Purpose: To examine the relationship between change scores on the log mean Wolf Motor Function Test (lmWMFT) and the intensity of supervised Constraint Induced Movement Therapy (CIMT) in participants with subacute and chronic stroke.
Methods: A retrospective analysis of data from 169 EXCITE participants who received CIMT either immediately after randomization or one year later was undertaken. During waking hours, participants wore a restraining mitt on the less affected extremity. The lmWMFT was administered before and after the two week treatment block.
Results: Significant relationships were seen between the intensity of training and functional score in the immediate, lower functional group for whom more training in adaptive task practice resulted in poorer outcomes (p=0.01) and in the immediate, higher functioning group for whom more training in repetitive task practice resulted in poorer outcomes (p=0.02). Female participants in the immediate group showed less progress in lmWMFT scores with greater amounts of total training (p=0.01). Functional level, gender, and concordance did not modify any other relationship. Both higher functioning participants who trained within the normal ratio (N=50) and who were exposed to more than the prescribed adaptive task practice (N=11) experienced a significant improvement in the lmWMFT score (p=0.03 and p=0.02, respectively) compared to those higher functioning participants who experienced excessive repetitive task practice.
Conclusion: Applying CIMT to a large sample of participants with stroke resulted in directionally inappropriate but significant relationships between intensity and lmWMFT scores in the immediate but not the delayed group. Our data also suggest that functional improvements observed in the EXCITE Trial might be attributable to training components other than the designated ratio of training approaches (adaptive and repetitive task practice).