“Validation of the Timed Up and Go Test for Assessing Balance Variables in Adults Aged 65 and Older”

by Nightingale CJ, Mitchell SN, Butterfield SA

Journal of Aging and Physical Activity
© 2018 Human Kinetics, Inc.

Note: This article will be published in a forthcoming issue of

the Journal of Aging and Physical Activity. This article appears
here in its accepted, peer-reviewed form; it has not been copy
edited, proofed, or formatted by the publisher.

Section: Original Research

Article Title: Validation of the Timed Up and Go Test for Assessing Balance Variables in
Adults Aged 65 and Older

Authors: Christopher J. Nightingale1, Sidney N. Mitchell2, and Stephen A. Butterfield1

Affiliations: 1School of Kinesiology, Physical Education, and Athletic Training; 2School of

Educational Leadership, Higher Education and Human Development; University of Maine,
Orono, ME.

Running Head: Tug test validation

Journal: Journal of Aging and Physical Activity

Acceptance Date: July 5, 2018

©2018 Human Kinetics, Inc.

DOI: https://doi.org/10.1123/japa.2018-0049
 “Validation of the Timed Up and Go Test for Assessing Balance Variables in Adults Aged 65 and Older”
by Nightingale CJ, Mitchell SN, Butterfield SA
Journal of Aging and Physical Activity
© 2018 Human Kinetics, Inc.

Abstract

The Timed Up and Go (TUG) Test is recommended by the U.S. Centers for Disease Control and

Prevention as an easy to administer clinical test to evaluate a senior citizen’s fall risk. Limited

evidence has been presented in the literature validating the TUG Test. In this study we sought to
Downloaded by USD UNIVERSITY on 08/18/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}

assess correlations between the TUG Test and various balance markers utilizing the OptoGait

system. Fifty-one healthy seniors completed randomized trials of the TUG Test and a Gait Test

utilizing OptoGait photoelectric technology. Correlations among mean and standard deviation

values for these variables and TUG performance were calculated. Utilizing a Bonferroni

adjustment and an alpha level of .05, eight significant correlations of a moderately strong degree

(absolute r scores between .51 and .78) emerged. Correlation results indicate that the TUG Test is

a valid tool for screening balance deficits that lead to increased fall risk in senior citizens.

Keywords: Balance, Fall Risk, Evaluation

 “Validation of the Timed Up and Go Test for Assessing Balance Variables in Adults Aged 65 and Older”
by Nightingale CJ, Mitchell SN, Butterfield SA
Journal of Aging and Physical Activity
© 2018 Human Kinetics, Inc.

Introduction

With the high prevalence of debilitating injuries associated with accidental falls in senior

citizens, the need for clinicians to have valid and reliable methods of risk assessment that can be

administered efficiently at low financial cost is important. The Timed Up and Go (TUG) Test has
Downloaded by USD UNIVERSITY on 08/18/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}

been endorsed by the U.S. Centers for Disease Control and Prevention (CDC) as clinically useful

tool for evaluating gait, strength, and balance in determining fall risk in elderly patients. However,

to date limited research exists to determine the validity of the TUG Test as a measurement of

balance in senior citizens. The purpose of this study was to examine correlations between the TUG

Test and various indicators of balance utilizing the OptoGait Photoelectric System.

The Timed Up and Go (TUG) Test is used to assess mobility and balance. Specifically,

the patient is asked to rise from a seated position, walk three meters, turn around, and return and

sit in the starting point chair while timed. Patients that fail to complete the test in fewer than 12

seconds are considered to have elevated fall risk. While there is limited research demonstrating

the veracity of this test, several studies have been conducted that show promise. Shumway-Cook,

Brauer, and Woollacott (2000) found the TUG Test to be both sensitive (87%) and specific (87%)

in identifying community-dwelling older adults who are prone to falls. Podsialdlo and Richardson

(1991) evaluated the TUG Test by measuring 60 patients at a Geriatric Day Hospital and found

TUG Test performance results to correlate well with the Berg Balance Scale (r = -.81), gait speed

(r=-.61) and the Barthel Index of ADL (r=-.78). Likewise, Steffen, Hacker, and Mollinger (2002)

evaluated 96 community-dwelling older adults and found high test-retest reliability between TUG

Test performance with the Six-Minute Walk Test and the Berg Balance Scale (ICC [2,1] =.95-.97).

