Reliability of upper limb kinematic variables: adults with unilateral cerebral palsy performing a standardized drinking task.

OBJECTIVE: To evaluate the test-retest reliability and agreement of kinematic variables measured with an optical motion capture system during a standardized drinking task in adults with unilateral cerebral palsy (CP). DESIGN: Test-retest. SUBJECTS/PATIENTS: 25 subjects (12 males) with spastic unilateral CP, aged 18-60 years. METHODS: Kinematic variables were collected while participants performed a drinking task with the affected upper limb on 2 consecutive days. Inter-session reliability and agreement were estimated using the intra-class correlation coefficient (ICC) 2.1, and standard error of measurement (SEM). RESULTS: Reliability was excellent for all investigated variables (ICCs: 0.90-0.99). Agreement for elbow flexion/extension, shoulder flexion, and abduction: SEM ≤ 3.1 degrees, max trunk displacement during reaching: SEM = 9.9 mm, number of stops/starts during total movement: SEM = 2.1 movement units, total movement time: SEM = 1.0 s. CONCLUSION: Optical motion capture of a standardized drinking task was reliable for evaluating upper limb movements in adults with spastic unilateral CP. The method may be used as an outcome measure to evaluate the effect of treatments of upper limb function, as well as to follow the function over time.