To the Editor:
I read the article reporting the randomized crossover trial by Lee et al. (2021) in the January/February 2021 issue of the American Journal of Occupational Therapy with great interest. I appreciate the authors’ efforts to assess the effects of robot-assisted rehabilitation on the hand function of people with stroke. However, I would like to point out two concerns regarding the trial.
First, the authors indicated that the robot-assisted therapy produced significant improvements in upper extremity motor control and activities of daily living ability, although the difference between the robot-assisted therapy group and conventional therapy group was not significant. Such comparisons against a baseline within randomized groups can be misleading (Bland & Altman, 2011). This point seems to contradict the purpose of the randomized crossover trial.
Second, the authors reported that the robot-assisted therapy–first group was compared with the conventional therapy–first group in the Results section, but this is inconsistent with the description in Table 2.
Yoshitaka Wada, MD
Department of Rehabilitation Medicine
Showa University Fujigaoka Rehabilitation Hospital
Kanagawa, Japan
