Mobility and balance

Mobility (active range of motion across joints) and balance (the ability to keep your centre of mass over a moving base of support) are the most-neglected pillar of training in midlife — and the one most directly responsible for whether you'll be functionally independent in your eighties. The hard end of the problem is falls: they are the leading cause of injury-related death in adults over 65, and about one in four falls each year — 27.5% of community-dwelling over-65s reported at least one fall in 2018, and 10.2% of all over-65s reported an injury from a fall, roughly a third of those who fell.^[1] About 88% of the emergency visits and hospital admissions for hip fracture are caused by a fall, and once one occurs the trajectory is brutal.^[2] Pooled across 36 countries and 229,851 patients, about 22% of hip-fracture patients are dead within a year — a figure that has been falling, and that varies enormously by country.^[3] Roughly half never regain their pre-fracture walking ability or their independence in everyday tasks.^[4] The signal lights up long before any of that, in single-leg stance time and floor-to-stand ability, both of which are independently and prospectively associated with mortality.

This page makes the case for why mobility and balance matter. For the drills, progressions, and validated programs — how to actually train them — see Training mobility and balance.

The mortality signal from functional tests

A handful of cheap, no-equipment tests carry surprisingly large prognostic weight (Strong as predictors for gait speed, grip strength and the Short Physical Performance Battery, which rest on large multi-country pooled cohorts; Moderate for the one-legged stance, the Sitting-Rising Test and whole-body flexibility, which come from a single exercise-medicine cohort in Rio de Janeiro studied by one research group — the stance test has since been independently replicated, the other two have not. In every case the associations are prognostic, not proof that training the test itself extends life).

