TL;DR
This study investigated the optimal number of categories for visually rating the single-leg squat test (SLST) to balance detailed assessment with reliable results. They found that a three-category scale offers the best compromise, providing sufficient detail for clinical assessment of trunk deviation, hip adduction, and lower extremity internal rotation without sacrificing reliability, unlike more granular scales.
What Problem Does This Paper Address?
The single-leg squat test (SLST) is a common functional assessment for evaluating lower extremity movement quality, particularly in individuals with injuries. However, the visual rating of the SLST can be subjective, and there's no consensus on the ideal number of categories for scoring. Previous research has used scales with two to ten categories, leading to varying reliability results. This ambiguity makes it challenging for clinicians to consistently and accurately interpret SLST performance, hindering effective patient management and progress tracking. The authors aim to determine if a two, three, or four-category ordinal scale provides the most reliable and informative assessment of specific SLST components.
Methods
The researchers conducted a cross-sectional study involving 58 participants with lower extremity injuries. They evaluated the reliability of two-, three-, and four-category ordinal scales for rating three specific SLST components: trunk deviation, hip adduction, and lower extremity internal rotation.
- Intra-rater reliability was assessed by having one physical therapist rate the SLST performance of 29 participants at two time points, one week apart.
- Inter-rater reliability was assessed by having two experienced physical therapists (with 35 and 11 years of experience) independently rate the SLST performance of another 29 participants at the same time point.
The SLST itself was performed with participants standing barefoot, performing squats to approximately 60° knee flexion, with specific instructions to maintain balance and control. The rating criteria for each category (trunk deviation, hip adduction, lower extremity internal rotation) were detailed in an appendix. Reliability was quantified using unweighted kappa (κ) values, with established benchmarks for interpreting agreement strength (poor, fair, moderate, good, excellent).
Datasets
Not discussed in the paper. The study utilized data collected from 58 participants with lower extremity injuries.
Key Insights and Ideas
- The "sweet spot" for SLST rating: The study suggests that more categories don't necessarily mean better reliability or informativeness. A three-category scale appears to hit a sweet spot, balancing detail with consistency.
- Component-specific reliability: Reliability can vary significantly depending on which aspect of the SLST is being rated. While overall scores might show lower reliability with more categories, specific components can still be reliably assessed.
- Real-time vs. video assessment: The study used real-time visual assessment, which is more representative of clinical practice but potentially introduces more variability than video-based assessments that allow for pausing and replaying. This highlights a trade-off between ecological validity and control.
- Training is crucial: The study emphasizes the importance of standardized training for raters, even experienced ones, to ensure consistent application of rating criteria.
Key Results
- Overall Score Reliability:
- Inter-rater reliability for the overall score was good for two-category (κ=0.60) and three-category (κ=0.61) scales, but only fair for the four-category scale (κ=0.33).
- Intra-rater reliability for the overall score was moderate to good for the two-category scale (κ=0.60), but poor for the three- (κ=0.35) and four-category (κ=0.20) scales.
- Component-Specific Reliability:
- Trunk Deviation: Inter-rater reliability was good to excellent for two- (κ=0.83) and three-category (κ=0.82) scales, and fair to good for four-category (κ=0.55). Intra-rater reliability was moderate to good for two- (κ=0.65) and three-category (κ=0.67) scales, and fair for four-category (κ=0.45).
- Hip Adduction: Inter-rater reliability was good for two- (κ=0.65) and three-category (κ=0.69) scales, and fair for four-category (κ=0.59). Intra-rater reliability was moderate to good for two- (κ=0.58) and three-category (κ=0.70) scales, and good for four-category (κ=0.65).
- Lower Extremity Internal Rotation: Inter-rater reliability was good to excellent for two- (κ=0.83) and three-category (κ=0.80) scales, and fair to good for four-category (κ=0.79). Intra-rater reliability was moderate for two-category (κ=0.47), but poor for three- (κ=0.30) and four-category (κ=0.36) scales.
Limitations and Future Work
- Methodological Limitations: The study acknowledges that the generalizability of its findings might be limited by:
- The use of only two experienced physical therapists, suggesting similar training is necessary for reliable results.
- A 1-week interval for intra-rater reliability, which may not reflect longer reassessment periods common in clinical practice.
- A specific demographic (young, active individuals with recent lower extremity injuries), limiting broader relevance.
- A small sample size, which may have limited the statistical power to detect precise reliability estimates.
- Future Work: The authors suggest that future studies should:
- Include larger sample sizes to enhance statistical power and generalizability.
- Investigate the impact of different training protocols and rater experience levels on reliability.
- Explore longer reassessment intervals for intra-rater reliability.
- Further research could also explore how changes in specific movement components are best captured by different rating scales to monitor patient progress effectively.
Summary updated