Considerations when assessing cognition in children with cerebral palsy.

Objective: To describe cognitive assessment results in relation to clinical characteristics in children with cerebral palsy (CP) and to compare methods for calculating composite scores, taking their motor impairments into account. Method: All children registered in the Norwegian Quality and Surveillance Registry for Cerebral Palsy born 2002-2014 were included. The contribution of clinical characteristics per assessment results were investigated using one-way ANOVA (eta-squared (η2) effect size) and multiple linear regression analyzes. Composite scores were compared with paired samples t-test (Cohen's d effect size). Results: Of the 1532 children with CP registered, 972 (63.4%) had a cognitive assessment. An Intelligence Quotient (IQ) score was available for 451 (46.4%) of the assessed children (range 40 - 129, mean 82.3). Twenty-two percent had an intellectual disability. Mean IQ was lowest in children with spastic quadriplegic CP, epilepsy, and/or severe hearing impairment (effect sizes were small, η2 = 0.05, 0.05 and 0.01, respectively). The Verbal Comprehension Index score was highest and Visual Spatial Index lowest, with moderate effect size, d = 0.7. Fine motor skills and speech ability contributed most towards explaining variability in Wechsler Full Scale IQ (FSIQ). General Ability Index, Fluid Reasoning Index and Nonmotor Full Scale Score were higher than FSIQ, with small to moderate effect sizes of d = 0.61, 0.46 and 0.62, respectively. Conclusions: Variability in cognitive functioning underscores the importance of individual assessments. Better scores were obtained on composite measures using subtests that did not place high demands on fine motor skills, indicating a need for tailoring assessment practice to children with CP.