• The 10-second one-legged stance. A 2022 cohort study in British Journal of Sports Medicine tracked 1,702 adults aged 51–75 over a median of 7 years. Inability to balance on one leg for 10 seconds at baseline was associated with 84% higher all-cause mortality (hazard ratio 1.84) — a signal that survived adjustment for age, sex, body mass index, and comorbidities.^[5] Roughly one in five participants failed at baseline, with failure rates climbing steeply by age; those who failed also carried worse metabolic profiles, including about threefold higher type 2 diabetes prevalence. To standardise it: barefoot, the free foot resting against the back of the standing leg, arms at the sides, gaze fixed on a point at eye level; up to three attempts in total, and passing on either foot counts as a pass. A 2025 Japanese cohort of 3,278 community-dwelling adults over 65 suggests timing both legs and reading the worse one. Compared with those who managed 30 seconds on their weaker side, people who managed under 10 seconds were more than twice as likely to die or lose their independence over the next seven years (hazard ratio 2.17; the 95% confidence interval — the range within which the true value plausibly sits — ran from 1.67 to 2.84), and the signal held even among people whose better leg was perfectly normal.^[6]
• The Sitting-Rising Test (SRT). Sitting cross-legged on the floor and rising to standing without using hands, knees, or external support, scored 0–10. A 12-year follow-up of 4,282 adults aged 46–75 found scores of 0–4 associated with roughly 4× higher natural-cause mortality and ~6× higher cardiovascular mortality versus a perfect 10; the earlier study that established the test found the lowest scorers had about 5× the all-cause mortality of the highest over 6 years.^[7] The SRT integrates lower-extremity power, core stability, dynamic balance, joint flexibility, and body composition in a single one-minute test. One point is deducted for each support used (a hand, forearm, knee, or hand on thigh) and half a point for visible wobble. Interpret a low score with the obvious confounders in mind — severe knee or hip osteoarthritis, or a large midsection, can drag the score down without reflecting poor systemic fitness. Whole-body flexibility tracks the same way: in 3,139 adults followed for about 13 years, those in the least-flexible group on a 20-movement joint-mobility score had markedly higher death rates from natural causes (roughly double in men; the signal in women was large but too imprecise to pin down) — an association, not proof that stretching extends life.^[8]
• Timed Up and Go and gait speed. The Timed Up and Go (TUG) test — rise from an armchair, walk 3 metres, turn, walk back, sit — is a standard measure of functional mobility. In the Norwegian Tromsø study, 1,005 adults averaging 77 years old were timed on the test and followed for up to twelve years. The slowest fifth had about 80% higher all-cause mortality than the quickest fifth adjusting for age and sex, and about 60% higher once body mass index, self-rated health, smoking and education were added (hazard ratios 1.79 and 1.63).^[9] Plain walking speed carries the same signal — it has been called a "sixth vital sign."^[10] Pooling nine cohorts totalling 34,485 community-dwelling adults aged 65+, each 0.1 m/s of extra gait speed tracked with about 12% lower mortality (hazard ratio 0.88 per 0.1 m/s); speeds of 1.0 m/s and above consistently marked survival longer than age and sex alone would predict, while about 0.8 m/s corresponded to plain median life expectancy.^[11] That same 0.8 m/s is the European working group's cut-off for low physical performance in sarcopenia, which is why it turns up as a clinical threshold.^[12]
• Lower-body power tests. Two quick clinical tests add prognostic value. The five-times sit-to-stand (stand from a chair five times as fast as possible, no hands) is, at a threshold of 12 seconds, among the best-evidenced functional measures of an individual's fall risk — though the same review is blunt that no single test performs strongly on its own.^[13] The Short Physical Performance Battery (SPPB) bundles balance, gait speed, and a chair-stand into a 0–12 score; pooling 17 studies in 16,534 adults averaging 76 years old, lower scores predict progressively higher all-cause mortality.^[14] The European working group uses a score of 8 or below to flag low physical performance.^[15]
• Grip strength. Each 5 kg lower grip strength tracks with about 16% higher all-cause mortality across 139,691 adults in 17 countries in the Prospective Urban Rural Epidemiology (PURE) cohort — a stronger mortality predictor than systolic blood pressure in head-to-head models. One caveat matters on this page in particular: in the same cohort grip strength showed no significant association with injury from a fall, or with fracture. It is a marker of whole-body reserve, not a fall-risk test.^[16]
• Everyday walking counts too. It isn't only laboratory tests. Among 11,539 Australian participants in ASPREE, all aged 70 or over and free of major disease at entry, those who did any weekly walking to get around had about a quarter lower all-cause mortality over a median of 8.6 years than those who never did (hazard ratio 0.73). The benefit arrived at the lowest exposure and then stopped growing — more than once a week gave 0.76 and every day 0.74, indistinguishable from each other. Read it as a nudge rather than a dose: only 314 people never walked for transport, and they were older, frailer and more sedentary in general than everyone else.^[17] Small amounts of regular incidental mobility move the needle.
• And the cheapest test of all is a question. Asking an older adult whether they are afraid of falling carries a mortality signal in its own right: pooled across seven cohorts and 27,714 people, fear of falling tracked with about 29% higher mortality.^[18] The likely mechanism is the self-reinforcing loop this whole page is about — fear leads to restricted activity, restricted activity to deconditioning, deconditioning to more falls. The usual caution applies with more than usual force here: this is observational, pooled from only seven cohorts, and reverse causation is not excluded.

These tests aren't causing mortality reduction by themselves. They're capturing the same underlying physiological reserve — muscle, nerve, balance, body composition — that determines how an aging organism handles physical challenge. Failing them is a signal that the reserve has eroded, not a curse. Each one is also trainable. One caution before you time yourself: these tests sort populations far better than they sort individuals. The Timed Up and Go is the best-studied example — at the usual 13.5-second threshold it identifies only about a third of the people who go on to fall, and pooled across the ten studies with the data for it, a Timed Up and Go score was not a statistically significant predictor of an individual's falls at all.^[19] A pass is not reassurance and a single fail is not a diagnosis; both are prompts to train, and to retest in a year.

Why these tests track mortality

Standing upright is not a single skill. It is the live integration of three sensory streams — vision, the vestibular (inner-ear) balance organ, and proprioception (the joint- and muscle-position sense) — feeding a central controller that drives the muscles. Mortality tracks with these tests partly because most of them load several of these systems at once, and the loop degrades on a fairly predictable schedule.

