Does More Practice Improve Arm Movement After Stroke?

Dose has emerged as a key factor promoting functional recovery after stroke. Currently, a lack of data on the dose-response relationship impedes progress in the field of stroke rehabilitation. The goal of the proposed project is to define the range of doses of movement practice that produce the greatest improvements in outcomes in people with chronic stroke. Borrowing from animal models of stroke, dose in humans can be quantified by the number of repetitions of task-specific practice.

Our central hypothesis is that there exists a range of doses for people with stroke, below which, there is minimal benefit, and above which, further practice does not result in further benefit. The range of beneficial doses is likely to vary based on the severity of motor deficits and the presence of non-motor deficits in other domains. Using a randomized, parallel dose-response design, we will evaluate the benefits of four different doses of task-specific upper extremity training with matched schedules of 1 hr sessions, 4 sessions/wk for 8 wks, in 100 people with chronic stroke. Total repetition doses to be evaluated (3200, 6400, 9600, & individualized-maximum) are based on our preliminary data. The individualized-maximum group may extend their sessions beyond 8 wks until meeting defined stop criteria.

Our primary aim will test whether larger total doses result in better outcomes than smaller total doses. Benefits of the four doses will be evaluated at the impairment, activity, and participation levels, since understanding the dose-response relationship at all levels of measurement is critical for advancing rehabilitation research. We hypothesize that improvements will be greatest in the 9600 and individualized-maximum, followed by the 6400, and then the 3200 repetition dose groups. Our secondary aim is to characterize the dose-response relationship of upper extremity task-specific practice. With data from multiple assessment points, individual curve modeling will be used to estimate dose ranges, below which, there is minimal benefit, and above which, further practice does not result in further benefit. Furthermore, we will determine how various factors modify the dose estimates. We hypothesize that the severity of motor deficits will be the primary modifier of the dose-response relationship, with larger doses needed for those with more mild motor deficits. We further expect that needed doses will be larger for those with depression and hemispatial neglect.

Our team is well-positioned to investigate the critical issue of dose because of our expertise in stroke rehabilitation research and measurement, our understanding of the challenges of clinical practice and clinical research, and our ready access to this patient population. Expected outcomes from this project are empirically-driven estimates indicating the amount of movement practice required to drive maximal improvements and how these estimates can be individually modified for people undergoing stroke rehabilitation. Our estimates will immediately impact rehabilitation research and clinical practice. The importance of this project transcends stroke rehabilitation; our primary results will be of high value to many other rehabilitation populations also impeded by the lack of knowledge regarding dose-response relationships.