TUG Fall Risk Assessment: A Complete Summary of Values and Interventions
[Positioning of this magazine] This article is positioned as Step 2: Major Balance Assessment Scales. TUG is a standard tool for fall risk screening that can also be used in the acute phase. Please use this in conjunction with Step 5, "Clinical Decision Flowchart," of this magazine.
Magazine: Evidence-Based Functional Assessment and Clinical Judgment
"If TUG is 13.5 seconds or more, it is a high risk for falls."
Just remembering that is actually insufficient.
In this article, based on 7 papers, I will summarize
cutoff values, MDC, and interventions for high fall risk all at once.
What you will learn in this article
Summary of TUG cutoff values by disease (with sensitivity and specificity)
MDC values by disease and how to use them
Specific interventions for patients with high fall risk
5 clinical points to keep in mind when using TUG
3 numbers you should remember first
List of cutoff values by disease
Points
Low sensitivity = many missed cases → It is dangerous to judge "safe" based on TUG alone.
Specificity of 0.74 = When it indicates high risk, the reliability is relatively high.
Cutoff values range from 10 to 33 seconds depending on the literature, so check the target population.
List of MDC by disease
Calculation example of the 15% rule
Baseline TUG = 15 seconds → A true improvement is 12.75 seconds or less
Baseline TUG = 20 seconds → A true improvement is 17 seconds or less
The golden rule when using MDC
Values vary significantly by disease → Do not use MDC from a different disease
Changes below the MDC should be interpreted as being within the margin of error
MDC < MCID is ideal (changes smaller than the margin of error are not clinically meaningful)
If TUG determines a high risk of falling
Intervention 1: Perturbation-based training (PBT)
From an RCT by Brüll et al. (Gerontology, 2023).
Compared two types of PBT in 71 elderly individuals (mean age 74.9) at risk of falling.
Two types of PBT
Program overview
3 times a week × 6 weeks (18 sessions total)
Adherence rate: PBTtreadmill 91%, PBTstability 87%
Serious adverse events: None
Main effects (PBTstability)
Significant improvement in fall risk index (Brief-BEST) (p=0.009, η²=0.131)
Significant improvement in Limits of Stability (p=0.020, η²=0.110)
Main effects (PBTtreadmill)
Significant improvement in reactive balance (Stepping Threshold Test) (p<0.001, η²=0.395)
💡 If equipment is unavailable, PBT stability using a balance pad is a practical option.
It has been reported to be more effective (-48% reduction)
than standard balance training (-24% reduction in falls).
Intervention ②: Multifactorial Approach
From an umbrella review (31 studies) by Beck Jepsen et al. (BMC Geriatrics, 2022).
TUG alone for fall prediction often yields "inconsistent results"
There is moderate evidence for gait speed (cutoff reference: 0.8 m/s)
No single assessment metric can predict fall risk with high certainty
Recommended combinations
TUG + Gait Speed + Berg Balance Scale (BBS)
Comprehensive assessment including cognitive function, environmental factors, and medication status
5 Clinical Considerations
① Do not judge fall risk based on TUG alone
Sensitivity of 0.31 = 70% of high-risk cases are missed
Useful for rule-in, but unsuitable for rule-out
Combine with BBS, gait speed, and Dual Task assessment
② Refer to MDC by disease and population
Knee OA: approx. 1.1 seconds
Chronic stroke: 3.2 seconds
Elderly neurological conditions: 15% of baseline
Do not apply MDC from different diseases
③ Always record On/Off status for Parkinson's disease
Results vary significantly between On (medication effective) and Off states
Comparisons are impossible unless status is recorded and standardized at each assessment
4. Standardize the evaluator and protocol
Values differ between intra-rater MDC (1.10 seconds) and inter-rater MDC (1.14 seconds)
Use the inter-rater MDC (1.14 seconds) when multiple evaluators are involved
5. TUG responsiveness (ability to detect treatment effects) has limitations
TUG responsiveness in stroke patients: SRM = 0.53 (moderate)
DGI (Dynamic Gait Index) for the same subjects: SRM = 0.89 (large)
Consider adding DGI or BBS if detecting treatment effects is a priority
Standard TUG Implementation Procedure
Standardizing the measurement protocol is a prerequisite for utilizing accurate cutoff values and MDC. When referencing values from research papers, it is important to ensure that measurements are taken under the same conditions as those in the paper.
Standard Procedure (based on Podsiadlo & Richardson, 1991)
1. Chair: With backrest and armrests, seat height approximately 46 cm
2. Starting position: Start from a seated position leaning against the backrest
3. Assistive devices: Walking aids normally used (canes, walkers, etc.) may be used as is
4. Sequence of movement: Stand up at the signal "Start" -> Walk to a marker 3m away -> Turn around and return -> Complete sitting down
5. Measurement: Measure the time in seconds from the "Start" signal until the subject is fully seated using a stopwatch
6. Number of trials: 1 practice trial + 1-2 actual trials is standard
Information that must be recorded: Presence and type of assistive device, name of evaluator (for inter-rater reliability management), On/Off state for Parkinson's disease patients, and footwear status
By standardizing and recording these conditions, the reproducibility of assessment results is improved, and comparison with values from other studies becomes possible.
Summary of Reference Papers
⚠️ This article is intended to introduce the contents of research papers. Each numerical value (MDC, cut-off values, etc.) varies depending on the target population and measurement conditions. Please make decisions regarding the selection of clinical evaluation indices based on the original papers and the condition of the patient in your care.
💡 For those who want to understand the concepts of MDC and MCID more deeply
We have published an article explaining from the basics how the MDC and MCID values covered in this article are calculated and why "change" alone is insufficient. If you want to organize the differences and proper usage of SEM, MDC, and MCID, please take a look at this as well.
→ What are MDC and MCID? — "Changed" and "Improved" are different stories
Related articles in this magazine
→ [Archived Edition] A Comprehensive Explanation of the Reliability of the Berg Balance Scale (BBS)
→ Clinical Decision Flowchart — Visualizing Fall Risk and Discharge Eligibility Judgments