The mix of inputs the body leans on appears to shift with age. In one cross-sectional comparison of 152 older adults — one group aged 65–74, the other 75 and over — joint-position sense correlated with balance and walking scores in the younger group but not the older one, while muscle strength correlated with both in both groups. The authors read that as a reason to train proprioception harder after 75, not to give up on it: enhancing it, they argue, "should be included in the exercise prescriptions for fall prevention among older adults aged 75 and older."^[20] Treat this as one suggestive study rather than a threshold — the two groups averaged 68 and 81 years, so the contrast is late-sixties against early-eighties, and no falls were recorded in it. That fall rates and hip fractures climb steeply in the over-75s is separately established.^[21]

On the motor side, aging muscle preferentially shrinks its fast-twitch (Type II) fibres — exactly the fibres that produce the rapid burst of force needed to catch a stumble. Pooling 27 studies in more than 750 adults, fast-twitch fibres are smaller in older people while slow-twitch fibres are not, and the proportion of each type is essentially unchanged: the fast fibres are still there, they are just thinner.^[22] The magnitude is substantial — in one comparison of healthy men, fast-twitch fibre cross-section was 29% smaller at 71 than at 23, and six months of resistance training grew it back by about a quarter.^[23] Since a fast fibre produces roughly five to six times the power of a slow one, a small loss of cross-section costs disproportionate power.^[24] A minor perturbation a younger nervous system would correct automatically becomes a fall.

There is also a subtler change in how balance is controlled: a younger person's quiet-standing sway is paradoxically irregular and complex, the signature of a fast, automatic control system, and aging brings a more predictable pattern — the "loss of complexity" idea that has been applied across many physiological systems.^[25] The effect is narrower than it sounds. In 69 young and 61 older adults, the complexity of sway was lower with age in the forward–backward direction only, and was identical side to side; how much people swayed while standing still barely differed — the area covered was the same in both groups, the speed only slightly higher in the older one. The clear age gap appeared when the floor moved unexpectedly.^[26] The authors read the lower complexity as a more attention-driven control strategy rather than simply lost adaptability — which is why the age gap in balance is narrower than the familiar demonstration suggests. Reviewing the postural dual-task literature, older adults hold a stable stance about as well as younger adults while doing a cognitive task; the age difference emerges once the postural task itself gets harder, on a moving or compliant surface.^[27] Balance that must run while you're distracted — so-called dual-task training — is therefore a distinct target, covered in Training mobility and balance.

Power matters more than strength after midlife

The evidence here is Moderate — a consistent mechanism plus observational mortality data. Muscle power declines earlier and faster than muscle strength. The best-known cross-sectional estimate, from 100 healthy adults spread across ages 65 to 89, put the annual difference at 1–2% for strength against about 3.5% for leg-extensor power — though that gap reached significance only in the men.^[28] Following people over time gives a more modest picture: across 25 years of repeated arm measurements in the Baltimore Longitudinal Study of Aging, power declined about 10% more than strength in men, with no detectable difference in women — whose follow-up was much shorter.^[29] Either way the direction is consistent, and it's power, more than maximum strength, that lets someone push up out of a low chair or recover a stumble.

Lower-limb power is the version that matters here: in 9,320 adults aged 60 to 103, those below a low-power threshold on a timed chair-rise test had roughly ten to fourteen times the odds of a mobility limitation, and power fell about twice as fast as it had in the decades before 50.^[30] The mortality data point the same way. Following 3,889 adults aged 46–75 for a median of nearly eleven years, those in the bottom tenth of upper-body power relative to body weight had about six times the mortality of the top tenth (hazard ratio 5.88 in men, 6.90 in women — the women's estimate resting on very few deaths and correspondingly imprecise).^[31]

This is why "I lift weights" doesn't automatically mean "I won't fall." Heavy strength training preserves maximum force production, but if every rep is slow, the power dimension goes untrained — and whether strength work alone carries over into better balance is genuinely unsettled. Across 29 randomised trials of progressive resistance training in 2,174 older adults — too varied in method to pool — only 15 of 68 balance measures improved more than in controls, and the authors concluded that resistance training alone may not be a robust intervention for balance control.^[32] A later synthesis of 13 trials reached the opposite conclusion, with meaningful gains in single-leg stance time and Timed Up and Go.^[33] The safe reading is that lifting may help balance and may not — which is reason enough to train balance directly rather than hope for transfer. Explosive power work and balance training are separate line items; Training mobility and balance covers how to program both.