Additionally, Alexandre, Meira, Rico, and Mizuta (2011) screened 63 community-dwelling elderly

individuals and found that those subjects with a history of falls (n=21) performed the TUG Test
 “Validation of the Timed Up and Go Test for Assessing Balance Variables in Adults Aged 65 and Older”
by Nightingale CJ, Mitchell SN, Butterfield SA
Journal of Aging and Physical Activity
© 2018 Human Kinetics, Inc.

significantly slower (p=.02) than seniors that had not fallen (n=42). Beyond these studies,

however, little research has been published that investigates the relationship between TUG Test

performance and other measures of balance.

The purpose of this study was to assess correlations between the TUG Test and various
Downloaded by USD UNIVERSITY on 08/18/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}

balance markers utilizing the OptoGait system. OptoGait is a highly sensitive photocell system

that can be used to conduct many different assessment tests to evaluate movement patterns, body

mechanics, and balance. The OptoGait system can measure changes in foot position, flight time

and movement variations to within 0.001 second. Data is then analyzed by specialized software

for the purpose of motion analysis and presented as means, standard deviations, and variability

coefficients for different spatiotemporal parameters of gait. OptoGait has been found to be both a

reliable and valid tool for assessing gait and balance in senior citizens (Stöckel et al., 2015;

Matsumoto et al., 2015; Lienhard, Schneider, & Maffuletti, 2012). It was hypothesized that TUG

Performance will correlate highly with multiple parameters of balance as assessed by the OptoGait

Gait Test.

Method

Fifty-one participants were recruited via flyers placed at local community centers, health

care provider offices, and senior centers. Potential participants completed a brief health history

questionnaire (See Table One) and were disqualified if they suffered from neurocognitive disease

that impaired their balance, suffered from unresolved lower extremity injury within the last 6

months that impaired their gait, or were unable to walk unassisted. All participants were 65 years

old or older at the time of testing. All testing was conducted under the auspices of the Primary

Investigator’s Institutional Review Board, and all participants completed written informed consent

documentation.
 “Validation of the Timed Up and Go Test for Assessing Balance Variables in Adults Aged 65 and Older”
by Nightingale CJ, Mitchell SN, Butterfield SA
Journal of Aging and Physical Activity
© 2018 Human Kinetics, Inc.

Participants were instructed to wear regular footwear and comfortable clothing on the day

of testing. Participants completed the Timed Up and Go (TUG) Test and OptoGait Gait Test

protocols in random order. Two practice trials of each test were allowed for participants to

familiarize themselves with the protocols. The TUG Test protocol consisted of participants sitting
Downloaded by USD UNIVERSITY on 08/18/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}

in a standard arm chair at the starting point. After a “3-2-1” countdown, a timer was begun as

participants initiated the process of standing from the chair. Participants were asked to rise from

the seated position, walk at a normal, self-selected pace to a line marked on a hardwood floor three

meters from the start line, turn around, walk back to the chair and sit down. Timing stopped when

the participants sat back down into the chair.

For the OptoGait Gait Test protocol, participants were instructed to begin in a standing

position and walk at a normal, self-selected pace between two parallel OptoGait bars. Participants

began at a line marked on the floor two meters before entering the system. This gave participants

the opportunity to be at their regular speed during the recorded aspect of the test. Participants

walked through 10 meters of OptoGait pathway, recording data for approximately 12 strides. (See

Figure One).

Participants continued to walk until they passed a line two meters beyond the OptoGait

system to ensure they did not slow down or otherwise alter their gait during the final steps of data

collection. Data collected during the OptoGait gait test included mean and standard deviation

scores for seven balance variables, including step length, stride length, contact time, contact phase

time, foot flat time, propulsive phase time, and average speed.

