Exercise for sarcopenia has a sweet spot, and more stops helping

Sarcopenia, the progressive loss of muscle strength and physical function with age, leaves people weaker, slower and more vulnerable, and exercise is the frontline response. Yet how much exercise, and which type, remain surprisingly open questions. Most guidance treats dose as if more were straightforwardly better, and rarely asks whether grip strength and everyday mobility respond to the same recipe. This review set out to map how dose and exercise type interact, and to test whether the relationship between dose and benefit is actually linear at all.

The authors searched the Cochrane Library, Web of Science, PubMed, Embase and SPORTDiscus through 5 February 2026 for randomised controlled trials of exercise in middle-aged and older adults with sarcopenia. Twenty-nine trials met the criteria. Exercise doses across different programmes were standardised using metabolic equivalents so they could sit on one comparable scale. A Bayesian network meta-analysis then modelled non-linear dose-response curves for two outcomes, handgrip strength and functional mobility, and compared how different exercise types performed across dose ranges.

Overall exercise dose traced an inverted U shape for both outcomes. Handgrip strength began improving from a dose of 140 metabolic equivalent-minutes, with the largest effect at 690 and an optimal range of 560 to 830. Functional mobility followed a similar curve, improving from 140 and peaking at 830. When results were broken down by exercise type, whole-body vibration produced the largest grip strength effect at a dose of 280, while functional mobility showed a stable peak only under resistance training at 580 metabolic equivalent-minutes.

The inverted U pattern complicates the common assumption that total exercise volume is what counts, and it supports the idea that different functional goals may need different prescriptions. It extends earlier work by modelling dose continuously rather than comparing broad categories. Open questions remain, though: the optimal ranges are model estimates rather than doses tested head to head in trials, and it is unclear how far the findings travel beyond people who already have sarcopenia. Individual variation in response also sits outside what this analysis can answer.

For anyone thinking about strength and mobility in later life, this is a reason to believe the middle of the dose range, not the top of it, is where the benefit concentrates. It also suggests the goal matters: grip strength and getting around responded best to different combinations of exercise type and dose in these trials. None of this fixes a personal prescription, and the exact figures are estimates, but it is evidence that dose and modality both deserve attention rather than total volume alone.