Falls: the stakes and what the evidence supports

The stakes are set out at the top of this page: falls are the leading cause of injury death after 65, and a hip fracture is a healthspan-ending event for many. The good news is that the intervention evidence is Strong for the outcome it was built to measure — falls themselves — and it is one of the largest bodies of randomized-trial evidence in physiotherapy. Be clear about the boundary, though: pooling the exercise trials, the US Preventive Services Task Force found no significant reduction in fall-related fractures or in all-cause mortality.^[34] The closest thing to a hard-outcome win is the LIFE trial, which randomised 1,635 sedentary adults aged 70–89 to a structured walking-and-strength programme and cut the risk of losing the ability to walk 400 metres from 35.5% to 30.1% over about two and a half years — a real result, and about mobility rather than survival.^[35]

A 2019 Cochrane review of 108 trials in more than 23,000 community-dwelling older adults found that exercise cuts the rate of falls by about 23% and the number of people who fall by about 15%, with the largest single estimate from multi-component programmes that combine balance training with progressive resistance.^[36] In 2024 the US Preventive Services Task Force reached the same conclusion and recommended exercise to prevent falls — but specifically for community-dwelling adults 65 and over who are at increased risk of falling, which in practice means a history of falls or a problem with mobility or physical function. That grade-B recommendation rests on 37 exercise trials in 16,117 people; the wider review it sits in covered 83 trials and 48,839 people across six different kinds of intervention.^[37]

Which kind of exercise matters most. The reliable answer is narrower than a single winner: what works is exercise that challenges balance, usually alongside progressive strength work. Balance-and-functional training carries the highest-certainty evidence of any category — about 24% fewer falls, from 39 trials — while programmes combining several types have a larger but less certain estimate, about 34% from 11 trials; the two are not statistically distinguishable.^[38] The largest review of the field, covering 219 trials and 167,864 participants, does not separate them at all — it treats balance-and-resistance as a single category, and finds the most certain benefits in supervised, longer-running programmes of that kind and in group Tai Chi.^[39] What the evidence does not support is a programme built on resistance training, walking or dance alone: reviewers repeatedly report themselves uncertain of the effects of those.

And staying mobile looks like prognosis as well as prevention. Among 2,454 US adults over 65 who had already had a fall injury, the 81% who fell short of the recommended weekly activity level had about 50% higher all-cause mortality than the 19% who met it (hazard ratio 1.50). Read it as a marker rather than a lever: activity and fall injury were reported in the same survey, so frailty could be driving both.^[40]

The response should be matched to risk rather than applied one-size-fits-all. The 2022 World Falls Guidelines — a consensus of 96 experts across 39 countries — sort older adults into low, intermediate and high fall risk. Low risk means no falls in the past year, or one fall that caused no injury in someone with no gait or balance problem: education plus general exercise, and reassess yearly. Intermediate risk is a minor fall in someone who does have a gait or balance impairment: tailored balance-and-strength exercise. High risk means a fall plus at least one of — injury, two or more falls in the past year, frailty, being unable to get up off the floor unaided for an hour or more, or the fall being accompanied by a suspected blackout. That last one changes the destination: it points at the heart or the nervous system, and the guidelines say it should trigger a syncope work-up rather than an exercise plan. Everything else in the high-risk group goes for a full multifactorial assessment and a personalised plan.^[41] One caveat on that last tier: the Task Force gave the multifactorial route a grade C — individualise the decision, because the average net benefit is small — while exercise got the stronger grade B.^[42]

For the specifics — the balance progressions, perturbation and step training, power work for fall-arrest, and the best-evidenced programmes (Tai Chi, the Otago programme, and others) — see Training mobility and balance.

What isn't exercise

Two things belong in any honest account of falls, and neither is training.

