Resistance training is the intervention, not the accessory

Exercise is not required for a GLP-1 to work. Appetite suppression and weight loss happen without it. That is exactly why the training question gets deferred, and why deferring it is the most consequential mistake in the programme.

The size of the problem

Published body composition analyses put roughly 26% to 40% of weight lost during GLP-1 treatment in the lean mass column, and the figure varies by drug: STEP 1 came in near 45%, SURMOUNT-1 near 26%. The exact proportion moves with diet, training, population and genetics. It is a range, not a constant, and the low end of that range is still a great deal of muscle.

What makes it a durability problem rather than a cosmetic one is what happens next. Muscle is metabolically active, so every pound lost lowers resting metabolic rate, and a lower metabolic rate makes weight harder to hold once medication is reduced or discontinued. The muscle you fail to protect during treatment is the reason maintenance fails after it. For patients past 50, the same loss also feeds directly into sarcopenia, falls, and functional decline.

What the guidance supports

The American College of Sports Medicine’s 2009 position stand on progression models is the origin of the familiar parameters: 2 to 4 days per week at moderate intensity, 8 to 12 repetitions per set, 2 to 4 sets per exercise. In practice, two to three full-body sessions a week is sufficient for most people starting out.

Worth stating, because it changes how confidently these numbers should be quoted: ACSM issued a new position stand in 2026, its first major update in seventeen years, drawing on 137 systematic reviews. Its headline finding is that the largest gain by far comes from moving from no resistance training to any resistance training, and that equipment choice, training to absolute failure, and complex periodisation do not consistently change outcomes for the average adult. The specific rep and set numbers matter less than the older guidance implied. That is good news for adherence and bad news for programmes that sell precision.

A workable beginner structure: two resistance days of 30 to 45 minutes built on compound movements (squats, push-ups or chest press, rows, lunges, overhead press, core), 2 to 3 sets of 10 to 12 reps, 60 to 90 seconds rest, with progressive overload over time. Two to three low-intensity cardio days of 20 to 45 minutes fill the rest. Bodyweight and bands are enough to begin; no gym membership is required, and the 2026 stand says so explicitly.

Where the two compete for the same hour, the resistance session wins.

Protein, and the timing question

Total daily protein is the load-bearing variable. A 2024 systematic review found intake above 1.3g per kilogram per day associated with preserved or increased muscle mass during weight loss, and intake below 1.0g/kg/day associated with decline; the AACE 2025 consensus statement recommends at or above 1.2g/kg/day for most individuals. Consuming 20 to 40g within 30 to 60 minutes of training has support, but it is a marginal gain layered on top of hitting the daily total: a bonus, not a requirement. Programmes that lead with timing and neglect the total have the priorities inverted.

Where the evidence runs out

This is worth stating precisely, because it is frequently overstated. The trials establishing lean mass loss (STEP 1, SURMOUNT-1) were designed to test medication efficacy. Neither was designed to optimise muscle preservation. The supporting evidence for resistance training as the mitigation comes from the broader exercise science literature applied to this context, not from a trial that randomised GLP-1 patients to training versus no training and measured the difference at scale.

That is a real gap, and it is a tractable one. It is among the more obvious studies waiting to be run.

Building meaningful new muscle in a caloric deficit also remains difficult. Preservation is the realistic goal during treatment; accrual becomes achievable in maintenance.

What we do with it

Training is not an optional adjunct bolted on alongside medication. It is the intervention that determines whether the medication produces a durable result or a temporary one, and it belongs in the plan from week one rather than at the point where a plateau forces the conversation.

One practical caveat: exercise tolerance drops when nausea is worst, typically in the first weeks or after a dose increase, and usually recovers within 2 to 3 weeks. Walking is always safe on the difficult days. Programmes that treat early non-adherence as a motivation problem rather than a tolerability one lose people unnecessarily.