23683: Predicting Cognitive and Motor Functional Outcomes Using Admission Cognition in Patients PostAcute Stroke

Document Type


Publication Date


Publication Title

Journal of Neurologic Physical Therapy: Academy of Neurologic Physical Therapy Poster Presentations


Purpose/Hypothesis: Stroke is the leading cause of long-term disability in the United States. Stroke can lead to cognitive impairments that compromise functional independence, such as memory loss and impaired attention. The purpose of this study was to determine whether a cognitive assessment could prospectively predict functional outcomes in patients post-stroke attending an inpatient rehabilitation facility (IRF). We hypothesized that a cognitive assessment would predict cognitive but not motor functional outcomes. Number of Subjects: Thirty subjects (68.1±15.7 years, 15 female) post-acute stroke admitted to an IRF completed the study. Materials and Methods: Cognition was assessed with the Trail Making Test Version A (TMT-A) and Version B (TMT-B), and functional independence was assessed with the cognitive and motor aspects of the Functional Independence Measure (FIM), administered near admission and discharge from the IRF. Admission and discharge FIM scores were used to calculate a cognitive and motor Montebello Rehabilitation Factor Score (MRFS), a well-established measure of relative gain in function during an IRF stay. Descriptive statistics and linear regression analyses were calculated using SPSS v24. Results: The overall regression model was statistically significant for the ability of admission TMT-A scores or TMT-B scores to independently predict cognitive and motor functional outcomes, as assessed by the MRFS (p<0.05). Specifically, poorer cognition at admission, as assessed by the TMT-A or TMT-B, predicted lower cognitive and motor MRFS. In contrast, change in TMT-A or TMT-B scores from admission to discharge from the IRF did not predict cognitive or motor MRFS (p>0.05). Conclusions: Results from this study indicate that a cognitive assessment near admission to an IRF, such as the TMT, can prospectively predict changes in functional outcomes in patients post-acute stroke. Of interest, the cognitive assessment predicted motor functional outcomes in addition to cognitive functional outcomes. The TMT-A and TMT-B assess perceptual speed and fluid intelligence, suggesting the importance of intact executive functioning at admission to an IRF in functional recovery post-stroke. Conversely, change in cognition from admission to discharge from an IRF did not predict cognitive or motor functional outcomes. The inability of changes in TMT scores to predict even the cognitive MRFS may be attributed to the communication and social cognitive aspects heavily assessed by the cognitive FIM, which may be unaffected by improvements in cognitive speed and fluid intelligence assessed by the TMT. Clinical Relevance: Poorer performance on a cognitive assessment near admission to an IRF was predictive of less improvement in cognitive and motor functional recovery, suggesting that patients with better cognition at admission have greater opportunities for improvement, despite the common contradictory assumption. Physical therapists should consider performance on cognitive assessments, such as the TMT-A or TMT-B, when determining a patient’s prognosis and establishing the plan of care.

Clinical Institute

Neurosciences (Brain & Spine)