Timed Up and Go times and OptoGait data were recorded and encrypted in a password

protected spread sheet software package and health history sheets and other potentially identifying

information were destroyed upon completion of data collection. Pearson’s r correlations were
 “Validation of the Timed Up and Go Test for Assessing Balance Variables in Adults Aged 65 and Older”
by Nightingale CJ, Mitchell SN, Butterfield SA
Journal of Aging and Physical Activity
© 2018 Human Kinetics, Inc.

calculated using SPSS analysis software. A total of fourteen correlations were calculated,

OptoGait mean and standard deviation variable scores of each of the seven balance variables were

correlated with Timed Up and Go scores. Due to the nature of the repeated calculations, a

Bonferroni adjustment was made, utilizing a P < .004 value as a determinant of significance.
Downloaded by USD UNIVERSITY on 08/18/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}

Results

All participants were able to complete both test protocols. Of the fifty-one participants,

only seven (13.7%) were unable to complete the TUG test in less than 12 seconds. The 44

participants that were deemed ‘not at-risk’ for falls performed the TUG in a mean time of 8.97

seconds. The mean time for the seven ‘at-risk’ participants was 18.43 seconds.

Preliminary tests of significance were measured for the two groups (‘at-risk’ and ‘not at-

risk’). Utilizing an alpha level of .05, it was determined that the differences in TUG Test

performance were significant between groups (p<.00001). Calculation of Cohen’s d for effect size

generates a value of 1.92, indicating a very large effect size.

Means and Standard Deviation Scores for each of the OptoGait gait test variables are listed

in Table Two. Each of these scores was then correlated with the TUG performance time to

generate a total of fourteen correlations. Correlations of mean gait scores varied from very weak

(mean contact time and TUG = -.067) to strong (mean step length and TUG = -.61, mean average

speed and TUG = -.68). Correlations of standard deviation of gait scores also varied from very

weak (SD stride length and TUG = -.01) to strong (SD foot flat time and TUG = .76, SD propulsive

phase time and TUG = .78). It is worthy of noting that those values with negative correlations

indicate that an increasing speed results in a smaller time variable.

Table Three presents the significance of each measured correlation. Three of seven mean

value correlations were found to be significant at the Bonferroni adjusted level of P<.004. Four
 “Validation of the Timed Up and Go Test for Assessing Balance Variables in Adults Aged 65 and Older”
by Nightingale CJ, Mitchell SN, Butterfield SA
Journal of Aging and Physical Activity
© 2018 Human Kinetics, Inc.

of the seven standard deviation value correlations were significant at P<.004, with a fifth value

nearly significant at that level (SD average speed and TUG correlations P=.04). The five strongest

correlations (mean step length (r=-.51) mean stride length and TUG (r=-.61), mean average speed

and TUG (r=-.68), SD foot flat time and TUG (r=.76), and SD propulsive phase time and TUG
Downloaded by USD UNIVERSITY on 08/18/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}

(r=.78)) were all highly significant with P values less than .0001.

Discussion

As people reach their senior years, balance issues and the risk of fall-related injuries greatly

increase. The financial cost of medical care associated with senior citizen falls is considerable,

and fear of falls leads to changes of behaviors, including voluntarily withdrawing from society due

to fear of leaving the home, that can greatly reduce quality of life for aging seniors. Loss of

vestibular function and slowing reaction time lead to changes in gait parameters and utilization of

different, potentially less effective, strategies in maintaining balance (Stockel et al, 2015).

Improved tests for assessing those at risk of developing poor balance and increased

likelihood of falls are vital to prevent negative outcomes associated with common aging-related

balance deficits. One clinical test recommended by the U.S. CDC is the Timed Up and Go (TUG)

Test. The TUG Test is quite popular as a clinical assessment tool for fall risk, primarily because

it requires minimal equipment, can be administered in a short amount of time, and is easily

interpreted. Despite its popularity, the test is somewhat lacking in evidence that supports its

validity and reliability as a balance prediction and fall risk assessment tool. Our goal was to

evaluate the TUG Test as an indicator of balance.

