Cognitive approaches in the rehabilitation of upper limbs function in children with cerebral palsy: a systematic review and meta-analysis.

The trials were small: 7 to 35 children per group, and the pooled analyses combined only 33 to 49 children per group, so they cannot reliably detect modest effects. The trials differed widely in population, dose, duration and delivery (action observation ran from 9 days to three months, 3 to 5 sessions a week, 15 to 120 minutes a session), so the pooled numbers average over genuinely different treatments. One trial's Assisting Hand Assessment data could not be included because they were reported in logits, and the authors could not obtain them from the trial authors even after asking. Mirror therapy could not be pooled at all, because the trials used incompatible scales or bundled mirror therapy with other treatments. Four of those five trials scored only 4 to 6 out of 10 on the PEDro quality scale. None of the mirror therapy trials measured children after treatment ended, so nothing is known about whether gains persisted. Children ranged from 3 to 16 years old, and the authors note it is still unclear at what age children can reliably use mental strategies, which makes the youngest participants hard to interpret. No included trial tested motor imagery, so the review says nothing about one of the three approaches named in its title. The limitations section of the paper as supplied introduces a list but the items themselves are not present in the text.

The authors declare no conflict of interest with any financial organisation regarding the material discussed in the manuscript. The supplied text does not state who funded the work.

Do not take the authors' conclusion that video action observation therapy provides better results at face value. The five trials they pooled showed no significant effect on any upper limb measure, and the encouraging results come from individual small trials rather than the combined analysis.