A Quarter of What You Lose Is Muscle. Lifting Is What Changes That.

The short version

You are eating far less than you used to, and nobody handed you a plan. Here is what the research actually supports, in the order it matters.

The headline you have probably absorbed — that these drugs strip muscle — does not survive the comparison. A 2026 meta-analysis pooled 20 randomized trials and 15,782 participants, measuring body composition by DXA or MRI, and found the share of weight lost as lean mass was statistically indistinguishable between the drugs and intensive lifestyle programs².

The comparison between the drug rows and the lifestyle row came out at p = 0.42 — no meaningful difference². Losing a lot of weight costs lean mass whichever way you do it.

One row is different, and it is the one with a barbell in it.

This is educational only and not medical advice.

So the first answer is not food

If you have twenty minutes a week to spend on this, spend them lifting before you spend them meal-prepping.

Resistance training is the only intervention in that table that changed the number, cutting the lean-mass share by roughly a third. The 2026 EASO, EFAD and ECPO consensus statement — a joint European position from obesity clinicians, dietitians and a patient organization — reaches the same place, recommending adequate protein and progressive resistance exercise together to preserve fat-free mass and physical function¹.

Progressive is the operative word: the load has to go up over time. Two short sessions a week that get incrementally harder beat five that never change. Preserving muscle on a busy schedule covers how to fit that around a workday.

Then protein — with an honest caveat

Every article on this subject gives you a grams-per-day target. Almost none of them tell you where the number came from, and most are borrowed from bariatric surgery guidance rather than derived from anyone on these drugs.

Here is what has actually been measured. In the 2026 CRAVE study, 28 adults starting semaglutide or tirzepatide were followed for 24 weeks in a real-world clinic with no structured nutrition support. Higher absolute protein intake was associated with greater preservation of skeletal muscle mass — but the correlation was r = 0.41 at p = 0.0498, which clears the significance threshold by a hair, in 28 people, in a study its own authors describe as exploratory and hypothesis-generating³.

That is a real signal and a thin one. Prioritizing protein is sensible and low-risk. Treating a specific gram target as established is not, and this page will not invent one for you. Ask your prescriber or a dietitian for a number tied to your body weight rather than taking one off a blog.

The finding that should actually change your grocery list

CRAVE measured something more useful than protein, and it is the part nobody talks about.

Over 24 weeks, total energy intake fell sharply — as designed — and intake fell across multiple micronutrients at the same time. Diet quality did not improve. Food cravings did not change overall³.

Read that carefully, because it contradicts the story most people tell themselves at the start. The drug shrinks how much you eat. It does not, on its own, improve what you eat. Left alone, a smaller version of the same diet is what you get — and a smaller version of the same diet is a smaller supply of everything in it.

This lands on a starting point that is already imperfect. A 2026 NHANES analysis of 16,143 adults found that 43.5% of US adults meet the prescribing criteria for these medications, and that this group had modestly poorer diet quality and modestly higher inadequacy for vitamin C, vitamin A, magnesium, vitamin E, vitamin D, fiber, potassium and vitamin K⁴. Notably, that gap largely disappeared after adjusting for waist circumference⁴ — it tracks with central adiposity rather than with willpower, which is worth knowing if you have ever been told the reverse.

The practical version

• Lift twice a week, progressively. It is the only lever in the evidence that changed the outcome².
• Put the protein first on the plate. When appetite is small, whatever you eat first is what you actually finish.
• Density over volume. Fewer calories means every bite has to carry more nutrition than it used to — the micronutrient decline is the measured risk here, not the calorie decline³.
• Do not assume the drug fixed your diet. It did not in the study where nobody intervened³.
• Ask for a nutrition referral, not a printout. The consensus statement treats medical nutrition therapy as part of the treatment rather than an optional extra¹.
• If eating has become fraught, say so. The same consensus flags shifts in food reward, coping and social connection, and recommends psychological screening with support where needed¹. That is in a clinical guideline, not a wellness column.

If nausea is what is shaping your eating rather than any of the above, the side-effect quick-fix reference is the more useful page today.

The efficient takeaway

The muscle you lose is mostly a function of losing weight, not of the drug you used to do it — that is what 20 trials and 15,782 people say. Resistance training is the intervention with real evidence behind it, protein is sensible on thinner evidence, and the risk nobody warns you about is not muscle at all: it is quietly eating less of every nutrient because you are quietly eating less of everything. None of that requires a meal plan. It requires two sessions a week and a plate that earns its space.