To assess the TUG test, we selected the OptoGait Gait Test as a tool to which we could

compare TUG performance. The OptoGait Photoelectric System has been validated as an effective

tool for precise measurement of spatiotemporal gait parameters in senior citizens (Granacher et al,
 “Validation of the Timed Up and Go Test for Assessing Balance Variables in Adults Aged 65 and Older”
by Nightingale CJ, Mitchell SN, Butterfield SA
Journal of Aging and Physical Activity
© 2018 Human Kinetics, Inc.

2014; Lee, 2017; Lienhard, Schneider, & Maffuletti, 2012; Stockel et al, 2015). That said, there

is a general absence of normative data for OptoGait parameters with the population we

investigated. As research with this relatively new technology continues to be conducted, we expect

that holes in the body of research regarding OptoGait parameters will be reduced. We opted to
Downloaded by USD UNIVERSITY on 08/18/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}

examine correlations between TUG Test performance and mean and standard deviation scores of

seven different gait variables (for a total of fourteen correlations). By identifying consistency in

the relationship between TUG Score and the validated OptoGait gait tests, we hope to establish

the validity of the TUG test. Eight of the fourteen correlations (when assessed with a Bonferroni

adjustment) were significant. Of these significant correlations, six showed a moderate to strong

relationship (R>.50). The strength of these relationships indicates that aspects of gait such as mean

step length, mean stride length, and mean average speed serve as good predictors for balance.

Based on these results, we state with confidence that the Timed Up and Go Test as a valid predictor

of balance in senior citizens.

The current state of research into the diagnostic accuracy of fall risk assessment shows a

lack of validated instruments that accurately predict fall risk with high sensitivity and high

specificity. Strupeit, Buss, and Wolf-Ostermann (2016) investigated three commonly used

methods of assessing fall risk and found that all three (the St. Thomas’s risk assessment tool,

clinical assessment, and self-reported fear of falling) were prone to elevated rates of false-negative

and false-positive results. Of the three tests evaluated, only the self-reported assessment (self-

reported fear of falling) was found to be sensitive at a significant level (Strupeit, Buss, & Wolf-

Ostermann, 2016). Longitudinal research is needed to support the Timed Up and Go test as having

appropriate sensitivity and specificity as a clinical diagnostic test, but initial findings show

promise. By showing a high degree of correlation with various balance markers, it is not
 “Validation of the Timed Up and Go Test for Assessing Balance Variables in Adults Aged 65 and Older”
by Nightingale CJ, Mitchell SN, Butterfield SA
Journal of Aging and Physical Activity
© 2018 Human Kinetics, Inc.

unreasonable to expect that the TUG Test might effectively rule out false-negative and false-

positive results in predicting which older adults are truly at risk of falls and in need of balance

improving interventions.

The justification for the use of the TUG Test is legitimate. The evidence that supports this
Downloaded by USD UNIVERSITY on 08/18/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}

test as an accurate predictive tool for balance in seniors gives credence to its use. Because this test

requires no more equipment that an armchair, a watch, and three meters of open floor space, it is

both simple and fast to administer and interpret. Healthcare providers can utilize this screening

test to determine who is at risk of falls associated with poor balance for virtually no cost and

minimal time investment. These providers can then utilize more costly and/or precise diagnostic

tools like OptoGait or Force Plate technology to develop interventions and monitor progress of

patients to reduce fall risk and increase the quality of life.

There are some limitations with this study. Subject recruitment may have created a non-

representative sample. Many of the participants in this study were recruited from assisted living

and community centers. These volunteers were often members of senior exercise classes or

independently maintained higher levels of physical activity than many of their peers. These higher

levels of activity might impart better degrees of balance that are not representative of many senior

citizens. Of our participants, 86% (n=44) performed the TUG Test in less than 12 seconds,

indicating a decreased risk of falls. This could potentially affect the findings of our study.

