Relative and absolute test–retest reliabilities of biomechanical risk factors for knee osteoarthritis progression: benchmarks for meaningful change
Journal Articles
Overview
Research
Identity
Additional Document Info
View All
Overview
abstract
OBJECTIVE: Biomechanical factors are important treatment targets in knee osteoarthritis. The knee adduction (KAM) and flexion (KFM) moments, quadriceps strength and power, load frequency, and body mass index (BMI) all have the potential to affect knee articular cartilage integrity by modulating forces across the joint. To identify clinically meaningful change, however, these measurements must be reliable and sensitive to change. This study estimated relative and absolute test-retest reliabilities over long periods of biomechanical risk factors for knee osteoarthritis progression. METHOD: Data from a longitudinal, observational study were analyzed for knee osteoarthritis patients with data at baseline, 6-month and 24-month follow-ups. Gait kinematics and kinetics, quadriceps strength and power, daily load frequency and BMI were collected. Relative and absolute test-retest reliabilities of these measures were estimated using intraclass correlation coefficients (ICCs) and standard errors of measurement (SEMs), respectively. Minimal detectable change at the 95% confidence level (MDC95) was also calculated. RESULTS: Data from 46 participants [36 women; age 61.0 (6.6) years] were included. Good-to-excellent relative reliabilities (ICC ≥ 0.80) indicated that KAM peak and impulse, quadriceps strength and power, and BMI had a strong ability to discriminate amongst participants. Absolute reliabilities were high for quadriceps strength and BMI, which demonstrated reasonable within-participant variability (SEMs ≤ 11% of the mean). The MDC95 values supported use of clinical interventions effective in reducing BMI and KAM, and increasing quadriceps strength. CONCLUSION: These data are useful in interpreting findings from interventional or longitudinal investigations by determining whether observed changes are beyond measurement error and interpretable as true change.
