• Jakob Hansen
4. semester, Sports Technology, Master (Master Programme)
Objective: To validate visual examination of arthroscopy patients during challenging ambulation tasks, such as stair descent and single leg squat.
Method: Three arthroscopy patients and 12 healthy subjects performed a stair descent and single leg squat task. 3D motion capture measurements and 2D video recordings were collected. From the 3D motion capture measurement kinematics for trunk, pelvis, knee and foot were calculated, along with the kinetics of the knee; external knee adduction moment (KAM) and knee flexion/extension power. The 2D videos were visually rated by three physiotherapists and intra-rater and inter-rater reliability was calculated using first order agreement statistics (AC1). The agreement between the 3D motion capture measurements and the visual ratings were calculated, also using AC1.
Results: The combined intra-rater agreement was almost perfect (AC1 = 0.91-0.94). The inter-rater agreement was slight for the combined mean (AC1 = 0.29) and substantial deviations in agreements were found (range 0.96). Pairwise comparison of the raters showed discrepancy in agreements, with the highest combined mean of AC1 = 0.55 (range = 0.69) and the lowest combined mean of AC1 = 0.09 (range = 1.49). The comparison between the visual examination and the 3D motion capture method, resulted in a moderate combined mean agreement (AC1 = 0.50) with a substantial deviation (0.80). Comparing the individual raters to the 3D motion capture measurements resulted in a good agreement as the highest (AC1 = 0.72; range = 0.88) and a slight agreement as the lowest (AC1 = -0.04; range = 1.49).
Conclusion: The findings from this study imply that visual rating has an almost perfect intra-rater reliability, but a slight inter-rater reliability, when combining ratings from stair descent and single leg squat. The agreement between raters and 3D motion capture measurements were moderate and rater specific. The agreement between the KAM measured and the overall visual ratings, resulted in slight to moderate agreements- Therefore, the overall visual ratings does not seem to comply with the medial knee loadings of the subject.
Publication date2 Jun 2016
Number of pages16
ID: 234619778