Insulin Resistance and PCOS: What's Actually Happening, and What Helps

Why insulin resistance sits at the center of PCOS

Insulin sensitivity runs about 27% lower in women with PCOS than in women without it, according to a meta-analysis of 28 gold-standard clamp studies — the most accurate way insulin sensitivity can be measured directly. That gap held regardless of BMI. A higher BMI made the resistance worse by another 15%, but it did not create it. Something about PCOS itself blunts the signal, in thinner bodies too, which is why “just lose weight” was never a complete answer to begin with.

That matters because of a quirk in how insulin behaves once it is elevated. In muscle and fat, insulin resistance means those tissues stop responding well to insulin’s request to take up glucose, so your pancreas releases more insulin to compensate. But the ovary listens to insulin through a largely separate pathway. There, insulin acts as what one major review calls a co-gonadotropin — it partners with luteinizing hormone to push the ovary toward producing more testosterone. So the same excess insulin that fails to lower your blood sugar succeeds at raising your androgens. Resistant where it should be working, sensitive where you’d rather it wasn’t.

That compensating cycle is also the physiology behind a pattern many people with PCOS already recognize: real hunger and shakiness two or three hours after a carbohydrate-heavy meal. A spike in insulin drives blood sugar down hard on the way back, and the hunger that follows is a chemical signal, not a lack of discipline.

You may now see this condition written as polyendocrine metabolic ovarian syndrome (PMOS), after a 2026 global consensus of more than 50 medical organisations renamed it. Nothing about the biology above changed. Insulin still does the same two things, under either name.

How insulin resistance actually gets tested

No single blood draw diagnoses insulin resistance in PCOS. Clinicians build the picture from a few numbers read together, not any one of them alone.

Fasting glucose. A single blood-sugar snapshot. Normal on its own does not rule out insulin resistance — it often stays normal precisely because your pancreas is compensating.

Fasting insulin. Measured alongside glucose. A high insulin next to a normal glucose is the classic PCOS pattern — proof the pancreas is working harder to hold blood sugar in range.

HOMA-IR. A calculated ratio: fasting glucose × fasting insulin ÷ 405. It’s used more in research than as a strict diagnostic cutoff, and a single value means little without your own lab’s reference range attached to it.

2-hour oral glucose tolerance test (OGTT). You drink a glucose solution and get tested at intervals. The 2023 international guideline folds metabolic risk — including diabetes risk — into routine PCOS care generally, which is why glucose-tolerance testing can come up even when your main concern is cycles or skin, not blood sugar.

HbA1c. A three-month average blood-sugar marker. Useful for the bigger picture; not sensitive enough to catch early insulin resistance by itself.

What actually moves insulin resistance

Two approaches have real trial evidence behind them for insulin resistance specifically in PCOS: changing the composition of what you eat, and metformin. Weight loss is often proposed as a third lever, but the trials below moved insulin markers without reliably moving weight — which is the more useful finding, not a footnote.

A 2021 systematic review and meta-analysis of ten randomised trials compared lower-glycemic-index diets against higher-GI diets of the same general pattern — not against eating less overall. A separate, older meta-analysis of 13 randomised trials did the same for metformin against placebo. Both are summarized below.

Approach	Trial evidence	What moved	What didn’t move
Lower-glycemic-index diet	10 RCTs, 403 women	HOMA-IR down 0.78; fasting insulin down ~2.4 µIU/mL; waist circumference down 2.8 cm; total testosterone down	Body weight — no significant difference vs higher-GI diets
Metformin	13 RCTs, 543 women	Fasting insulin reduced; ovulation odds nearly 4x placebo (OR 3.88)	BMI and waist-to-hip ratio — no significant difference vs placebo

Read the last column twice. Both interventions changed insulin markers without reliably changing weight. If the number you’re tracking is on a scale, neither trial promises to move it. If the number is what’s happening in your blood vessels and ovaries, both did.

This sits inside the weight-loss section, which — despite the name — is really about metabolic markers: insulin, blood pressure, cholesterol, cycle regularity. Not a target weight. The same 2023 guideline that set these priorities also explicitly names weight stigma as something clinicians should be aware of when discussing metabolic care in PCOS — that’s the guideline’s own language, not a softened marketing line. A plan built around forcing one number on a scale down is not what this evidence supports. A plan built around glucose, insulin and cycles moving in a better direction is.

Where your phenotype changes this

Insulin resistance is common in PCOS, but it isn’t universal, and how central it is to your case depends on which pattern you have. It shows up most consistently in research on the classic, higher-androgen phenotypes with irregular cycles. If your presentation is driven more by post-pill androgen rebound or by the stress axis, insulin plays a smaller role, and cutting carbohydrates harder is unlikely to be the lever that moves your symptoms. The four types of PCOS walks through how to tell which pattern you’re working with.

The waist-circumference change in Table 1 is one reason abdominal fat distribution gets particular attention in PCOS research specifically — see PCOS belly for what that pattern does and doesn’t tell you. And because the excess insulin described above acts on androgen production, it’s also worth reading high testosterone in women if hair growth or acne are what brought you here rather than a lab number.

If cutting refined carbohydrate hasn’t shifted your cycles, skin, or hair growth after a few months, that’s useful information — it’s a reason to ask your clinician about androgen levels directly rather than assuming you haven’t tried hard enough.

This is also why two people can follow the same eating pattern and get different results. One is mostly correcting an insulin-driven process, and the levers in Table 1 have real work to do. The other’s cycles and skin are driven more by cortisol or by androgens that never fully settled after stopping hormonal birth control, and food composition was never going to be the biggest lever for that picture — regardless of how carefully it’s followed.

Your next 24 hours

You don’t need a full metabolic workup to start. Pick one thing tonight: swap the fastest carbohydrate on your plate — white rice, white bread, a sugary cereal — for a slower one, and notice nothing more dramatic than whether you’re hungry again within the hour.

Then write down two questions for your next appointment: “Can we check my fasting glucose and fasting insulin together?” and “What’s my HOMA-IR, and what does that mean for me specifically?” Those two numbers turn a vague sense that insulin “might be involved” into something you can actually act on, rather than another round of guessing from symptoms alone.

Bring the answer back to what you’re actually trying to change — cycle regularity, skin, energy crashes, or a lab value your doctor flagged — and let that decide whether diet, metformin, both, or neither one is the next conversation worth having.