Effect of task-oriented training assisted by force feedback hand rehabilitation robot on finger grasping function in stroke patients with hemiplegia: a randomised controlled trial.

Single centre, single-blind design: patients and therapists knew which group they were in, which could influence effort and performance. Four-week duration with no follow-up, so it is unknown whether the between-group differences persist. Small sample (40 analysed) with no stratification by age, sex, stroke type, or time since onset. The glove assists finger flexion only and has no extension function; the therapist had to help the patient release objects. No objective neurophysiological measures (EMG, fMRI, fNIRS) to confirm a mechanism. Four participants dropped out (one in the control group had unrelated health issues and attended only one session; three were discharged).

Do not read the cylindrical and spherical grasp sub-item differences (p = 0.031, p = 0.015) as evidence the robot specifically improves those grasp types. The authors applied Bonferroni correction across seven sub-items, and neither difference survived (p > 0.007). The between-group advantage is in the total scores, grip strength, and range of motion, not in any single grasp pattern.