Sample size was limited because only 51 participants met the inclusion criteria. As these

criteria included no recent history of orthopedic or neurologic condition that could impair balance

or gait, our sample might not be representative of all senior citizens. Salzman (2010) found that

between 30 and 60 percent of persons older than 65 report waling impairments over moderate

distances and 35 percent of tested seniors had abnormal gait due to one or more of our exclusion
 “Validation of the Timed Up and Go Test for Assessing Balance Variables in Adults Aged 65 and Older”
by Nightingale CJ, Mitchell SN, Butterfield SA
Journal of Aging and Physical Activity
© 2018 Human Kinetics, Inc.

criteria. By limiting the sample we evaluated, we strengthened the reliability of our correlations,

but might have limited the validity of our overall findings. Finally, our participants came from

a localized geographic area, which might also influence generalizability of our findings. Future

testing with a larger and more diverse sample is recommended, despite the promising initial
Downloaded by USD UNIVERSITY on 08/18/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}

findings.

Additionally, correlations among the TUG test and other balance assessments should be

investigated. There are many valid and reliable ways to measure various aspects of balance. For

example, the OptoGait System alone has over 40 different test protocols that could measure aspects

of balance differently than the gait test we utilized. Investigating the relationship of the TUG Test

with these different variables could add to further verify the Timed Up and Go as clinically valid

and reliable screening test for fall risk. Many other balance evaluation tests exist and considering

the relationship of the TUG Test to other tests only strengthens our ability to determine who is at

risk for falls.

In conclusion, the purpose of this study was to investigate the Timed Up and Go (TUG)

Test and evaluate its validity as a clinical tool for diagnosis of balance impairments that could lead

to increased fall risk in senior citizens. TUG performance time was correlated with several

spatiotemporal gait variables that were collected via the OptoGait System. A higher number than

expected of these correlations were significant, thus indicating that the TUG performance was

associated with varying degrees of balance in otherwise unimpaired senior citizens. Based on

these findings, we assert that the TUG Test is a clinically valid tool for the evaluation of fall risk

in senior citizens.
 “Validation of the Timed Up and Go Test for Assessing Balance Variables in Adults Aged 65 and Older”
by Nightingale CJ, Mitchell SN, Butterfield SA
Journal of Aging and Physical Activity
© 2018 Human Kinetics, Inc.

References

Alexandre, T.S., Meira, D.M., Rico, N.C., & Mizuta. S.K. (2012). Accuracy of Timed Up and
Go Test for screening risk of falls among community-dwelling elderly. Brazilian Journal of
Physical Therapy, 16(5), 381-388. http://dx.doi.org/10.1590/S1413- 35552012005000041.

Granacher, U., Lacroix, A., Roettger, K., Gollhofer, A., & Muehlbauer, T. (2014). Relationships
between trunk muscle strength, spinal mobility, and balance performance in older adults.
Downloaded by USD UNIVERSITY on 08/18/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}

Journal of Aging and Physical Activity, 22, 490-498.

Lee, K. (2017). Effects of single and dual tasks during walking on spatiotemporal gait
parameters of community-dwelling older. Journal of Physical Therapy Science, 29(10), 1874-
1877.

Lienhard, K., Schneider, D., & Maffuletti, N.A. (2013). Validity of the Optogait photoelectric
system for the assessment of spatiotemporal gait parameters. Medical Engineering and Physics,
35, 500-504.

Matsumoto, H., Hagino, H., Sageshima H., Osaki, M., Tanishima, S., & Tanimura, C. (2015).
Diagnosis of knee osteoarthritis and gait variability increases risk of falling for osteoporotic
older adults: The GAINA study. Osteoporosis and Sarcopenia, 1, 46-52.

Podsiadlo, D., & Richardson, S. (1991). The timed “Up & Go”: a test of basic functional
mobility for frail elderly persons. Journal of the American Geriatrics Society, 39(2), 142-148.

Salzman, B. (2010). Gait and balance disorders in older adults. American Family Physician
Journal, 82(1), 61-68.

Shumway-Cook, A., Brauer, S., & Woollacott, M. (2000). Predicting the probability for falls in
community-dwelling older adults using the timed up & go test. Physical Therapy, 80, 896-903.

