Are the 4 Types of PCOS Real? What the Phenotypes Actually Are

Is “4 types of PCOS” an actual medical classification?

No guideline anywhere defines PCOS by “insulin-resistant,” “inflammatory,” “post-pill” or “adrenal” categories. The 2023 international evidence-based guideline for PCOS — built from a systematic review of the entire published literature — does not use any of those four labels once in its recommendations. What it uses instead is the Rotterdam classification, agreed by an ESHRE/ASRM consensus workshop in 2003, which diagnoses PCOS when two of three features are present: irregular or absent ovulation, clinical or biochemical androgen excess, and polycystic ovarian morphology. Those two-of-three combinations create four real phenotypes, labelled A, B, C and D — four combinations of features, not four causes. See how PCOS is actually diagnosed, step by step for the full work-up this fits into.

The wellness version circulating online groups women by presumed root cause instead, and assigns the label a symptom quiz decides, with no blood test, ultrasound finding or diagnostic cutoff attached to any of the four names. That doesn’t make the ideas behind them baseless — insulin resistance, inflammation, hormonal rebound after contraception, and adrenal androgen excess are all real, studied mechanisms in PCOS. It means the packaging — four discrete “types” identified from a checklist — isn’t how any of them actually sorts women in the research. Below is what’s true in each claim, and what your doctor is actually checking for when your chart gets a phenotype letter instead.

What are the real PCOS phenotypes, A through D?

Phenotype A — androgen excess, irregular ovulation and polycystic ovaries together — was the most common presentation in a prospective study of 1,212 women with PCOS, found in 48.2% of diagnoses, compared with 30.7% for phenotype B, 11.4% for phenotype D, and just 9.7% for phenotype C. Each letter is simply a different pairing of the two Rotterdam features a person meets.

Phenotype D is the only combination with no measurable androgen excess, which is also why it’s the hardest to tell apart from other causes of irregular ovulation — including lean PCOS presentations and functional hypothalamic amenorrhea, where the same ultrasound picture can show up without PCOS being the cause at all.

Is “insulin-resistant PCOS” a real subtype?

Insulin resistance is measurably more common in two of the four real phenotypes, not a fifth category layered on top of them. In that same 1,212-woman study, both normal-weight and overweight/obese women with phenotype A or B were significantly more insulin resistant than BMI-matched women without PCOS. Phenotype C showed no difference from controls at any weight, and phenotype D only showed excess insulin resistance in the overweight or obese subgroup — not in normal-weight women with that phenotype. So the popular idea maps loosely onto something real: androgen excess plus irregular ovulation (A and B) does track with more insulin resistance. But it isn’t a diagnosable “type” layered on top of your real phenotype — it’s a gradient that already lives inside the phenotype letter you were assigned at diagnosis. Insulin resistance has its own separate work-up, independent of which letter you carry, and it’s worth asking for regardless of phenotype.

Is “inflammatory PCOS” real?

Chronic low-grade inflammation is present in PCOS independent of body weight, according to a 2012 review of the mechanistic evidence: glucose ingestion triggers oxidative stress and cytokine release from immune cells in women with PCOS even without obesity, and markers of inflammation correlate closely with circulating androgens. That is a genuine, well-documented mechanism. What doesn’t exist is a lab test, a cutoff, or a guideline that sorts a subset of women into an “inflammatory type” separate from the rest — the inflammation described in that research shows up across phenotypes and weight categories, not in one identifiable slice of PCOS patients. Framed as a mechanism operating to different degrees in most PCOS, the claim is accurate. Framed as one of four discrete, mutually exclusive types, it overstates how cleanly the biology actually sorts.

Is “adrenal PCOS” real?

