Functioning and activity outcomes of the Akwenda Intervention Program for children and young adults with cerebral palsy in Uganda: A cluster-randomized trial.

The control group improved significantly on the GMFM-66 and on many daily-function scores, which the authors did not expect. They suggest contamination from the community communication and advocacy work reaching control villages, antiseizure medication given to both groups, newly opened non-governmental clinics nearby, and the general effect of taking part in a study. Before and after assessments were done by two different teams, and whether the two teams scored children the same way was never checked. The authors say the second team could have scored systematically higher, though this would have affected both groups. Blinding may not have held: assessors could have guessed which group a child was in from where the family lived or whether the child had an assistive device. The program was many things at once, so no single component can be credited with the improvement, and the paper never separates them. Three children in the control group died before follow-up and were not included in the analysis, and 35 younger children were recruited separately, so the cohort was no longer strictly population-based. All participants lived in one rural district of Uganda and had almost no previous therapy, which the authors suggest is why the changes were large. The size of improvement may not transfer to children already receiving rehabilitation services. This paper reports only the child functioning and activity outcomes. Outcomes for caregivers and for the community are in separate, later reports.

Funded by the Swedish Research Council and three Swedish foundations: Stiftelsen Frimurare Barnhuset i Stockholm, Stiftelsen Sunnerdahls Handikappfond and Stiftelsen Promobilia. The authors declared no interests that might be seen as a conflict or bias. No company that sells a treatment is named as a funder.

Do not read the differences between the groups as the program's true effect. The control group also improved significantly on the gross motor measure and on many daily-function scores, and the team that measured the children after the program was different from the team that measured them before, with no check that the two teams scored children the same way.