STEADI (n.d.) Algorithm for Fall Risk Assessment & Interventions. Retrieved from
(https://www.cdc.gov/steadi/pdf/Algorithm_2015-04-a.pdf)

Steffen, T.M., Hacker, T.A., & Mollinger, L. (2002). Age- and gender-related test performance
in community-dwelling elderly people: Six-minute walk test, Berg balance scale, timed up & go
test, and gait speeds. Physical Therapy, 82, 128-137.

Stöckel, T., Jacksteit, R., Behrens, M., Skripitz, R., Bader, R., & Mau-Moller, A. (2015). The
mental representation of the human gait in young and older adults. Frontiers in Psychology, 6,
943-952.

Strupeit S., Buss A., & Wolf-Ostermann, K. (2016). Assessing risk of falling in older adults – A
comparison of three methods. Worldviews on Evidence-Based Nursing, 13(6), 349-355.
 “Validation of the Timed Up and Go Test for Assessing Balance Variables in Adults Aged 65 and Older”
by Nightingale CJ, Mitchell SN, Butterfield SA
Journal of Aging and Physical Activity
© 2018 Human Kinetics, Inc.
Downloaded by USD UNIVERSITY on 08/18/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}

Figure 1: simulated OptoGait Set-Up

 “Validation of the Timed Up and Go Test for Assessing Balance Variables in Adults Aged 65 and Older”
by Nightingale CJ, Mitchell SN, Butterfield SA
Journal of Aging and Physical Activity
© 2018 Human Kinetics, Inc.

Table 1: Health Screening Questions.

1. Do you currently have ankle, knee, and/or foot pain?

2. Have you had foot, knee, or hip surgery within the past six months?

3. Do you have a history of hearing loss?

Downloaded by USD UNIVERSITY on 08/18/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}

4. Have you fallen at anytime in the last six months?

5. Have you been dizzy when standing or walking anytime in the last six months?

6. Do you need a walker, cane, or other assistance when walking?

7. Have you been diagnosed with any conditions that could make walking or balance
difficult for you in the last six months?
 “Validation of the Timed Up and Go Test for Assessing Balance Variables in Adults Aged 65 and Older”
by Nightingale CJ, Mitchell SN, Butterfield SA
Journal of Aging and Physical Activity
© 2018 Human Kinetics, Inc.

Table 2: Mean and Standard Deviation Scores for OptoGait Test Variables.

Variable Mean Score Standard Deviation

Step Length 60.39 cm 7.62

Stride Length 118.00 cm 8.58

Downloaded by USD UNIVERSITY on 08/18/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}

Contact Time .62 sec .07

Contact Phase .08 sec .03

Foot Flat .45 sec .06

Propulsive Phase .21 sec .05

Average Speed 1.11 m/s .05

 “Validation of the Timed Up and Go Test for Assessing Balance Variables in Adults Aged 65 and Older”
by Nightingale CJ, Mitchell SN, Butterfield SA
Journal of Aging and Physical Activity
© 2018 Human Kinetics, Inc.

Table 3: TUG and Gait Test Variable Correlations and Significance.

Correlation R p

Mean Step Length & TUG -.5133 0.000061 *

Mean Stride Length & TUG -.6086 <.00001*

Downloaded by USD UNIVERSITY on 08/18/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}

Mean Contact Time & TUG -.0677 .6369

Mean Contact Phase & TUG -.0718 .6137

Mean Foot Flat & TUG .2635 .0617

Mean Propulsive Phase & .2159 .1281

TUG

Mean Average Speed & TUG -.6762 <.00001*

SD Step Length & TUG .2988 .0332

SD Stride Length & TUG -.0078 .9567

SD Contact Time & TUG .1878 .1869

SD Foot Flat & TUG .7616 <.00001*

SD Propulsive Phase & TUG .7842 <.00001*

SD Average Speed & TUG .2885 .0401

* = significant at Bonferroni Corrected Value of p<.004