PCOS and belly fat: why it is stubborn, and what actually moves it

PCOS — polycystic ovary syndrome — affects somewhere between 8% and 13% of women globally. In South Asian populations, reported figures run considerably higher, with studies citing up to 22%, and a South-East Asian regional figure of 14.3% using the standard diagnostic criteria.

So if it feels like a lot of women you know have it, that is not an impression. It is roughly one in seven to one in five.

It is also routinely diagnosed late, and often first noticed because of exactly the thing this article is about: weight that settles around the middle and will not respond to the things that used to work.

It is diagnosed using the Rotterdam criteria, set in 2003 and still the most widely used. You need two of these three:

1. Irregular or absent ovulation — periods that are infrequent, unpredictable, or missing.
2. Excess androgens — either shown on a blood test, or visible as symptoms like unwanted facial and body hair, acne, or hair thinning at the scalp.
3. Polycystic ovaries on ultrasound.

Two of three. Which is why the name misleads so many people: you can have PCOS without having cysts on your ovaries at all. Plenty of women are told their scan was clear and conclude they must be fine, when the other two criteria were already met.

Only a doctor can diagnose this. Nothing on this page can, and nothing you read online should be treated as a substitute for that conversation.

This is the part that explains the frustration, and it is worth reading slowly.

In PCOS, cells respond poorly to insulin. The body compensates the only way it can: it makes more insulin. Higher circulating insulin has two consequences that matter here.

First, insulin is a storage signal. Persistently high levels favour fat storage and make fat release harder — so the same effort produces less result than it would in someone without insulin resistance. That is not a motivation problem. It is an endocrine one.

Second, high insulin pushes the ovaries to produce more androgens, and androgens drive fat towards the abdomen specifically rather than the hips and thighs. Studies measuring this find a higher trunk-to-leg fat ratio in women with PCOS compared with controls.

Then it closes the loop: abdominal fat is itself metabolically active and worsens insulin resistance. More insulin resistance, more abdominal fat, more insulin resistance.

Between 40% and 80% of women with PCOS are overweight or obese — but the loop runs in lean women too, and South Asian women in particular show insulin resistance and raised cardiometabolic risk at comparatively low BMI.

The encouraging part is that the loop runs both ways. Reduce the insulin load and the same mechanism starts working in your favour.

Protein at every meal. This is the highest-value change and the one most often missing from a desi plate. Eggs, chicken, fish, daal, chana, dahi, paneer. Protein blunts the glucose rise of the meal it is eaten with and keeps you full long enough that the 5pm biscuit stops being inevitable.

Do not eat carbohydrate alone. Roti by itself, chai with rusk, plain rice — each of these is a sharper insulin signal than the same food eaten with protein, fat and fibre alongside. The fix is not removal. It is company.

Fibre, early and often. Kachumber before the meal, sabzi taking up real space on the plate, chana and beans as actual portions rather than garnish.

Resistance training. Muscle is where glucose goes. More of it means better insulin sensitivity, independent of weight lost. Two sessions a week is a real dose, and it does not require a gym.

Sleep. One bad night measurably worsens insulin sensitivity the next day. For a condition defined by insulin resistance, this is not a soft recommendation.

Walk after eating. Ten to twenty minutes after your largest meal blunts the glucose peak. It is the cheapest intervention on this list.

PCOS attracts more bad advice than almost any other condition, because it is common, frustrating, and mostly managed by the woman herself.

Anything that names a single food as the cause or the cure. Not rice, not wheat, not dairy, not one particular seed. If it were that simple it would be in every guideline.

"PCOS tea" and detoxes. There is no mechanism by which these address insulin resistance, and some are diuretics, which change your weight for a day and nothing else.

Cutting carbohydrates to near zero. It produces fast early results and, in a household where roti and chawal are what gets cooked, it is abandoned within weeks — and the regain is demoralising in a way the original weight never was.

Being told to "just lose weight". It is circular advice for a condition that makes losing weight harder, and it skips the mechanism entirely.

PCOS is managed, not cured. Anyone promising otherwise is selling something.

What the evidence supports is that even a modest reduction in weight and insulin load can improve cycle regularity and symptoms — and that improvements in insulin sensitivity often show up before the scale moves much at all. Energy first, cycles later, measurements after that, in roughly that order.

No fixed timeline is promised here, because pace genuinely depends on the person. What can be said is that the thing most likely to work is the plan you can still be doing in six months, cooked from the food already in your kitchen.