Can strength training protect your bones after menopause?

Does Lifting Weights Build Bone After Menopause? 17 Trials in 690 Women

2026-11-01 · By Anthony Falcone

Verdict: Keep it, with a caveat. Yes, at the hip, and it's the steadiest finding here: across 17 randomised trials, women who did resistance training ended up with better hip bone density than women who didn't. The spine and femoral neck results looked bigger but varied hugely between trials, the trochanter showed no clear effect, and the review didn't look at fractures.

What the researchers did

Bone loss speeds up after menopause as oestrogen falls; the paper cites an estimate that about 40% of women over 50 will have an osteoporotic fracture in their remaining lifetime. Resistance training is often recommended to slow that down, but advice on how heavy, how often and for how long varies.

This team pooled every randomised trial they could find, up to March 2025, in which healthy postmenopausal women aged 50 or over did resistance training for at least 4 weeks and were compared with women who kept to their normal activity. Women with high blood pressure, diabetes or heart disease were excluded. Bone density had to be measured by DXA, the standard bone scan.

That gave 17 trials with 690 women, 375 in training groups and 315 controls, from 11 countries, published between 1994 and 2024. The programmes lasted 24 to 52 weeks, two or three sessions a week, at loads from 40% to 85% of 1RM (the most you can lift once). The researchers looked at four sites: the lower (lumbar) spine, the femoral neck (the narrow top of the thigh bone), the total hip and the trochanter (the bony bump on the outside of the hip).

What they found

Compared with controls, training improved bone density at three of the four sites. At the total hip the effect was small but consistent (SMD 0.30, 9 trials, low variation between them). At the femoral neck (SMD 0.89, 15 trials) and the lower spine (SMD 0.88, 13 trials) the pooled effects were larger, but the trials disagreed a lot with each other (I² of 87% and 91%). At the trochanter there was no clear effect (SMD 0.23, P = 0.06).

An SMD is a standardised effect size, measured in units of how much people differ from each other rather than in grams or percent, so the hip effect is the smallest of the three. The paper doesn't translate the results into how much bone the women gained or kept.

The spine result is the shakiest. When the authors removed trials one at a time, taking out a single trial changed the spine result a lot, while the femoral neck and total hip results held up.

How heavy, how often

The authors split the trials by how they trained. With heavier loads, at least 70% of 1RM, the effects at the total hip and femoral neck were significant. With lower or moderate loads, none of the four sites showed a significant effect, which fits the idea that bone needs a strong enough strain to respond.

Three or more sessions a week gave significant effects at all four sites, the trochanter included. With fewer than three sessions a week, only the femoral neck result was significant.

Trials of 48 weeks or longer showed significant effects at the femoral neck and total hip; shorter trials showed them at the spine and femoral neck. Sessions of 40 minutes or more did better at the spine.

Treat these splits as hints, not rules: some subgroups held only one or two trials, and the variation between trials stayed high for the spine and femoral neck even within subgroups. The authors' own suggestion is heavier training (at least 70% of 1RM), three times a week, kept up for longer.

What this means for you

If you're past menopause, resistance training is a reasonable way to look after your hips: the hip result was the most consistent in these trials.
Load matters. The benefit showed up with weights of at least 70% of the most you can lift once, not with light ones. Several of the trials started lighter and worked up over the first weeks.
Think in months, not weeks. The trials ran for 24 to 52 weeks, two or three times a week, and three sessions did better than two.
This doesn't show fewer fractures, and it isn't a replacement for a bone scan or treatment your doctor recommends. If you have osteoporosis, have had a fracture or take medication for your bones, ask your doctor which loads are safe for you first.

How much to trust it

The authors' own: only 17 trials, some subgroups based on very few of them, high variation between trials for the spine and femoral neck that the subgroups didn't explain, English-language papers only, and possible publication bias. They call the findings preliminary.
Small and mostly older trials: training groups had 7 to 37 women each, and many of the trials are from the 1990s and 2000s.
No fracture data. Bone density is a stand-in; none of the pooled results show whether training means fewer broken bones.
Results come as standardised effect sizes, so you can't read off how much bone density changed in grams or percent.
Some reporting slips: the methods describe one effect measure while the results use another, and two study counts in the text don't match the figures. The main numbers agree across the summary, results and tables.
Healthy women only, aged 50 or over, without high blood pressure, diabetes or heart disease. It doesn't tell you what's safe or effective with osteoporosis. The authors report no external funding and no competing interests.

The study

Zhao F, Su W, Sun Y, Wang J, Lu B, Yun H. Optimal resistance training parameters for improving bone mineral density in postmenopausal women: a systematic review and meta-analysis. Journal of Orthopaedic Surgery and Research, 2025. doi.org/10.1186/s13018-025-05890-1

Our Perimenopause Protocol builds a strength and impact plan for muscle and bone on the LIFTMOR trial and the ESSA guidelines, sets protein, calcium and vitamin D within EU limits, and comes with a 12-week plan and symptom diary. See the protocol