The short version
- In a meta-analysis of nine studies and 394 young men, HMB produced a 9.9% lower-body strength benefit in untrained lifters but trivial strength effects in trained lifters, with a trivial 3.7% overall.
- The same meta-analysis found HMB's effects on fat mass and fat-free mass inconsequential, and its effects on the muscle damage marker creatine kinase unclear.
- A 2003 meta-analysis of 250 supplements found only creatine and HMB significantly increased lean mass and strength, at 0.36 and 0.28% per week of lean mass respectively.
- In ten trials of 596 older adults, adding HMB to resistance training improved handgrip strength borderline-significantly (SMD 0.24, p = 0.05) and physical performance moderately (SMD 0.54, p = 0.01), with no effect on lean mass or gait speed.
- The 2025 ISSN position stand puts the HMB dose at 38 mg per kilogram of bodyweight daily and reports that benefits in trained athletes are mixed while effects in untrained and aging populations are more consistent.
Every few years a supplement arrives with a mechanism so tidy that the trials feel like a formality. HMB — beta-hydroxy-beta-methylbutyrate — is the long-running example. It is a metabolite of leucine, it appears to both raise muscle protein synthesis and suppress muscle protein breakdown, and that dual action is exactly what a lifter would order if mechanisms were on a menu. The problem is what happened when it was tested in lifters.
The 2003 result that built the category
Nissen and Sharp screened 250 supplements for a meta-analysis and found only six with more than two studies meeting their criteria. Two of the six cleared significance: creatine and HMB. HMB produced net lean mass gains of 0.28% per week and strength gains of 1.40% per week; creatine produced 0.36% and 1.09%. Chromium, DHEA, androstenedione and protein did not significantly affect either outcome.
That paper is where most of the marketing still lives. It is a real result and it is over twenty years old, and the field spent the following decade finding out which population it applied to.
What happened when the studies were split by training status
Rowlands and Thomson meta-analysed nine qualifying studies yielding 14 comparisons across 394 subjects, averaging about five weeks of intervention, and separated trained from untrained lifters. In untrained men there were small benefits: lower-body strength 9.9%, average strength 6.6%, and a negligible 2.1% for upper body. In trained lifters, all strength outcomes were trivial. Combined across every study, the overall average strength increase was trivial at 3.7%, though the authors noted the uncertainty leaves room for a small benefit. Effects on fat and fat-free mass were inconsequential, and the muscle damage results were unclear.
That is a specific and awkward pattern: the supplement helps most in people who would have improved anyway, doing almost anything. It is the same shape you see with most training supplements when the beginner effect is not controlled for, and it is the reason "significant in a meta-analysis" is not the end of a sentence.
The population where it may genuinely matter
The 2025 ISSN position stand is the most complete summary available and it is more measured than the headline claims. It puts the effective dose at 38 mg per kilogram of bodyweight daily, states that HMB may improve strength and power in untrained individuals while benefits in trained athletes are mixed, and notes that those benefits increase with study duration beyond six weeks. Its two most interesting points are elsewhere: that HMB may be important in a sedentary and aging population for muscle strength, functionality and quality, and that it may be effective in countering disuse atrophy during periods of inactivity due to illness or injury.
The older-adult claim now has a meta-analysis behind it, and it is a modest one. García-Alonso and colleagues pooled ten trials in 596 older adults comparing resistance training plus HMB against resistance training alone. Handgrip strength improved borderline-significantly (SMD 0.24, 95% CI 0.00 to 0.48, p = 0.05) and Short Physical Performance Battery scores moderately (SMD 0.54, 95% CI 0.12 to 0.95, p = 0.01). There was no significant effect on gait speed, appendicular lean mass, muscle quality, fat mass or body weight. Five of the ten trials were rated at high risk of bias, which the authors say limits confidence in the pooled estimates. They are right, and we would rather quote them saying it than not.
Who paid for the evidence
This matters more here than in most categories, and the position stand discloses it openly rather than burying it. Its first author is employed by a company that conducts HMB research and is listed as an inventor on active HMB-related patents; a second author is a former employee and current consultant of the same company and is also named on HMB patents; several other authors list HMB patents or industry funding. None of that makes the document wrong. It does mean the most favourable summary of HMB in the literature was written largely by people with a stake in HMB, and that the independent meta-analyses read cooler.
