In brief
If you have ever been told that the only way to protect your bones is to lift heavy weights, here is a fact to hold against it. In the 1990s, Finnish researchers measured both arms of 105 national-level tennis and squash players and 50 women who didn't play. In the non-players, the two arms were within a few percent of each other. In the players, the racquet arm carried 8 to 16 percent more bone than the other arm, in the same woman, on the same diet, with the same hormones. Women who had started playing as adults still had more bone in the playing arm, just less of it1.
Nobody lifted a weight. They hit a ball, thousands of times, with sudden force, from angles the arm wasn't expecting. That is what bone responds to.
Our bone loss article says to lift heavy things, and that advice stands. Heavy lifting is the best-studied way to load bone in women past menopause. But it is not the only way, and the research is clearer about that than most advice admits. If you would rather be on a court, a field or a trail than under a barbell, this article is for you.
Bone is living tissue with its own sensors, and they follow three rules. Bone adapts to loads that move, not loads that sit. It needs only a short burst of loading to start rebuilding. And it gets used to whatever it feels every day, and stops responding to it.
Think of a thermostat. Nothing happens until the room crosses the line you set. Bone has a set point for strain: a load below it does nothing, a load above it turns the building on. And once bone has thickened enough to handle a load, the set point rises to match, so yesterday's big load becomes today's ignored one. Walking hasn't crossed your skeleton's line since you were a toddler.
How short a burst? In a 1984 experiment, researchers applied loads to a bird's wing bone at a normal strength but in an unusual pattern. Four loads a day stopped the bone from thinning. Thirty-six loads a day increased its mineral content by a third. Going from 36 loads a day to 1,800 added nothing more2. Bone counts the first few dozen and then stops counting.
So the recipe is a load big enough to cross the line, arriving fast, from a direction bone isn't braced for, and not much of it. Weights supply force. Landing supplies force and speed. A sport supplies all three.
The clearest ranking comes from a Finnish study that measured the hips of 255 elite female athletes across different sports and compared them with 30 women who didn't train. The researchers grouped the sports by how they load bone.
Two groups stood out. Athletes in high-impact sports (volleyball, hurdling) had hip bone density 23 percent higher than the non-athletes. Athletes in what the researchers called "odd-impact" sports, where the force comes from unusual directions (squash, soccer, speed skating, step aerobics), had 29 percent higher. Both groups also had wider, stronger hip bones, by about a quarter.
The surprise was the weightlifters. They had denser hips than the non-athletes, but no significant advantage in the measure of bone strength the study used. Runners and cross-country skiers, whose sports are repetitive and lower-impact, got about half the benefit of the impact groups. Swimmers and cyclists looked like the women who didn't train at all3.
Those are lifelong athletes, so the percentages are bigger than anyone will gain starting at 45. But the order is what matters. For the hip, the bone that matters most in old age, the winners were sports with landings and changes of direction. Tennis, squash, pickleball, soccer, netball, volleyball, dance with jumps, step classes. Running counts, at about half rate. Stairs count.
Walking is good for nearly everything, and this is the one place it lets you down. A review of eight trials of walking programs in women past menopause found no significant effect on spine density over six months to two years, and a borderline effect at the hip of about one percent4. Bone has been walking too long to be surprised by it.
Swimming and cycling are worse for bone, not because they harm it but because they never load it. In a 15-week trial in sedentary women averaging 45, one group played recreational soccer three times a week, another swam hard three times a week, a third swam at moderate effort, and a fourth did nothing. The soccer group's hip density rose 1.7 to 2.4 percent and their bone-building markers jumped by a third to a half. Neither swimming group changed at all5. Swim and cycle for your heart and your mood. Count them as zero for bone.
The most encouraging studies are the smallest. In a UK trial, 61 premenopausal women hopped on one leg, 50 hops a session, on zero, two, four or seven days a week for six months. The other leg was each woman's own control. Hopping twice a week did nothing. Four days a week added about one percent at the hip. Every day added 1.8 percent, in six months, from about a minute of hopping6. Daily mattered more than anything.
Here is the honest part. In 1998, British researchers gave the same program, 50 vertical jumps a day, six days a week, to two groups of women. Before menopause, it added 2.8 percent at the hip in five months. After menopause, over 12 and 18 months, it did nothing measurable, whether or not the women were on hormone therapy7.
Falling estrogen changes how bone reads its sensors. Impact still works after menopause, but it stops working on its own.
What does work is pairing it. LIFTMOR, the Australian trial that made "lift heavy" famous, is itself a pairing: its title is "high-intensity resistance and impact training," and the program included jumping chin-ups with drop landings alongside the deadlifts and squats8. The longest-running evidence points the same way. In Erlangen, Germany, 137 early-postmenopausal women with low bone mass were followed for 16 years. Those who did a mixed program of impact and strength work lost 1.5 percent of spine density over that time, against 5.8 percent in the comparison group, and had half as many low-trauma fractures9.
So after menopause the rule isn't "lift instead of land." It's "land, and also make your muscles work hard." That half doesn't need a gym. A heavy backpack on a hill, a pack of stairs taken two at a time, loaded carries, bands, bodyweight work done to the point of effort. What it does need is to be hard. Light and easy doesn't count as the strength half any more than walking counts as the impact half.
Density is only half the story, and it isn't the half with the best evidence.
Most fractures start with a fall. The largest review of fall prevention, 108 trials and 23,407 people over 60, three-quarters of them women, found that exercise cut the rate of falls by 23 percent and the number of people who fell by 15 percent, and probably cut fall-related fractures by about a quarter. The exercise that did it was balance and functional training. Walking on its own didn't reduce falls. Neither did strength training on its own10.
Canada's osteoporosis guideline reads that evidence the same way. Its one strong exercise recommendation is balance and functional training at least twice a week. Progressive resistance training is a weaker, conditional suggestion. Impact exercise, walking, yoga and pilates are to be "encouraged" for anyone who enjoys them, as long as they can be done safely11.
This is where the sports win twice. A tennis rally, a soccer drill or a dance class trains reaction, footwork and balance at the same time as it loads the hip. A leg-press machine does one of those things.
Tennis, pickleball, squash, soccer, netball, volleyball, dance, step class, trail running, skipping rope. The best one is the one you'll still be doing in six months.
If your exercise is walking, finish with 20 hops on each leg, or take a flight of stairs two at a time. If you run, add a few strides with a hard push-off. Bone only needs the first few dozen.
Fifty hops a day beat 100 twice a week. Daily and brief beats weekly and long.
Add something that makes your muscles work hard two or three times a week. It can be a gym, or a hill with a heavy pack. It has to be effortful.
Standing on one leg while the kettle boils, tai chi, a balance class, or a sport with footwork in it. This is the part with the strongest evidence for keeping you out of a cast.
If you have osteoporosis or a broken bone behind you, Canada's guideline says to get advice on which movements to modify, especially fast or loaded twisting and bending of the spine11. Impact isn't off the table, but it needs to be chosen with someone who knows your scan.
Bone doesn't count your steps. It counts your landings, and it stops counting after a few dozen. Weights are one way to give it what it wants. A racquet, a ball, a rope or a staircase are others, and they train your balance while they do it. Before menopause, a minute a day is enough to move the number. After menopause, land and lift both, at whatever version of "lift" you'll actually do.
The only wrong answer is the one you quit.
This article is general information, not medical advice. The medical professional who sees you in person may reach a different assessment, and their judgement takes precedence.