Chair-stand test
Five rises from a chair, timed, predict disability and death in older adults about as well as walking speed does. Another proxy, not another workout.
goodWell supported, with the usual caveats about effect size.Revised 01 Oct 2026
No dynamometer, no corridor. A chair you already own.
Definition
Five repetitions of rising from a chair and sitting back down, timed, usually with arms folded so the legs do the work. It is the lower-body third of the Short Physical Performance Battery, alongside standing balance and a short timed walk. No equipment beyond a chair and a stopwatch, and no skill beyond standing up.
The cohort that made it famous
Guralnik and colleagues at the National Institute on Aging recruited one thousand one hundred twenty-two community-dwelling men and women aged seventy-one and older who reported no disability, could walk half a mile, and could climb stairs without help. At baseline each completed the three-test battery. Four years later the researchers asked who had become disabled.
The gradient was steep and orderly. After adjusting for age, sex, and chronic disease, those with the lowest performance scores were between four point two and four point nine times as likely to show disability in daily activities as those with the highest scores, and those in the middle were between one point six and one point eight times as likely. The authors concluded that objective lower-extremity function identifies a preclinical stage of disability: people who report no problems yet measure as if problems are coming.
Note what the test battery does here. The published risk ratios cover the combined score, not chair rises alone. The chair test earned its place by being one of three cheap windows onto the same declining reserve, next to gait speed and balance.
The pooled mortality evidence
Cooper’s team pooled the prospective evidence linking four capability measures (grip strength, walking speed, chair rises, and standing balance) to death from any cause in community-dwelling populations. All four pointed the same way: people who performed worse died sooner. The two estimates the paper prints in full give the scale. Weakest against strongest quarter of grip strength, across fourteen studies and fifty-three thousand four hundred seventy-six people: hazard ratio one point six seven. Slowest against fastest quarter of walking speed, five studies, fourteen thousand six hundred ninety-two people: hazard ratio two point eight seven, after adjusting for age, sex, and body size.
Chair rises belong in that company with one caveat the authors state plainly: studies of walking speed, chair rising, and balance were done only in older populations, average age above seventy, while grip strength predicted mortality in younger groups too. The chair test is an older person’s readout. Exporting it to midlife is extrapolation, not evidence.
Why it works
Like every entry in this shelf, the chair test is not really measuring what it appears to measure. Five fast rises draw on quadriceps power, joint integrity, balance, vision, cardiorespiratory tolerance of effort, and freedom from pain, so a slow time is usually the visible end of several quiet declines at once. That is why it predicts: it summarises the body the way a balance sheet summarises a company. See frailty for what happens when several of those lines turn red together.
What it does not mean
Practising chair rises does not move the underlying risk. This is the thermometer trap described under grip strength, and it applies unchanged. Drilling five fast stands will improve your timed number while leaving the reserve it stood for exactly where it was. The number is useful because it summarises something; it stops being useful the moment it becomes the training goal.
What actually moves it
Whole-body leg strength built over months, balance practised in varied positions, and treatment of whatever is draining reserve: painful joints, breathlessness, anaemia, deconditioning after a hospital stay. Squats, step-ups, carries, and hill walking raise chair-rise speed as a by-product, and raise the thing it was measuring at the same time.
Using it
As a tracked number over years, tested occasionally on the same chair, not as a score to beat or a workout to repeat. A falling time across annual checks is information about the legs and everything attached to them. A rising one after an illness is recovery made visible. Either way the chair is the messenger. Do not shoot it, and do not train it.
What the data do not show
No trial randomised anyone by chair-rise time, so nothing here is a treatment effect: slow risers were not made slow by skipping practice. Guralnik’s ratios describe the combined battery, not the chair test in isolation, and both sources studied community-dwelling older adults, so hospital populations and younger adults are outside the frame. The Cooper review found real heterogeneity between studies, and its chair-rise evidence base was thinner than grip or walking speed. What the data support is a cheap, honest signal. Everything past that (cut-offs, training zones, longevity scores sold from a stopwatch) is somebody else’s invention.