Predictive ability of 2-day measurement of active range of motion on 3-mo upper-extremity motor function in people with poststroke hemiparesis

OBJECTIVE: We determined (1) whether active range of motion (AROM) of shoulder flexion and wrist extension measured at the initial therapy evaluation in the acute hospital predicted upper-extremity (UE) motor function 3 mo after stroke and (2) whether the presence of nonmotor impairments influenced this prediction. METHOD: We collected AROM data from 50 people with stroke during their initial acute hospital therapy evaluation and UE motor function data 3 mo later. Multiple regression techniques determined the predictive ability of initial AROM on later UE motor function. RESULTS: Initial AROM explained 28% of the variance in UE motor function 3 mo poststroke. Nonmotor deficits did not contribute to the variance. CONCLUSION: Compared with later AROM measurements, initial values did not adequately predict UE motor function 3 mo after stroke. Clinicians should use caution when informing clients of UE functional prognosis in the early days after stroke.