After stroke many patients remain with difficulties in using the upper limbs, balance, transfers, and walking. The Motor Assessment Scale (MAS) assess the movements needed to do those activities. Social restrictions to combat the Covid-19 pandemic have increased the telerehabilitation, but the remote assessment is also important for rural areas or geographic regions where neurorehabilitation specialists are scarce, and when patients have difficulties on transport to the clinic. Although the measurement properties of the MAS applied in person are established, the validity and reliability of the MAS applied via teleconsultation is unknown.

To investigate the validity and reliability of the Motor Assessment Scale (MAS) when administered remotely by videoconference (Tele-MAS).

This is a study of investigation of measurement properties, following the recommendations of COSMIN, for validity and reliability of Tele-MAS. The sample was 18 participants with a diagnosis of stroke, Braztel-MMSE score ≥13 points and with internet access and mobile device. The order of the evaluations (remote or in person) was randomly defined. The application sequence of the MAS items was adapted to allow remote application in addition to verbal commands during the evaluation and a specific instruction manual for application by videoconference was developed. For assessment by videoconference (rater A and B) the participant was instructed to position the camera in a way that the therapist can observe from the ground to above the head. The raters are positioned similarly to the participant to demonstrate the items and score synchronously. The application in person takes place in the participant's house, by rater A. The three collections took place within a period of eight days, with an interval of 2 days. The validity between the in person and remote evaluation was analyzed by Pearson's correlation coefficient and the reliability between the remote and in person evaluation was analyzed using the Bland-Altman limits of agreement. The interrater reliability for the sum of the score of the items was analyzed by the Intraclass Correlation Coefficient (ICC) with a confidence interval of 95% (95%CI). Data were analyzed using MedCalc software®.

A strong positive correlation (r=0.96; 95%CI=0.90-0.99) was found between the MAS and Tele-MAS scores. The analysis of the Bland-Altman graph for the total MAS scores indicated that the mean difference between in person and remote scores was -0.6 points. The 95% limits of agreement are +6.5 and -7.6 points. Excellent interrater reliability (ICC≤0.94; 95%CI =0.84-0.98) of the total Tele-MAS score.

The preliminary results indicate that Tele-MAS should be considered as a valid measure and has interrater reliability.

MAS is widely used in clinical practice, however, the measurement properties of the remote version applied by videoconference were unknown. The results of the final study should present the validity and reliability of Tele-MAS to remotely assess post-stroke motor function through videoconference.