What we'd actually tell you
If you are a trained lifter chasing size or a number on a bar, the meta-analytic answer is that HMB does close to nothing for you, and your money buys more elsewhere. Progressive overload and adequate protein are doing the work; the leucine threshold argument covers why a leucine metabolite is unlikely to beat the leucine you would get from eating. HMB is also a cleaner idea than BCAAs but lands in a similar place: a real mechanism, a real but small effect, and a claim that outran the trials.
If you are older, training against age-related muscle loss, or coming off an injury or a period of forced inactivity, the case is genuinely different and worth raising with whoever is managing your rehabilitation. That is the population the ISSN singles out and the one the newest meta-analysis found a functional signal in — modest, borderline on two outcomes, and drawn from a literature that is half high-risk-of-bias. Honest is the most this evidence supports.
Good questions
Should I take HMB if I already lift regularly?
Probably not. The meta-analysis that separated trained from untrained lifters found all strength outcomes trivial in trained men, and inconsequential effects on body composition. The clear benefits in that analysis appeared in untrained lifters, who improve on almost any stimulus. If you have been training consistently for a year or more, this is one of the easier supplements to skip.
Is HMB just an expensive form of leucine?
It is a metabolite of leucine, not a substitute for it, and the position stand argues its activation of the mTORC1 pathway is independent of the leucine-sensing route. That is a mechanistic distinction rather than a practical one for most people. If your daily protein is already adequate, you are producing HMB from the leucine you eat, and the supplemental version has not reliably beaten that in trained lifters.
Can HMB help while I am injured or can't train?
This is the strongest version of the case for it. The 2025 ISSN position stand states HMB may be effective in countering muscle disuse atrophy during periods of inactivity due to illness or injury, and that it may matter in sedentary and aging populations for strength and function. It is worth raising with whoever is managing your rehabilitation rather than deciding alone.
How much HMB is an actual dose?
38 mg per kilogram of bodyweight daily is the figure the ISSN position stand uses, which works out at about 3 g for an 80 kg adult. Two forms exist, calcium HMB and a free acid form, and the free acid form appears to raise blood HMB more, though the position stand notes recent results are mixed. Chronic use up to at least one year is described as safe.
Does HMB work for older adults?
Modestly, on function rather than on muscle mass. A 2025 meta-analysis of ten trials in 596 older adults found adding HMB to resistance training improved handgrip strength borderline-significantly and physical performance battery scores moderately, with no significant effect on appendicular lean mass, gait speed, muscle quality or fat mass. Five of the ten trials were at high risk of bias, so the authors themselves urge caution.
Who funded the research that says HMB works?
A lot of it, and the field is unusually open about it. The 2025 ISSN position stand's first author is employed by a company that runs HMB research and is named on active HMB patents, a second author is a former employee and current consultant of the same company, and several co-authors disclose HMB patents or industry funding. That does not make the document wrong, but the independent meta-analyses reach cooler conclusions, and you should know both exist.
Sources
- Rowlands DS, Thomson JS. Effects of beta-hydroxy-beta-methylbutyrate supplementation during resistance training on strength, body composition, and muscle damage in trained and untrained young men: a meta-analysis. J Strength Cond Res, 2009. View study
- Nissen SL, Sharp RL. Effect of dietary supplements on lean mass and strength gains with resistance exercise: a meta-analysis. J Appl Physiol, 2003. View study
- Rathmacher JA, Pitchford LM, Stout JR, et al. International society of sports nutrition position stand: β-hydroxy-β-methylbutyrate (HMB). J Int Soc Sports Nutr, 2025. View study
- García-Alonso A, Sánchez-González JL, Navarro-López V, Méndez-Sánchez R, Polo-Ferrero L. The Role of HMB Supplementation in Enhancing the Effects of Resistance Training in Older Adults: A Systematic Review and Meta-Analysis on Muscle Quality, Body Composition, and Physical Function. Nutrients, 2025. View study
These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease.
