Background: This study aimed to investigate the use of actigraphy (accelerometry) to measure disuse of the impaired arm in acute stroke patients. We correlated the National Institute of Health Stroke Scale (NIHSS) and the Fugl-Meyer Assessment arm section (FMA) findings with actigraphic data as a measure of validity. Methods: Thirty-nine acute ischemic stroke patients were included within 1 week after stroke onset. At inclusion, motor deficits were assessed by the NIHSS, FMA and 48-hour actigraphic recordings of both wrists were performed. Results: Moderate but highly significant correlations (Spearman’s rho) between actigraphic recordings and total NIHSS (ratio r = –0.59 and activity of impaired arm r = –0.75; p < 0.001) and FMA (ratio r = 0.54 and activity of impaired arm r = 0.69; p < 0.001) scores were found. Based on actigraphic motor activity scores, ROC curves were calculated following dichotomization of the population based on NIHSS = 7 and FMA = 45, showing good sensitivity and specificity, with negative predictive value of 100% and positive predictive value of 91% for the ratio variable. Conclusions: Moderate but highly significant correlations were found between actigraphy and the stroke scales NIHSS and FMA. Actigraphy was able to reliably discriminate less impaired from more impaired stroke patients with excellent sensitivity and specificity values. Actigraphy is a simple, valid, objective and reliable clinical research tool that can be used to determine motor impairment of the upper limb in stroke patients.