Medications are the most modifiable non-exercise risk factor (Moderate). The multifactorial assessment the guidelines send high-risk people for is defined as covering postural blood pressure, medication, environment, cognition and psychological health — and medications are where the exposure is concentrated. Pooling 46 studies of people who had actually fallen, 68.6% were on at least one potentially inappropriate prescription at the time, averaging 2.21 each; sedatives and hypnotics, opioids, diuretics and antidepressants dominate the list.^[43] The honest caveat has to travel with it: reviewing medications on its own has not been shown to reduce falls — pooled across three community trials the number of fallers was unchanged (relative risk 1.05, meaning no difference) — so it belongs inside a broader plan rather than instead of one.^[44] See Insomnia treatment for the hip-fracture data on the sleep drugs in particular.

Vitamin D does not prevent falls in people who are not deficient (Strong). The US Preventive Services Task Force recommended against supplementing to prevent falls in 2018 and has not reversed that — the 2024 update simply moved the question out of scope pending a separate review.^[45] The evidence has since firmed up: across 69 randomised trials in 153,902 adults, vitamin D had little to no effect on fractures or falls, with high certainty for any fracture.^[46] Large intermittent doses appear to make things worse: in 200 adults over 70 with a prior fall, monthly high-dose vitamin D produced more falls over a year than a standard dose, 67% versus 48%.^[47] Correcting a genuine deficiency is a different question — see Vitamin D.

