The TUG, short for Timed Up and Go, times a single sequence: stand up from an armchair, walk three metres, turn around, walk back, sit down again. It yields one number, in seconds. The calculator is right below, with the PDF export.
The stopwatch starts on the command and stops once the patient is seated again. A change of 4.09 seconds is needed before calling it a real change in older adults: the thresholds are detailed below.
Figures taken from the validation studies listed at the end of this page, every value carrying its source where it is written. The test procedure is published in open access by the CDC.
What the TUG measures
The TUG measures neither balance, nor strength, nor gait. It measures a sequence, and that is what makes it valuable: standing up, walking, turning, walking back and sitting down are the five actions that make up most daily movement, and the turn is where falls happen most.
The original publication, in 60 geriatric day-hospital patients with a mean age of 79.5, immediately places the test against the others: the correlation is -0.81 with the Berg balance scale, -0.78 with the Barthel index of activities of daily living, and -0.61 with gait speed. The TUG therefore captures much of what those three tools capture, in a minute and with no equipment.
That is also what bounds its use: a single number summarising five actions does not say which of the five is the problem.
Running the test
The procedure is short and it tolerates no approximation, because every deviation is paid in seconds.
- An armchair with armrests, usual seat height, back against the backrest at the start.
- A floor marker at three metres, measured from the foot of the chair.
- The patient keeps their usual footwear and walking aid: the TUG measures their real mobility, not their ideal mobility.
- One practice run, then one timed trial.
- The stopwatch starts on the command and stops when the back touches the backrest again.
The procedure is published in open access by the Centers for Disease Control and Prevention in their STEADI toolkit, which makes it reproducible without permission. The text of the 1991 article, however, remains the property of its journal.
The twelve-second threshold
The number in circulation is twelve seconds, and it is older than people think: the Centers for Disease Control and Prevention assessment sheet has carried it since 2013 at least. The 2017 systematic review often cited as its source therefore is not, since it appeared three and a half years later: it corroborates the threshold, placing a TUG of 12 seconds or more among the best-evidenced functional measures out of 26 performance tests examined.
That same review states the limit the threshold makes people forget: no single test reaches a strong post-test probability. Twelve seconds point towards a full falls assessment, they diagnose nothing, and they do not transfer as such to an institutionalised or neurological population.
The other number, the forgotten one, is the minimal detectable change: the gap below which two measurements cannot be told apart from noise.
| What is measured | Value | Population and sample |
|---|---|---|
| Falls-risk screening threshold | 12 s | 65 and over at home, review of 59 studies |
| Reliability, within and between raters | ICC > 0.90 | 78 residents dependent in ADL |
| Test-retest reliability | ICC 0.973 | 51 people with Alzheimer disease |
| Minimal detectable change, 90% | 4.09 s | 51 people with Alzheimer disease |
| Improvement judged important | 0.8 to 1.4 s | 65 hip osteoarthritis, by method |
| Hip fracture, hazard ratio | 1.64 | 6 cohorts, 1,639,397 participants |
What the number does not tell you
The reliability of the TUG is excellent, and that is not where the problem lies. In 78 residents dependent in activities of daily living, mean age 84.8 and MMSE 18.7, the ICC exceeds 0.90 both within and between raters. In 51 people with Alzheimer disease, it reaches 0.973. The measurement is stable, even when cognition is not.
The weak link is interpretation. The difference judged clinically important by patients themselves is 0.8, 1.4 or 1.2 seconds depending on the method of calculation, in one and the same cohort of 65 hip osteoarthritis patients. Quoting a single threshold for important change therefore means picking a method without saying so.
What the number also fails to say: fear of falling, the home environment, medication, orthostatic hypotension. A TUG of 9 seconds in someone who is afraid to go out is not reassuring.
A screening threshold is not a diagnosis
Twelve seconds point towards a falls assessment, they do not replace one. The TUG assesses neither fear of falling, nor the home environment, nor current medication, and a time below the threshold puts nobody in the clear.
Three administration pitfalls
Skipping the practice run
The first trial is almost always slower than the second. Timing the first pass folds learning the protocol into the measurement, and inflates the baseline time, and therefore the apparent improvement at follow-up.
Taking the walking aid away
Running the test without the cane because it will look cleaner measures a mobility the patient does not have in daily life. The walking aid is part of the measurement and it goes in the notes: changing aid between two administrations makes the comparison void.
