PCOS Weight Loss: What Actually Works, and Why It Starts Harder

Why Is Losing Weight Harder With PCOS?

Insulin sensitivity runs about 27% lower in women with PCOS than in women without it, regardless of body mass index, according to a meta-analysis of 28 gold-standard clamp studies — the most direct way insulin sensitivity can be measured. That gap is the starting point for everything else on this page, because it changes what a calorie deficit is working against before a single meal is eaten. In muscle and fat, resistant tissue stops responding well to insulin, so the pancreas releases more of it to compensate. The ovary, though, listens to that excess insulin through a largely separate pathway, where insulin behaves as a co-gonadotropin that partners with luteinizing hormone to push androgen production up. The same hormone that fails to lower blood sugar succeeds at raising testosterone — which is the mechanical reason “eat less, move more” was never a complete answer for this condition. Insulin resistance and PCOS covers that mechanism and how it is actually tested, in full.

This does not mean every case looks the same. The four recognised PCOS phenotypes carry different degrees of insulin involvement, and a normal-weight presentation can carry the same 27% deficit with no excess weight attached to it at all — lean PCOS and weight covers that pattern specifically, because most of what follows on this page assumes there is excess weight for an intervention to act on.

What Actually Changes at 5% Weight Loss — and What Needs More?

Eighty-two percent of women who lost more than 5% of their starting weight regained menstrual function in a 1992 trial of 24 women with PCOS, versus 12.5% of those who lost less than that (Kiddy et al. 1992). A modern randomised trial found the same range: 149 women with PCOS-related infertility were assigned to 16 weeks of a preconception lifestyle programme targeting 7% weight loss, achieved 6.2% on average, and raised cumulative ovulation from 46% with contraceptive pretreatment alone to 60-67% with the lifestyle intervention (Legro et al. 2015). Neither trial found universal response. Both found a real, measurable shift in the odds at a modest percentage — not the large loss that “weight loss” implies in most general advice.

The mechanism behind why a small number moves the needle matters more than the number itself. In an 18-woman trial of gradual diet and exercise change, the women who regained ovulation showed an 11% drop in central fat, a 71% improvement in insulin sensitivity, and a 39% fall in luteinizing hormone — none of which shifted significantly in the women who did not regain ovulation, even though total body weight was not the variable most tied to the outcome (Huber-Buchholz et al. 1999). Two people can lose an identical percentage and see different reproductive results, because what actually happens to insulin and centrally distributed fat underneath that percentage is the real driver. How much weight loss restores ovulation in PCOS walks through both trials, and what they do and do not promise, in more depth than fits here.

What Does Diet Composition Actually Change?

Weight loss improved PCOS presentation under nearly every macronutrient composition tested in a systematic review of six trials and 137 women with PCOS, comparing low-glycemic-index, low-carbohydrate, high-protein, high-carbohydrate and monounsaturated-fat-enriched diets against each other (Moran et al. 2013). The review’s own conclusion was that weight loss should be targeted “irrespective of diet composition” — but composition still moved specific markers that weight loss alone did not touch equally. A low-glycemic-index pattern produced the largest gains in menstrual regularity and quality of life. A low-carbohydrate pattern produced the largest drop in insulin resistance, fibrinogen and total and HDL cholesterol. Free androgen index moved in the wrong direction on a higher-carbohydrate diet in one trial within that same small evidence base. No single ratio won on every outcome, which is why the honest answer to “what should I eat” depends on which marker matters most to you specifically — covered at the food and portion level in the PCOS diet guide, and broken down outcome by outcome in PCOS macros for weight loss.

None of the trials behind that review, or any other trial cited on this page, tested a specific calorie target, macro split, or rate of weight loss that this article can respectably hand you as advice — every number above describes what a specific trial’s protocol used, in a specific, supervised population, not a prescription. The low-glycemic diet for PCOS covers the pattern that moved cycles and mood most reliably, in practical, food-level terms.

Does Exercise Add Anything Diet Alone Doesn’t?