About 20% of white and 30% of Black women with PCOS have adrenal androgen excess — a measurably elevated DHEAS, the androgen made by the adrenal gland rather than the ovary — according to a study of 213 women with PCOS matched against 182 controls. That’s a real, testable finding: DHEAS is a specific blood value, and a raised level does point toward an adrenal contribution to androgen excess. Where the wellness framing overreaches is treating “adrenal” as one of four mutually exclusive boxes. A separate study of 238 hyperandrogenic women found elevated DHEAS in statistically similar proportions across classic PCOS (39.6%), ovulatory PCOS (29.1%) and non-PCOS hyperandrogenism (48.3%) alike. Adrenal androgen excess overlaps every phenotype rather than defining a distinct one — and it’s a real, useful thing to test for through a DHEAS blood draw, regardless of which letter or label you’ve already been given.

Is “post-pill PCOS” real?

Stopping a combined oral contraceptive cannot cause PCOS, because the pill suppresses ovulation rather than producing it — what it can do is unmask an anovulatory pattern that was already present before you started taking it. The international guideline addresses this directly: because hormonal contraception itself alters cycle regularity, androgen levels and ovarian appearance, diagnosis is not made while someone is taking it, and testing is deferred until the natural cycle re-establishes, which is typically within three months of stopping for most women. If irregular cycles or acne appear only after coming off the pill, that is usually the underlying pattern becoming visible again, not a new condition the pill created and then handed you on the way out.

Which real phenotype carries the most metabolic risk?

Phenotype A carries the heaviest combined load of insulin resistance and androgen excess of the four real phenotypes, phenotype B is close behind it, and phenotype C shows the least metabolic disturbance of any group at any weight.

Circulating androgens follow the same order — higher in phenotypes A through C than in phenotype D, and higher in phenotype A than in phenotype B specifically among normal-weight women. That ranking, not a wellness label, is what a clinician means when they say your phenotype carries more or less long-term risk.

How to actually get tested for what each wellness label is gesturing at

Four separate lab values or histories cover almost everything the wellness “4 types” are trying to describe, and three of the four are standard, orderable tests rather than a symptom checklist.

Notice that only three of the four rows have an actual test attached. “Inflammatory” is the one wellness label built on a genuinely real mechanism — chronic low-grade inflammation is well documented in PCOS — that still has no validated clinical cutoff a lab can report back as abnormal or normal. That gap, more than anything else, is why “inflammatory PCOS” functions as a marketing label rather than a diagnosis: the biology behind it is real, but medicine hasn’t yet built a yardstick a clinician can use to measure it in you specifically.

Who the “4 types” framework does not help

This framework will not help you if you’re using a symptom-quiz result in place of the two blood tests and pelvic ultrasound your actual diagnosis requires — a self-assigned “type” carries no diagnostic weight, and leaning on one can delay a work-up for a look-alike condition, including rare conditions that mimic PCOS but need a completely different treatment path. It also doesn’t help if “adrenal type” or “inflammatory type” is being used to justify skipping a standard PCOS diagnostic panel in favour of a supplement protocol sold under that name — the research behind each mechanism doesn’t license a bespoke treatment pathway named after it. And it doesn’t replace phenotype A, B, C or D, which is what actually predicts metabolic risk and is what belongs in your medical chart.

What to ask for instead of finding “your type”

Ask your clinician which of the four real phenotypes is documented in your chart, since that single letter carries more information than any online quiz result: it tells you exactly which two Rotterdam criteria you met, and it’s tied to the measured differences in insulin resistance and androgens described above. If your BMI is in the normal range, ask specifically about lean PCOS’s higher rate of missed insulin resistance, since a normal number on the scale is often wrongly treated as reassurance that metabolic testing isn’t needed.

Note: in May 2026, PCOS was renamed polyendocrine metabolic ovarian syndrome, or PMOS, by a global consensus of more than 50 organisations. The Rotterdam phenotypes and everything above them carry over unchanged — only the name on the diagnosis changed. This article uses PCOS, since that’s still what most readers search.

Your next step

Ask your clinician to state your Rotterdam phenotype letter in plain terms at your next appointment, and request a fasting insulin and DHEAS alongside your standard PCOS panel if neither has been checked. That combination — a letter plus two numbers — tells you more about your actual risk than any online “type” quiz will.