Stopping the stopwatch on contact with the seat
Time runs until the back touches the backrest again, not until the first contact with the seat. That is half a second to a second, which is the order of magnitude of the differences we are trying to interpret.
Frequently asked questions
What does TUG stand for?
TUG stands for Timed Up and Go. It was published by Podsiadlo and Richardson in 1991, derived from the Get-Up and Go of Mathias, Nayak and Isaacs from 1986, which was not timed but scored on five points.
What is the TUG cut-off?
The most widely used operational threshold is 12 seconds in people aged 65 and over living at home. The CDC assessment sheet has carried it since 2013 at least, and a 2017 systematic review corroborates it, placing it among the best-evidenced measures while concluding that no single test reaches a strong post-test probability. It is a screening threshold, not a diagnosis.
How much change counts?
The minimal detectable change is 4.09 seconds at 90% confidence in older people with Alzheimer disease. Below that, the gap between two measurements cannot be told apart from noise. The difference judged important by patients varies from 0.8 to 1.4 seconds depending on the method.
Is the TUG free to use?
Yes. No royalty or prior registration was found, and the full procedure is published in open access by the Centers for Disease Control and Prevention in their STEADI toolkit.
Does the TUG predict anything besides falls?
A meta-analysis of 6 prospective cohorts totalling 1,639,397 participants associates degraded TUG performance with increased hip fracture risk, with a hazard ratio of 1.64 (95% CI 1.20 to 2.22).
References
7 sources, PMIDs included
- Podsiadlo D, Richardson S. The timed "Up & Go": a test of basic functional mobility for frail elderly persons. J Am Geriatr Soc 1991;39(2):142-8. PMID 1991946. The original publication, in 60 geriatric day-hospital patients with a mean age of 79.5. Correlations: Berg balance scale r = -0.81, gait speed r = -0.61, Barthel index r = -0.78.
- Mathias S, Nayak US, Isaacs B. Balance in elderly patients: the "get-up and go" test. Arch Phys Med Rehabil 1986;67(6):387-9. PMID 3487300. The direct antecedent: the untimed, 5-point version from which Podsiadlo and Richardson derived the timed one.
- Nordin E, Rosendahl E, Lundin-Olsson L. Timed "Up & Go" test: reliability in older people dependent in activities of daily living. Phys Ther 2006;86(5):646-55. PMID 16649889. 78 residents dependent in activities of daily living, mean age 84.8, MMSE 18.7. Reliability ICC above 0.90, both within and between raters.
- Ries JD, Echternach JL, Nof L, Gagnon Blodgett M. Test-retest reliability and minimal detectable change scores for the timed "up & go" test, the six-minute walk test and gait speed in people with Alzheimer disease. Phys Ther 2009;89(6):569-79. PMID 19389792. 51 people with Alzheimer disease, ICC of 0.973 or above. This is the source of the 4.09-second minimal detectable change at 90% confidence.
- Wright AA, Cook CE, Baxter GD, Dockerty JD, Abbott JH. A comparison of 3 methodological approaches to defining major clinically important improvement of 4 performance measures in patients with hip osteoarthritis. J Orthop Sports Phys Ther 2011;41(5):319-27. PMID 21335930. 65 hip osteoarthritis patients reassessed at 9 weeks: important improvement is 0.8, 1.4 or 1.2 seconds depending on the method. This is what forbids quoting a single threshold.
- Lusardi MM, Fritz S, Middleton A, Allison L, Wingood M, Phillips E, 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;40(1):1-36. PMID 27537070. 2,294 abstracts screened, 59 articles retained. It CORROBORATES the 12-second threshold rather than being its source: the CDC sheet already carried it in 2013, three and a half years earlier. It also concludes that no single test reaches a strong post-test probability.
- Hernandez-Martínez J, Cid-Calfucura I, Vásquez-Carrasco E, Perez-Carcamo J, Herrera-Valenzuela T, Aravena-Sagardia P, et al. Timed Up-and-Go as a predictor of fracture risk in older adults: a systematic review and meta-analysis. PMID 42200137. 6 prospective cohorts, 1,639,397 participants: degraded TUG performance is associated with increased hip fracture risk, hazard ratio 1.64 (95% CI 1.20 to 2.22).
Page written by Anthony Baillon, physiotherapist, co-founder of Physio Learning. The test itself is by Podsiadlo and Richardson, 1991, after the Get-Up and Go of Mathias, Nayak and Isaacs, 1986: this page documents it, it does not claim authorship of it.