Diet alone cut body weight by 8.9% over 20 weeks in a trial of 94 overweight and obese women with PCOS, a result statistically indistinguishable from the 8.7% and 10.6% losses posted by groups that added aerobic or combined aerobic-resistance training to the identical energy-restricted, high-protein diet (Thomson et al. 2008). Testosterone, free androgen index and sex-hormone-binding globulin all improved by a similar amount across every group, diet-only included — the hormonal benefit tracked with the diet and the weight lost on it, not with whether a workout was part of the plan.

Exercise still changed something a scale could not see: the groups that trained lost roughly 3 kilograms more fat and 2 kilograms less muscle than the diet-only group, for the same total weight change. That is a real, meaningful difference for long-term metabolic health even though it never shows up as “more weight loss” on a chart — PCOS weight change without exercise covers the full trial and what a larger, lower-quality pooled Cochrane review found by comparison (considerably less than 8.9% on average, rated low-quality evidence). Exercise also does more for insulin resistance specifically than diet alone does: a separate meta-analysis found vigorous exercise the single largest lever for lowering HOMA-IR among everything tested, ahead of diet changes or any supplement — ranked in full in treating insulin resistance in PCOS. Building strength without treating weight loss as the goal is covered directly in strength training for PCOS and in body recomposition with PCOS, for anyone whose priority is composition rather than a number on a scale. How HIIT and low-impact training actually compare goes further into which specific intensity moved HOMA-IR in the trials above, and where a lower-intensity default fits phenotypes more sensitive to cortisol.

What Does Metformin Do for Weight and Insulin?

Metformin nearly quadrupled the odds of ovulation compared with placebo (odds ratio 3.88) and reduced fasting insulin in a meta-analysis of 13 randomised trials and 543 women with PCOS — but did not produce a significant change in BMI or waist-to-hip ratio compared with placebo across those same trials (Lord et al. 2003). That split matters for what metformin is actually for in this context: it is a well-evidenced option for insulin resistance and ovulation, not a weight-loss drug, and treating it as one sets an expectation the trial evidence does not support. Metformin for PCOS covers the mechanism, dosing conversation and who should not take it; metformin for weight loss in PCOS covers the real average result specifically, rather than the anecdote.

What About GLP-1 Drugs Like Semaglutide?

GLP-1 drugs are not licensed for PCOS anywhere, and current prescribing guidance calls for stopping them before conception — a real constraint worth stating plainly, since most people reading a PCOS weight-loss page are also managing fertility. Semaglutide does have real, if limited, PCOS-specific trial data: a 2025 randomised, open-label trial of 100 overweight or obese women with PCOS compared metformin alone against metformin plus weekly semaglutide for 16 weeks and found 6.09 kg of weight loss in the combination group versus 2.25 kg with metformin alone, alongside greater improvement in testosterone, visceral adiposity and CRP (Chen et al. 2025). Twenty of the 100 women randomised did not complete the trial, and the pregnancy-rate comparison often quoted from this study — 35% versus 15% — is drawn from the roughly 40 women remaining per arm at week 40, after semaglutide had already been stopped in both arms and every participant was carried through the pregnancy window on metformin alone. That is a real, encouraging first data point, not a settled answer, and it says as much about stopping semaglutide before pregnancy as it does about starting it.

Roughly two-thirds of the weight lost on semaglutide returns within a year of stopping it, based on the best available long-term data — the general-population STEP 1 trial extension, not a PCOS-specific one. Participants who had lost an average 17.3% of body weight over 68 weeks regained 11.6 percentage points of that within a year of stopping the drug and its accompanying lifestyle programme, landing at a net 5.6% below their starting weight (Wilding et al. 2022). Tirzepatide has no PCOS-specific trial at all; everything attributed to it in this population is extrapolated from general obesity and type 2 diabetes data. GLP-1 drugs for PCOS covers the full mechanism, contraindications and stopping-before-conception timeline; GLP-1 access and cost covers coverage, pricing and the real risk attached to compounded versions. A different, older appetite-suppressant option sometimes discussed alongside these drugs — phentermine — works through neither of their mechanisms and carries its own 12-week licensing limit. For the non-prescription route some people try instead, what OTC weight-loss pills actually contain covers the evidence and safety picture separately.

