[Clinically Applicable] Summary of Measurement Properties of the TUG Test in Parkinson's Disease | Organizing Reliability, Validity, and Responsiveness from Systematic Reviews

Do you have these concerns in your clinical practice?

・How reliable is the TUG test when used for patients with Parkinson's disease?
・Are you worried about being able to answer when asked for the basis of its validity?
・To what extent does it have responsiveness (ability to detect treatment effects)?

In this article, based on a systematic review by Mollinedo & Cancela (2020, Journal of Exercise Rehabilitation), we will
organize the measurement properties of applying the TUG test to patients with Parkinson's disease (PD) and
summarize them.

What you will learn in this article

✔ Evidence levels for the reliability, validity, and responsiveness of the TUG test
✔ Relationship with balance, gait, and disease stage
✔ Precautions for use in clinical and research settings

Conclusion (Key Points)

① Reliability is "good to moderate":
With an ICC range of 0.69–0.99, it is generally good for both test-retest and inter-rater reliability.

② Validity is "good" when compared with balance assessments:
It shows a particularly high correlation with the Mini-BESTest and the Berg Balance Scale.

③ Responsiveness is "insufficient":
With an effect size (ES) of 0.11–0.16, it is low, and there are limitations in detecting changes due to treatment.

1. Background and Purpose of the Study

What is the TUG test?

• Developed by Podsiadlo & Richardson (1991)

• Measures the time from standing up from a chair, walking 3m, turning around, and sitting back down

• Used to evaluate dynamic balance, mobility, and fall risk

• Original ICC: 0.99 for both inter-rater and intra-rater reliability (for ages 60–90)

Why is a re-examination necessary for PD?

PD presents with the following characteristic motor impairments,
and different evaluation properties from the general elderly population are expected.

• Forward-leaning posture, shuffling gait, and freezing of gait

• Postural reflex impairment, bradykinesia, and rigidity

• Diurnal fluctuations in motor function due to On/Off states

Purpose of the Review

To systematically organize the reliability, validity, and responsiveness of the TUG test for PD patients and examine its applicability to clinical practice and research.

2. Overview of Methods

Criteria for evaluating measurement properties (from Table 1)

3. Main Results and Findings

① Reliability (9 studies)

Test-retest reliability

• ICC range: 0.69–0.99

• Evaluation: Moderate (+++) to Good (++++)

• Representative value: Morris et al. (2001) ICC = 0.99 (in both On/Off states)

Intra-rater and inter-rater reliability

• Inter-rater ICC: 0.95–0.99 (Good)

• Intra-rater ICC: 0.80–0.99 (Good to Moderate)

In cases using technology (iTUG, inertial sensors, etc.)

Van Lummel et al. (2016):
Test-retest ICC = 0.89–0.90, Inter-rater ICC = 0.95–0.97
Kleiner et al. (2018):
ICC = 0.995–0.997 for both optical and IMU
→ Reliability is ensured to be equal or better with digital measurement

Standard Error of Measurement (SEM)

• 0.59–3.43 (varies depending on the protocol)

• Significant differences in protocols between studies make simple comparisons difficult

② Validity (17 studies)

Comparison with Balance Assessment (Strongest Evidence)

Comparison with UPDRS (Parkinson's Disease Severity)

Correlation with UPDRS III (Motor Section): Moderate (+++)
Correlation with UPDRS Total Score: Weak (++) to Moderate (+++)
→ Higher association with motor domain than total score

Comparison with Freezing of Gait Assessment (FOG-Q, etc.)

Correlation with Freezing of Gait Questionnaire (FOG-Q): Poor (+) to Weak (++)
→ TUG is unsuitable for assessing freezing of gait

③ Responsiveness (Only 2 studies)

Foreman et al. (2011): ES = 0.11 (On state), 0.16 (Off state)
Johnston et al. (2013): ES = 0.16
Both evaluations are Poor (+)

4. Clinical Implications and Discussion

Recommended Situations for Use

✔ Quantitative assessment of dynamic balance: Validity is enhanced when combined with the Berg Balance Scale or Mini-BESTest

✔ Monitoring mobility: Suitable for tracking changes over time due to good test-retest reliability

✔ Screening for fall risk Spagnuolo et al. (2018): ROC analysis shows cutoff ≥ 2.2 seconds, sensitivity 0.85, specificity 1.0

Precautions for Use

Always record On/Off state: Standardizing evaluation conditions is a prerequisite for reliability as motor function fluctuates significantly depending on the state


Do not use for assessing freezing of gait: Use PD-specific assessment tools like FOG-Q in conjunction

Standardize protocols: Comparison between studies is difficult unless walking distance (standard 3m) and measurement methods are standardized

Interpretation by age group and disease stage: Interpretation of results differs between elderly and younger patients, and between stages I-II and III-IV

Future Research Topics

Extremely few studies have examined responsiveness (detectability of treatment effects) → High priority for future research. Lack of stratified analysis by gender and disease stage. Need for standardization of SEM using technology (sensors, smartphones).

Summary

The TUG test has established reliability and balance validity in PD, but
it has limitations in its ability to detect treatment effects (responsiveness).
In clinical practice, recording the On/Off state and using complementary assessments based on the objective are essential.

References

Mollinedo I, Cancela JM (2020).
Evaluation of the psychometric properties and clinical applications of the Timed Up and Go test in Parkinson disease: a systematic review.
Journal of Exercise Rehabilitation 16(4): 302–312.