targets, so this trial cannot fully separate “glycemic load specifically improved skin” from “a diet that also produced weight loss improved skin.” Both explanations point toward the same underlying insulin mechanism that PCOS diet strategies generally target, which is why the direction of this result is plausible for PCOS even though the trial itself enrolled only teenage and young-adult men, not women with PCOS.

Does Chocolate Actually Trigger Acne? What a Blinded Trial Found

A small, tightly controlled trial found that pure cocoa — isolated from the sugar and milk most chocolate bars also contain — measurably worsened acne on its own. Fourteen men aged 18 to 35 with a history of acne swallowed capsules containing unsweetened 100% cocoa, a placebo (hydrolysed gelatin powder), or a combination, without knowing which, at a single site. Thirteen completed the study. Total acneiform lesion counts rose significantly by day 4 (p = 0.006) and remained elevated at day 7 (p = 0.043) compared with baseline, and the amount of cocoa each person consumed correlated with how many new lesions they developed (Caperton et al., 2014).

Using capsules rather than an actual chocolate bar is the useful design choice here — it isolates cocoa from the sugar, milk and fat that a real bar also contains, which is what lets this trial say something a chocolate-bar study could not: cocoa itself, not just the sugar or dairy typically blamed alongside it, appears to move the needle. The trial is small — 13 completed participants, all men — so treat the direction as a real, blinded finding and the exact numbers as a starting point rather than a settled dose-response relationship.

What About Foods Marketed to Help — Zinc, Omega-3, Anti-Inflammatory Eating?

Two spices come up constantly in “anti-inflammatory for acne” content specifically — ginger and turmeric — and both are covered on their own pages with the same evidence-first standard rather than repeated here.

This article stays on what changes acne through diet, not what treats it through supplementation or medication — those sit on a separate, more established evidence base. Two supplements actually have trial data for PCOS acne specifically, and most marketed options do not; that comparison, and the hormonal mechanism behind PCOS acne along with the medical treatment options ranked by timeline, are covered in full elsewhere rather than restated here. The topical side of that same ladder gets its own separate comparison too — which OTC skincare actives are actually tested at the strengths the trials used.

Who This Won’t Help — When Acne Needs a Work-Up, Not a Food Change

Diet changes are not the right first response to acne that is nodulocystic, scarring, or not improving after several weeks of consistent topical treatment — that pattern needs a dermatologist regardless of what you eat, because the driver at that severity is androgen signalling deep in the follicle, not something a lower glycemic load reliably reaches in weeks. The same is true if acne shows up suddenly alongside rapid new hair growth, a deepening voice, or cycles that changed abruptly rather than gradually — those combinations point toward a hormonal picture that needs testing, not a dietary trial-and-error period first.

If what’s actually driving the search for a “PCOS diet” is something other than skin — hair thinning, or a baking or recipe question specifically — the diet evidence for PCOS hair loss is its own, separately thin story, and which baking swaps actually change the glucose response, with the numbers, rather than which ones just change the label covers the recipe side of PCOS eating directly.

Your Next Step

If you want to test the dairy signal specifically, swap skim or low-fat milk for whole milk or a non-dairy alternative for 8 to 12 weeks — long enough for a full follicle cycle — before judging whether it changed anything, rather than stopping after a few days. If cystic or scarring lesions are part of the picture, start that conversation with a dermatologist in parallel, not after a diet trial finishes; diet and medical treatment are not mutually exclusive options here.