VR-based cognitive rehabilitation for children with traumatic brain injuries: Feasibility and safety.

Only 26 children at a single hospital took part, and the study was never designed with enough participants to test whether the training works. The authors call the executive function analysis exploratory. No group comparison on executive function reached statistical significance, so the trend in children with moderate to severe TBI is a description of what the numbers did, not a result you can rely on. The subgroups were very small. The children with mild complicated TBI, whose effect size pointed the other way, were few enough that this trend is just as insecure as the positive one. All sessions took place in the hospital. That helped children stick with the programme, but it probably kept out families who lived further away, and it is not how VR would usually be used at home. Assessors could not be fully blinded to which group a child was in, because children might mention the VR games while answering questions. Only the children were asked how they found the games. Caregivers and treating clinicians were not asked, so their view of whether it helped is missing. Attrition and missing data came from several directions: the pandemic, scheduling problems and a technical fault with the VR system, leaving some measures incomplete.

The authors declare no conflict of interest. The text provided does not name a funding source.

Do not read the promising trend as proof that VR cognitive training improves thinking skills after childhood TBI. The authors state that none of the group comparisons reached statistical significance, and the trend in favour of the intervention appeared only in an exploratory look at children with moderate to severe injuries.