What protects your bones

What prevented falls and broken bones in trials, and where vitamin D fits. Every number links to its source.

The short version

Exercise that trains balance cut falls by about a quarter, and may cut broken bones from falls too.
Heavy strength and impact training built bone at the spine in women with thin bones.
If a scan shows osteoporosis, the medicine alendronate probably cut spine fractures by about half.
In care homes, vitamin D with calcium cut hip fractures a little, and hip protectors probably did too.
On its own, vitamin D didn't prevent broken bones in healthy midlife and older adults, and one huge yearly dose backfired.

Exercise that cut falls

At least a third of people over 65 who live at home fall each year. A 2019 review of 108 trials with 23,407 people, average age 76, found that exercise cut the number of falls by 23%, and the reviewers were highly confident in that (Sherrington, 2019). Among 1,000 people who'd have about 850 falls between them in a year, that's about 195 fewer falls, and 72 fewer people falling at all.

Exercises that train balance and everyday movements cut falls by 24%.
Programmes that mixed those with strength exercises probably cut falls by about a third.
Tai chi may cut falls by about a fifth.
The reviewers couldn't tell whether strength training on its own, dancing or walking prevent falls.

Fewer falls should mean fewer broken bones, and the trials point that way, though with less certainty. In the same review, 10 trials with 4,047 people had 27% fewer people breaking a bone in a fall, and the reviewers were less sure of that result. A 2013 review of 6 trials counted falls rather than people, and found 61% fewer falls that broke a bone (El-Khoury, 2013).

Where the 2019 review's trials reported side effects, they were mostly minor muscle and joint problems.

Strength and impact training

The LIFTMOR trial split 101 women past menopause with thin bones, average age 65, into two groups. 49 did 30 minutes of supervised heavy lifting and impact training twice a week for 8 months, lifting more than 85% of the heaviest weight they could manage once, in 5 rounds of 5. The other 52 did gentle exercise at home (Watson, 2018).

Bone density at the spine rose 2.9% in the training group and fell 1.2% in the home-exercise group.
At the hip, it held steady (up 0.3%) in the training group and fell 1.9% in the home-exercise group.
The training group also did better on all the strength and movement tests. The only side effect reported was one minor back spasm.

A 2021 review of 53 trials in healthy women past menopause found that high-intensity training added the most bone at the spine, though only 4 of the programmes were high-intensity. Strength training, perhaps combined with impact training, appeared to work best at the spine and hip (Kistler-Fischbacher, 2021).

Whether that extra bone means fewer fractures hasn't been settled. In a 2011 review of 43 exercise trials, the fracture results were too uncertain to show a difference (Howe, 2011).

If a scan shows osteoporosis

Alendronate is a medicine that slows the breakdown of bone. In a 2025 review, women past menopause at higher risk, meaning osteoporosis, a past spine fracture, very thin bones on a scan or age 75 and over, who took it for one to three years were probably about half as likely to have a spine fracture diagnosed. 24 of 1,114 women on alendronate did, against 51 of 1,055 on a placebo (a dummy pill), roughly 2 in 100 against 5 in 100 (Wells, 2025). It may also have cut hip and wrist fractures by about half. In women at lower risk, it may have cut spine fractures, but might make little or no difference to hip and wrist fractures.

Where calcium and vitamin D still help

Vitamin D and calcium taken together slightly lowered the risk of breaking a hip. In 9 trials with 49,853 older people, hip fractures were 16% less common (Avenell, 2014). The review worked out what that means at different levels of risk:

In care homes, where about 54 in every 1,000 people break a hip each year, that's about 9 fewer hip fractures per 1,000 a year.
Among people living at home, where about 8 in 1,000 do, it's about 1 fewer.

The pills also brought slightly more stomach upsets and kidney problems. A 2022 review of 32 earlier reviews found the benefit may come mainly from people in care homes. The reviews that looked only at people living at home found no drop in fractures (Chakhtoura, 2022).

Hip protectors

Hip protectors are pads worn in special underwear. In 14 trials with 11,808 people in care homes, they probably cut hip fractures slightly, with about 11 fewer people per 1,000 breaking a hip over the trials. They may slightly raise the small risk of a broken pelvis. In 5 trials of people living at home, they probably made little or no difference. Across the trials, many people offered them didn't keep wearing them (Santesso, 2014).

Vitamin D on its own

The VITAL trial gave 25,871 US adults aged 50 and over 2,000 IU of vitamin D3 a day or a placebo for about five years. Of roughly 12,900 in each group, 769 on vitamin D broke a bone, against 782 on placebo, and starting with low vitamin D didn't seem to matter (LeBoff, 2022).

People who broke at least one bone in about 5 years, out of about 12,900 per group

A 2018 review pooled 81 trials with 53,537 people. Vitamin D made no difference to fractures or falls, and no meaningful difference to bone density. Higher doses did no better than lower ones (Bolland, 2018).

Huge doses backfired

A trial split 2,256 women aged 70 and over, all at high risk of fracture, between a placebo and one 500,000 IU dose of vitamin D a year, in autumn or winter, for three to five years. The women on vitamin D fell 15% more often, and there were 171 fractures against 135 (Sanders, 2010).

In a Canadian trial, 311 healthy adults aged 55 to 70 took 400, 4,000 or 10,000 IU a day for three years. Bone density in the forearm fell 1.2% on 400 IU, 2.4% on 4,000 IU and 3.5% on 10,000 IU (Burt, 2019). Bone strength, estimated from the scans, didn't clearly differ.

Winter doses, and whether to get tested

The NHS advises everyone in the UK to consider 10 micrograms (400 IU) of vitamin D a day in autumn and winter, and advises people at high risk of running low to take it all year. It says not to take more than 100 micrograms (4,000 IU) a day (NHS). That advice is about avoiding a deficiency. It isn't a claim about fractures.

The Endocrine Society, a professional society of hormone specialists, suggests that healthy adults under 75 without a medical reason don't take more than the recommended daily amount to prevent disease, and skip routine blood tests. That goes for people with obesity or darker skin too. It found no trial evidence that testing helps. For people over 50 who need vitamin D, it suggests a daily dose rather than occasional big ones (Demay, 2024).

Where to start

If you're over 60, exercise that trains balance has the strongest evidence here, and tai chi counts.
If a scan has shown thin bones, heavy lifting built bone in a trial, with a coach watching every session.
If you've broken a bone in a minor fall, ask your doctor about a bone scan and about alendronate.

How far to trust these numbers

The results here come from trials and reviews of trials, most of which assigned people by chance, so the groups were comparable.
The overall falls result is solid. Tai chi and mixed programmes rest on less certain evidence, and the fracture results for exercise on fewer trials.
LIFTMOR is one trial of 101 women. If you have thin bones, don't start heavy lifting without someone watching your form.
Most people in VITAL weren't low in vitamin D to begin with, so it says little about treating a deficiency.
The vitamin D and calcium result, and the hip protector one, may rest mostly on people in care homes, who break hips far more often than people living at home.