Shoulder and Wrist Kinematic Variability is Different Between Box and Block Test Trials

Introduction: Evaluation of upper extremity (UE) movement is commonly achieved using functional tasks and activities of daily living (ADLs).1 The Box and Blocks test (BBT) is an example of a functional task used to measure manual dexterity.2 Motion capture methods have been used to quantify joint movements during functional tasks. Differences in movement variability between trials, which could also indicate task learning, could lead to kinematic data being misrepresented if the data that is analyzed is collected during the learning phase.3 Changes in variability could indicate kinematic changes made between trials to improve the BBT score, increasing the need for more practice trials to accurately observe UE motor improvement without the influence of task learning. Therefore, it is important to understand UE movement variability over time to better interpret joint kinematic metrics and ultimately inform rehabilitation progress. Therefore, the purpose of this study was to determine if the Coefficient of Variation (CV) is significantly different between trials. The CV is expected to significantly decrease between trials due to task learning.

Materials and

Methods: Through an institutional review board-approved study (#22-662), 21 typically developing children with full upper extremity mobility were recruited and tested. An 8-camera, markerless motion capture system (Theia Markerless Inc., Kingston, ON, Canada) was used to capture data at 85 Hz. Each participant completed three trials of the modified BBT for each limb. The modified BBT required the participant to move wooden blocks from one compartment to another over a divider while aiming to place the box in a target compartment as fast as possible. Participants completed three trials each lasting 60 seconds. Kinematic data was processed in Visual3D (C-Motion, Inc. Germantown, MD), to obtain hand segment position and velocity, shoulder, elbow, and wrist joint angle data. The middle 15 successful block movements (SBM) for the dominant limb for each trial were analyzed using a custom MATLAB code. The range of hand position and velocity and range of motion (ROM) of flexion/extension (FLEX/EXT), abduction/adduction (ABD/ADD), and internal/external (IN/EXT) rotation of the shoulder, elbow, and wrist were determined for each SBM within a trial, which was then used to determine the Coefficient of Variation (CV) for each outcome of interest for that trial. A linear mixed effects model was used with trial (T1, T2, T3), BBT score, and age as the fixed effects and participant ID as a random effect. It was expected that the CV would decrease between trials (α=0.05).

Results, Conclusions, and Discussions: The BBT score increased significantly (p < 0.001) between each trial, indicating performance improvement. Age did not have an impact on between trial differences in CV (p>0.05). The results showed that shoulder (p=0.028) and wrist (p=0.010) FLEX/EXT ROM CV were different between trials (Figure 1), while no other measures of interest were found to be significant. Shoulder FLEX/EXT ROM showed a decrease in CV between T1 and T2 (p=0.021), while T3 was similar to T1 (p=0.477) and T2 (p=0.250). Additionally, wrist FLEX/EXT ROM was similar between T1 and T2 (p=0.968) but showed an increase from T1 to T3 (p=0.017) and T2 to T3 (p=0.032). The changes in the shoulder and wrist only could indicate the importance of these joints to the BBT task given prior research that indicates that the wrist and shoulder are key to hand function and task completion.4,5 Our results partially agree with our hypothesis that shoulder FLEX/EXT did decrease with trial repetition. However, variability increased across trials in the wrist, suggesting that there is still task learning for the BBT after 3 trials. Based on these results it is important to ensure that clinical testing is completed after task learning has stabilized and there is little to no difference in movement variability between trials.

Movement variability decreased for the shoulder FLEX/EXT and increased for the wrist FLEX/EXT as task performance improved during a repetitive UE functional task. Therefore, the number of trials completed would influence kinematic metrics quantified during the task and should be considered if examining rehabilitation programs and motor control improvement over time. Future studies can determine how many practice trials are needed to minimize the impact of task learning on kinematic assessment. Further exploration into understanding how variability changes throughout the movement cycle is also needed to better understand task-learning strategies.

Acknowledgements (Optional):