Weight Loss With Type 2 Diabetes

A patient once told me she had been trying to lose weight for two years and had gained eleven pounds doing it. She was not lying to me and she was not lying to herself. She was on insulin and glipizide, and every time her blood sugar ran low she ate to correct it. Her prescription and her goal were pulling against each other, and nobody had told her.

That conversation is the reason this page exists. There is no shortage of advice about losing weight. Almost none of it accounts for the fact that you are taking a drug that changes the answer.

Your medication is part of the arithmetic

This is the piece that general weight-loss advice cannot give you, because writing it requires knowing what is in your prescription.

Two people can eat identically, move identically, and travel in opposite directions on the scale, purely because of what they were prescribed.

So the first question is not what you should eat. It is whether the drugs you take are on your side. That is a conversation with your prescriber, and it is worth having before you change anything else. Regimens can often be shifted — sometimes toward a weight-neutral or weight-lowering agent, sometimes by reducing a sulfonylurea rather than adding to it. If your weight will not move, consider that possibility before you conclude you lack discipline. In my experience it is the more likely explanation.

The dangerous fortnight nobody warns you about

Here is the thing I would most want you to take away.

Blood sugar responds within days. The scale takes weeks. Cut your carbohydrate load meaningfully on a Monday and your glucose readings can be different by Thursday, while your weight has barely moved.

That gap is good news — it means the change is working before you can see it. It is also exactly where people get hurt. You are eating considerably less while still taking a dose calculated for how you used to eat. If that dose is insulin or a sulfonylurea, the arithmetic now produces hypoglycaemia.

⚠️ If you take insulin or a sulfonylurea, talk to your clinician before you start, not after the first low. Not because a diet needs permission — because the dose does. A planned reduction alongside a planned dietary change is routine, safe, and something any competent prescriber will do gladly. An unplanned one at two in the morning is not.

Metformin alone, SGLT2 inhibitors alone, DPP-4 inhibitors alone: these do not cause hypoglycaemia by themselves, and the urgency is lower. It is insulin and the sulfonylureas that demand the conversation first.

Your target is higher than the general advice suggests

For someone without diabetes, losing 5–10% of body weight is a good result and buys real improvements in blood pressure and lipids.

With type 2 diabetes, that is where it starts rather than where it ends.

I am not telling you this to set an intimidating target. I am telling you because being aimed at 5% while quietly hoping for remission is a way to be disappointed by a genuinely good result.

What actually works, in the order I would do it

1. Audit the prescription before the plate

Covered above, and it comes first because everything downstream is easier if the pharmacology is not fighting you. Bring the list to your appointment and ask the specific question: is any of this making weight loss harder, and is there an alternative?

2. Track three ordinary days, honestly

Not a good week. Three normal days, including the bad one. Most people are surprised in one of two directions, and both are useful. You cannot change an intake you have not measured, and estimating it from memory is reliably wrong by several hundred calories.

3. Set the deficit from your own requirement

Not from a magazine number. Work out your basal metabolic rate and daily requirement, then take roughly 500 calories a day off it for about a pound a week. If your target is 10–15%, be realistic that this is a six-to-twelve month project, not a six-week one.

4. Protect your muscle, deliberately

Any substantial weight loss costs some lean tissue. How much is genuinely contested — studies of GLP-1 therapy report lean mass as anywhere from about a quarter to nearly half of total weight lost, and the measurement methods differ enough that I would not trust any single figure.³

What is not contested is what mitigates it: enough protein and resistance training. Aim for roughly 1.2–1.6 g of protein per kilogram of body weight daily, and lift something twice a week. This matters more with diabetes than without it, because muscle is where most of your glucose gets disposed of. Losing weight and losing muscle at the same time trades one problem for another, particularly past sixty.

5. Eat in a pattern you could still be eating in a year

Vegetables, legumes, whole grains, fruit, nuts, adequate protein. Mediterranean, plant-predominant and lower-carbohydrate patterns all have trial support, and the differences between them are far smaller than the difference between following one and following none. The best diet is the one still in place next winter.

My own bias, stated plainly: the Adventist Health Study-2 data on plant-predominant eating is the cleanest long-term evidence we have in nutrition, and diabetes prevalence in that cohort ran from 7.6% in meat-eaters down to 2.9% in vegans. I write about it often because it deserves the attention, not because I think everyone must become vegan.

6. Ask about sleep apnoea

Common in this group, frequently undiagnosed, and it makes weight loss harder while worsening glucose control. If you snore, wake unrefreshed, or your partner has mentioned that you stop breathing, ask. Treating it can change everything downstream.

7. Measure the right things

Weight weekly, not daily — daily readings are mostly water and will make you miserable. Waist circumference monthly. HbA1c every three months. And how many medications you are on, which is the outcome people forget to count and often the first one to improve.

Where the GLP-1 drugs fit

Honestly: they work, and pretending otherwise would be silly.

In STEP 2, semaglutide 2.4 mg produced 9.6% mean weight loss in people with type 2 diabetes over 68 weeks, against 3.4% on placebo.⁴ In SURMOUNT-2, tirzepatide produced 13.4% at 10 mg and 15.7% at 15 mg over 72 weeks, against 3.3% on placebo — and roughly half of participants lost 20% or more.⁵ Those figures are in the range the remission literature cares about, achieved pharmacologically.

Four things I tell patients considering them:

• They are treatment, not a shortcut past the rest. The trials paired them with lifestyle intervention. The muscle-preservation problem gets worse, not better, if you rely on the drug alone.
• Stopping generally means regaining. This is a chronic-disease medication, like a blood pressure tablet. Plan for that rather than being surprised by it.
• Cost and supply are real obstacles, and coverage varies enormously.
• Remission achieved on a drug is not the same as remission off one. The consensus definition requires an HbA1c under 6.5% sustained for three months after stopping glucose-lowering medication. Worth knowing what you are aiming at.

I am not against them. I am against them being offered instead of a conversation about food, movement and sleep rather than alongside one.

The honest limits

Two things the enthusiastic version of this page would leave out.

Look AHEAD — the largest and longest lifestyle trial ever run in type 2 diabetes, over 5,000 people followed for a decade — achieved real weight loss and better fitness, but did not reduce heart attacks and strokes, and was stopped early for futility on that endpoint.⁶ Anyone promising you that losing weight will prevent your heart attack is going beyond the best trial we have.

And weight is not the only thing worth changing. Better fitness, better sleep, less medication and a lower HbA1c are all worth having even if the scale is stubborn. I have had patients whose weight barely moved and whose diabetes control transformed. That is a good outcome and it should be recognised as one.

What to ask at your next appointment

Take this list. Fifteen-minute appointments reward specific questions.

If remission specifically is what you are after, that is a related but distinct question — what it means clinically, who achieves it, and what the trials required. I have written that up separately: type 2 diabetes remission, and what the evidence actually shows.

⚠️ This page is health education, not medical advice, and reading it does not make me your physician. It cannot see your kidney function, your other conditions or your prescription. Do not change or stop any medication on the strength of it. Take the plan to the clinician who manages your diabetes.