Further reading

• Araújo CGS et al. Successful 10-second one-legged stance performance predicts survival in middle-aged and older individuals. Br J Sports Med 2022.^[48]
• Hirayama Y et al. Association of shorter secondary one-leg standing time with loss of independence or death in community-dwelling older adults. J Am Med Dir Assoc 2025.^[49]
• Araújo CGS et al. Sitting-rising test scores predict natural and cardiovascular causes of death in middle-aged and older adults. Eur J Prev Cardiol 2025.^[50]
• Brito LBB et al. Ability to sit and rise from the floor as a predictor of all-cause mortality (Sitting-Rising Test). Eur J Prev Cardiol 2014.^[51]
• Araújo CGS et al. Flexibility and all-cause mortality (Flexindex). Scand J Med Sci Sports 2024.^[52]
• Bergland A et al. Mobility as a predictor of all-cause mortality in older men and women: 11.8-year follow-up in the Tromsø study. BMC Health Serv Res 2017.^[53]
• Studenski S et al. Gait speed and survival in older adults. JAMA 2011.^[54]
• Fritz S, Lusardi M. White paper: walking speed as the sixth vital sign. J Geriatr Phys Ther 2009.^[55]
• Cruz-Jentoft AJ et al. Sarcopenia: revised European consensus on definition and diagnosis (EWGSOP2). Age Ageing 2019.^[56]
• Pavasini R et al. Short Physical Performance Battery and all-cause mortality: systematic review and meta-analysis. BMC Med 2016.^[57]
• Lusardi MM et al. Determining risk of falls in community-dwelling older adults: a systematic review and meta-analysis using posttest probability. J Geriatr Phys Ther 2017.^[58]
• Barry E et al. Is the Timed Up and Go test a useful predictor of risk of falls in community-dwelling older adults: a systematic review and meta-analysis. BMC Geriatr 2014.^[59]
• Leong DP et al. Prognostic value of grip strength — findings from the PURE cohort. Lancet 2015.^[60]
• Shah S et al. Walking for transport and all-cause mortality: a prospective cohort study of Australian community-dwelling older adults (ASPREE). BMJ Public Health 2024.^[61]
• Ma W et al. Association between fear of falling and mortality in middle-aged and older adults: a systematic review and meta-analysis. Geriatr Nurs 2024.^[62]
• Wang Q et al. The relationships of postural stability with muscle strength and proprioception are different among older adults over and under 75 years of age. J Exerc Sci Fit 2022.^[63]
• Rubenstein LZ. Falls in older people: epidemiology, risk factors and strategies for prevention. Age Ageing 2006.^[64]
• Lee C et al. Effects of age on human skeletal muscle: a systematic review and meta-analysis of myosin heavy chain isoform protein expression, fibre size and distribution. Am J Physiol Cell Physiol 2024.^[65]
• Nilwik R et al. The decline in skeletal muscle mass with aging is mainly attributed to a reduction in type II muscle fibre size. Exp Gerontol 2013.^[66]
• Grosicki GJ et al. Single muscle fibre contractile function with ageing. J Physiol 2022.^[67]
• Lipsitz LA, Goldberger AL. Loss of complexity and aging: potential applications of fractals and chaos theory to senescence. JAMA 1992.^[68]
• Rizzato A et al. Age-related changes in static and dynamic postural balance performance. Front Aging Neurosci 2026.^[69]
• Boisgontier MP et al. Age-related differences in attentional cost associated with postural dual tasks. Neurosci Biobehav Rev 2013.^[70]
• Skelton DA et al. Strength, power and related functional ability of healthy people aged 65–89 years. Age Ageing 1994.^[71]
• Metter EJ et al. Age-associated loss of power and strength in the upper extremities in women and men. J Gerontol A Biol Sci Med Sci 1997.^[72]
• Alcazar J et al. Relative sit-to-stand power: aging trajectories, functionally relevant cut-off points, and normative data in a large European cohort. J Cachexia Sarcopenia Muscle 2021.^[73]
• Araújo CGS et al. Muscle power versus strength as a predictor of mortality in middle-aged and older men and women. Mayo Clin Proc 2025.^[74]
• Orr R et al. Efficacy of progressive resistance training on balance performance in older adults: a systematic review of randomized controlled trials. Sports Med 2008.^[75]
• Šarabon N, Kozinc Ž. Effects of resistance exercise on balance ability: systematic review and meta-analysis of randomized controlled trials. Life (Basel) 2020.^[76]
• Sherrington C et al. Exercise for preventing falls in older people living in the community (Cochrane Review). Cochrane Database Syst Rev 2019.^[77]
• Pillay J et al. Falls prevention interventions for community-dwelling older adults: systematic review and meta-analysis of benefits, harms, and patient values and preferences. Syst Rev 2024.^[78]
• Nicholson WK et al. Interventions to prevent falls in community-dwelling older adults: US Preventive Services Task Force recommendation statement. JAMA 2024.^[79]
• Guirguis-Blake JM et al. Interventions to prevent falls in older adults: updated evidence report and systematic review for the USPSTF. JAMA 2024.^[80]
• Pahor M et al. Effect of structured physical activity on prevention of major mobility disability in older adults: the LIFE study randomized clinical trial. JAMA 2014.^[81]
• Adeyemi O et al. Activity intensity and all-cause mortality following fall injury among older adults. Healthcare (Basel) 2025.^[82]
• Montero-Odasso M et al. World guidelines for falls prevention and management for older adults. Age Ageing 2022.^[83]
• Moreland BL et al. Hip fracture-related emergency department visits, hospitalizations and deaths by mechanism of injury among adults aged 65 and older. J Aging Health 2023.^[84]
• Downey C et al. Changing trends in the mortality rate at 1-year post hip fracture — a systematic review. World J Orthop 2019.^[85]
• Dyer SM et al. A critical review of the long-term disability outcomes following hip fracture. BMC Geriatr 2016.^[86]
• O'Reilly T et al. Potentially inappropriate prescribing and falls-risk-increasing drugs in people who have experienced a fall: a systematic review and meta-analysis. Age Ageing 2025.^[87]
• Seppala LJ et al. Medication reviews and deprescribing as a single intervention in falls prevention: a systematic review and meta-analysis. Age Ageing 2022.^[88]
• US Preventive Services Task Force. Interventions to prevent falls in community-dwelling older adults: recommendation statement. JAMA 2018.^[89]
• Massé O et al. Calcium, vitamin D, or combined supplementation to prevent fractures and falls: systematic review and meta-analysis. BMJ 2026.^[90]
• Bischoff-Ferrari HA et al. Monthly high-dose vitamin D treatment for the prevention of functional decline: a randomized clinical trial. JAMA Intern Med 2016.^[91]