About 1,700 people a month search for creatine as a supplement for seniors, the elderly or older adults, usually with one of three questions in mind: muscle loss, strength and falls, or bone. The trials in people over 50 do answer parts of all three, and the answer changes depending on one thing the headlines tend to leave out: whether the creatine came with resistance training.
This page reports what the trials and their meta-analyses measured, how large the effects were, who funded or ran them, and what is still open. It recommends nothing; the decision for any older adult belongs with them and their clinician.
With resistance training: small, consistent gains
Most older-adult trials give creatine or placebo to people who are also starting a supervised strength programme, so the comparison is training plus creatine against training alone.
| Review | Trials and people | What creatine added to training |
|---|---|---|
| Devries & Phillips, 2014 | Randomised placebo-controlled trials, 357 older adults, mean age ≈64, 12.6 weeks of training on average | More fat-free mass and total body mass, no change in fat mass; greater chest-press and leg-press 1RM; better 30-second chair stand. No difference in knee-extension or biceps 1RM, or isometric and isokinetic knee torque |
| Forbes et al., 2021 | Updated meta-analyses by dosing strategy | Lean tissue mass and strength gains overall; strength advantages disappeared when trials with a loading phase were excluded |
| Naddafha et al., 2026 | 7 RCTs, 608 postmenopausal women, mean age ≈62, 12–104 weeks | Lean mass +0.37 kg (95% CI 0.05 to 0.69); leg-press 1RM +7.5 kg; benefit when ≥5 g a day was combined with training; no measurable effect at ≤3 g a day without training; bone density unchanged |
| Davies et al., 2024 | 33 RCTs, 1,076 older adults and people with chronic disease | Sit-to-stand SMD 0.51 (95% CI 0.01 to 1.00); upper-body and handgrip strength small gains; lean mass +1.08 kg. Evidence quality low or very low for every outcome |
Read together, the size is the point. Fractions of a kilogram of lean tissue and single-digit kilograms on a leg press, measured by DXA scan rather than by imaging the muscle itself, on top of the larger gains the training produced in both groups. The same gap between DXA lean mass and directly imaged muscle runs through the younger-adult literature and is set out on creatine for muscle growth.
Who ran them. The Forbes 2021 authors disclose creatine advisory roles, industry-sponsored creatine research and product donations; the 2026 review's authors include the chair of a creatine scientific advisory board supported by a creatine manufacturer, which also paid part of the publication fee. The reviews point the same way whoever wrote them, which is worth saying, and so is the fact that the largest by trial count (Davies 2024, 33 trials) rated the evidence low or very low quality.
Without training: no effect found
The cleanest test of creatine alone in older people is a two-year Brazilian trial: 200 postmenopausal women with osteopenia, 3 g a day or placebo, no exercise programme (Sales et al., 2020). Bone density fell over time in both groups, with no difference. Lean mass and appendicular muscle mass rose in both groups, again with no difference (p = 0.731 and 0.397). Falls, fractures, bone markers and laboratory safety measures did not differ. The authors' conclusion is unusually direct: the result "refutes the long-lasting notion" that creatine alone has bone-building or muscle-building properties in the long run.
That trial is also where the 2026 review's dose split comes from: the trials at 3 g a day or less without training are the ones that showed nothing.
Bone: two two-year trials, no density change
| Trial | People | Creatine | Bone mineral density vs placebo | Other findings |
|---|---|---|---|---|
| Sales et al., 2020 | 200 postmenopausal women with osteopenia, São Paulo | 3 g a day, no training, 2 years | No difference (spine, femoral neck, total femur) | No difference in falls or fractures, which were secondary outcomes |
| Chilibeck et al., 2023 | 237 postmenopausal women, mean age 59 | 0.14 g per kg a day (≈10 g at 70 kg), with resistance training 3×/week and walking 6×/week, 2 years | No difference at femoral neck (primary outcome), total hip or lumbar spine | Femoral-neck section modulus and buckling ratio preserved (secondary); 80 m walk ≈1.5 s faster; no difference in bench or hack-squat 1RM; lean mass higher only in a completers sub-analysis |
Two well-sized, two-year trials, two null results on the primary bone outcome. The geometry finding in the second is a secondary measure from one trial and has not been repeated. Bone is the place where the creatine-for-ageing story has run furthest ahead of its trials. How the menopause-specific claims compare is on creatine for women.
Function, falls and independence
The outcomes older adults care about most are the least studied. The chair-stand and sit-to-stand improvements above are real measurements, but small, and the 2024 review gave a posterior probability of only 66.7% that creatine improves physical function at all. No trial found here measured falls or fractures as a primary outcome; the Brazilian trial counted them as secondary and found no difference. None of the trials above was run in care homes.
Safety in the trials
The 2026 postmenopausal review reported adverse events similar to placebo and renal indices unchanged; the two-year 3 g trial found no change in its laboratory safety panel. Those trials largely excluded people with kidney disease, which is common in older age, so they say nothing about it. Creatine also raises blood creatinine, the number used to estimate kidney function, which can confuse a routine blood test; that is explained on creatine in bloodwork, and the kidney trials are on creatine and the kidneys. Medicines common in later life, and how creatine sits with medical conditions, are on creatine and medical conditions. Any older adult taking regular medication or with a kidney, heart or diabetes diagnosis is the case where the decision belongs with a clinician.
What is not established
- Falls and fractures. No trial has used them as the primary outcome.
- People with diagnosed sarcopenia or frailty. Most participants were healthy or mildly limited volunteers able to train.
- Older men. The newest and largest trials summarised here are in postmenopausal women.
- Creatine without training at higher amounts. The one long trial of creatine alone used 3 g a day.
- Cognition. Claims about the ageing brain are a separate literature, graded on creatine and the brain.
The wider evidence grades for creatine by outcome, of which older adults are one row, are on creatine benefits.
Sources and dates
- Devries MC, Phillips SM. Creatine supplementation during resistance training in older adults — a meta-analysis. Med Sci Sports Exerc 2014;46:1194–203, PMID 24576864.
- Forbes SC, Candow DG, Ostojic SM, Roberts MD, Chilibeck PD. Meta-analysis examining the importance of creatine ingestion strategies on lean tissue mass and strength in older adults. Nutrients 2021;13:1912, PMID 34199420 (conflict-of-interest statement read).
- Naddafha S … Stout JR. Creatine monohydrate for lean mass, strength, and bone density in postmenopausal women: a systematic review and meta-analysis. J Int Soc Sports Nutr 2026;23:2668435, PMID 42141930 (abstract and disclosures read 2026-10-05).
- Davies TW … Puthucheary Z. Creatine supplementation for optimization of physical function in the patient at risk of functional disability: a systematic review and meta-analysis. JPEN 2024;48:389–405, PMID 38417175.
- Sales LP … Rodrigues Pereira RM. Creatine supplementation (3 g/d) and bone health in older women: a 2-year, randomized, placebo-controlled trial. J Gerontol A Biol Sci Med Sci 2020;75:931–8, PMID 31257405; registry NCT01472393.
- Chilibeck PD … Zello GA. A 2-yr randomized controlled trial on creatine supplementation during exercise for postmenopausal bone health. Med Sci Sports Exerc 2023;55:1750–60, PMID 37144634.
- PubMed census, 2026-10-05: creatine in title with older/elderly/aging/sarcopenia/postmenopausal in title, 24 records tagged meta-analysis and 222 tagged randomised controlled trial.