How Much Protein on Ozempic or Mounjaro? Why Every Source Gives You a Different Number

Ask the internet how much protein to eat on a GLP-1 and you will get a number. Ask it twice and you will get a different one. The published answers range from about 1.0 to 2.2 grams per kilogram — and they are not even measuring the same kilogram.

That is not a small discrepancy. For a 90kg person it is the difference between 90 grams a day and roughly 200. One of those is a normal amount of food. The other is a part-time job.

This piece does something the ranking pages mostly don't: it explains why the numbers disagree, what the actual consensus documents say, and why the target you pick matters less than whether you have any idea if you're hitting it.

#The numbers, side by side

Here is the spread you'll find across widely-cited sources, with the denominator each one uses:

• 1.0–1.2 g/kg of actual body weight — the low end, common on telehealth prescriber sites
• 1.2–1.6 g/kg of actual body weight — the most commonly repeated range
• 1.2–2.0 g/kg of actual body weight — the range cited in the clinical nutrition literature as "desirable"
• 1.6–2.2 g/kg of goal body weight — a different denominator entirely
• 0.7–1.0 g/lb of lean body mass — a third denominator, requiring a body composition scan to use

Four denominators, a twofold spread in the multiplier. No wonder people give up and guess.

#Why the sources disagree

There are two separate reasons, and they compound.

The denominator problem. Protein targets were originally developed in populations without obesity, where body weight and lean mass track together reasonably well. In someone carrying substantial excess fat, actual body weight inflates the target — fat tissue doesn't require protein to maintain. So clinicians reach for workarounds: goal weight, adjusted body weight, or lean mass. Each workaround produces a defensible number. They just aren't interchangeable, and almost nobody states which one they're using.

The evidence problem. This one is more uncomfortable. The trials that established how well these drugs work largely did not measure what people ate.

So when a website confidently gives you a GLP-1 protein target, it is almost always extrapolating from the general weight-loss, older-adult or sarcopenia literature. That extrapolation is reasonable. It is not the same as direct evidence, and the honest sources say so.

#What the consensus documents actually say

Strip out the content marketing and the picture is more settled than the spread suggests. Current GLP-1 consensus guidance recommends individualised higher-protein eating at roughly 1.2 to 1.5 grams per kilogram of actual body weight per day during active weight loss.

That figure has support behind it. Meta-analytic evidence from weight-loss interventions in adults with overweight or obesity indicates that higher protein intake attenuates muscle mass loss, with intakes above about 1.3 g/kg/day associated with better preservation than lower intakes.

So: for a 90kg person, roughly 108 to 135 grams a day. For a 70kg person, roughly 84 to 105. If you have kidney disease, this conversation belongs with your doctor rather than a blog.

The higher figures aren't wrong so much as borrowed — they come from contexts (resistance-trained athletes, aggressive cutting phases) that don't describe most people on a weight-loss injection.

#The number that matters more than your target

Here is the statistic that reframes the entire question. In a cross-sectional study of current GLP-1 receptor agonist users, mean protein intake was 77.3 grams per day — against estimated requirements spanning 74 to 169 grams across participants. Only 43% were reaching even the minimum 1.2 g/kg/day.

Read that again. Well over half of people on these medications are not hitting the floor, let alone an optimised target.

Which means the argument about whether the right answer is 1.2 or 1.6 or 2.2 is, for most people, entirely academic. You are not choosing between good and optimal. You are somewhere below the floor, and the gap between where you are and 1.2 is much larger than the gap between 1.2 and 2.2.

#Why hitting it is structurally hard

This is not a discipline problem, and framing it as one is both wrong and unhelpful. The medication works by suppressing appetite. Reported reductions in energy intake range from about 16% to 39%. When total food intake falls that far and nobody is actively counselling you on composition, protein falls along with everything else.

Three things make it worse:

Gastrointestinal symptoms narrow what you can tolerate. In one cohort of current users, 53.7% reported nausea, 30.3% fatigue and 27.8% diarrhoea. Meat and heavy foods are frequently among the first things that become unappealing — and they're the protein-dense ones.

Protein ends up bunched at night. In a secondary analysis of that same cohort, roughly 40% of daily protein was eaten at dinner. Low appetite pushes people toward a single larger evening meal, which is the least useful distribution: spreading protein across eating occasions makes it more likely each one reaches a threshold where it can be used for muscle.

Almost nobody gets professional support. Only 20% of users in that study reported being referred to a registered dietitian. Just 51% reported receiving any information about managing gastrointestinal side effects at all.

The finding that fewer than half of users reach the minimum target, despite that minimum being consensus guidance, reflects a structural difficulty inherent to the treatment — not a lack of awareness.

#What actually helps

The practical measures that appear consistently across current clinical guidance are unglamorous and mostly about sequencing:

• Protein first, literally. Start each meal with the protein component before vegetables or starch. When you fill up after a few bites, the order you eat in decides what you got.
• Move protein earlier in the day. If 40% of your protein is arriving at dinner, breakfast is the highest-leverage meal to change.
• Use liquid or semi-solid sources during titration. Dose-escalation weeks are when solid food tolerance is lowest. Greek yoghurt, skyr, milk-based drinks and soups are easier to get down than chicken breast.
• Add resistance training, two to three sessions a week. This is the one co-intervention with randomised support: in the liraglutide-plus-exercise maintenance trial, exercise-containing arms preserved bone mineral density substantially better than medication alone, and improved stair-climb performance by close to 9% over pharmacotherapy alone. Protein and loading work together; neither substitutes for the other.
• Watch more than protein. The same reduced intake that lowers protein also lowers fibre, calcium, iron and vitamin D. In a retrospective analysis of 461,382 adults with type 2 diabetes newly prescribed these medications, diagnosed nutritional deficiencies occurred in 12.7% within six months and 22.4% within twelve.

#The part nobody addresses: how would you know?

Every article on this topic ends at "prioritise protein." None of them answer the obvious follow-up.

You ate half a chicken thigh and some yoghurt, felt full, and stopped. Was that 30 grams or 12? Across a day of small, irregular, appetite-suppressed eating, the error compounds fast — and the 43% figure exists precisely because people believe they're doing fine when they aren't.

This is a measurement problem, and it's the one traditional food logging handles worst. Database-search apps assume discrete meals at normal times in normal amounts. GLP-1 eating is the opposite: four bites at 11am, half a portion at 2pm, a yoghurt before bed. Logging each of those through a search screen costs two to four minutes a go, which is exactly the friction that makes people stop tracking within a fortnight.

Voice logging exists for this shape of problem. You say "half a chicken thigh and a small greek yoghurt" and get an estimate in about five seconds, without opening a database. The estimate is directional rather than laboratory-grade — but directional and actually recorded beats precise and abandoned, and knowing you landed near 70g when you were aiming for 120g is the information that changes what you do tomorrow.

#If you stop the medication

Worth knowing while you're building the habit rather than after. A 2026 systematic review and meta-analysis of 37 studies covering more than 9,300 people found that people regained a mean of 0.4kg per month after discontinuation, tracking toward baseline weight within roughly 1.7 years. Regain was faster after stopping medication than after stopping behavioural interventions alone.

The reading of that isn't "never stop." It's that pharmacotherapy without integrated nutritional and behavioural support produces a more fragile result. The protein habit, the resistance training and the awareness of what you're actually eating are the parts that stay with you when the prescription doesn't.