How Long Before Any of This Actually Shows Up?

Hormonal and inflammatory markers moved within 16 weeks in the semaglutide-metformin trial above, while full ovulation-restoration trials ran 16 weeks to 6-7 months before assessing whether cycles had actually returned. That spread is the single most useful thing to know before starting anything on this page: different markers move on different timelines, and comparing your week two against a trial’s month six sets an expectation the physiology cannot meet.

What to Check When Nothing Is Moving

A stalled scale despite a genuine, sustained effort has a testable explanation behind it far more often than it has a missing-effort explanation behind it. Thyroid disease, undiagnosed sleep apnoea, a weight-affecting medication, adaptive metabolism from under-eating, and untreated insulin resistance itself are five of the most common, testable causes — cannot lose weight with PCOS walks through nine of them in the order worth checking. Whether PCOS actually slows resting metabolism at all is a separate, genuinely unsettled question — the direct evidence disagrees by study, and does PCOS slow your metabolism covers why. Why PCOS causes weight gain covers the four mechanisms behind the gain itself, for anyone trying to understand the starting point rather than the plateau.

It is also worth naming plainly that PCOS care has a documented weight-stigma problem: the 2023 international guideline itself flags weight stigma as something clinicians should account for when discussing metabolic care, not a softened talking point this site is adding on top (Teede et al. 2023). PCOS and weight stigma at the doctor covers what that looks like in an appointment and how to redirect it toward the markers that actually matter.

This Will Not Work for You If…

Weight loss is the wrong target entirely for two groups reading this page, and naming them plainly matters more than any trial number above. If your PCOS presents lean — normal or low body weight — insulin resistance is still present in a majority of PCOS cases even after researchers statistically adjust for BMI, in one of the largest studies to measure it directly across 271 patients and 260 controls (DeUgarte et al. 2005). None of the trials in Table 1 enrolled lean participants; every one of them selected for overweight or obese women, so nothing above tells you what to expect if there is no excess weight for an intervention to act on. Lean PCOS and lean PCOS and weight cover what the actual lever is instead.

If your relationship with food includes restriction, loss-of-control eating, or a diagnosed eating disorder, weight loss as a stated goal is also the wrong frame — women with PCOS carry 1.5 to 3.9 times the odds of a diagnosed eating disorder compared with women without it, per a 2024 meta-analysis of 20 studies and roughly 29,000 women with PCOS (Cooney et al. 2024), and calorie-restrictive advice is a documented setup condition for restriction-rebound cycles in that population specifically. PCOS and binge eating covers the full mechanism, a validated five-question screening tool, and where to get help, including crisis resources if that description fits what you are experiencing right now.

Beyond those two groups, none of the trials above ran longer than 20 weeks, and none tested what happens to insulin, fat distribution or weight a year or more past the study period. Pregnancy and breastfeeding sit outside every trial cited here as well — the postpartum weight picture for PCOS covers that window directly, including which weight-loss medications actually have breastfeeding data. Treat every percentage and kilogram figure on this page as a description of what a specific, supervised trial measured — not a target, and not a verdict on effort if your own result looks different.

You may 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 mechanisms or trial data above changes under either name — this page uses PCOS because that is still what most readers search.

Your Next Step

Pick one question from this page and bring it, specifically, to your next appointment, rather than a general “I can’t lose weight” complaint. If insulin resistance has never actually been tested, ask for the 75g oral glucose tolerance test the international guideline recommends. If you are overweight or obese and weighing a GLP-1 drug, ask what stopping it looks like before you start. If your PCOS presents lean, ask whether weight is even the right lever for your specific case. Each of those is a more useful sentence than “nothing works,” because it gives a clinician something specific to test or explain — the same approach every trial cited on this page took, one measured question at a time.