Rehabilitation increasingly addresses the International Classification of Functioning, Disability and Health's (ICF) concepts of activity and participation, but little is known about associations between changes in body functions and structures, activity, and participation. We conducted a before-and-after study of 35 ambulatory children with spastic diplegia or hemiplegic cerebral palsy, mean age 5 years 6 months (SD 2y 2mo). Children were in Gross Motor Function Classification System (GMFCS) Levels I (n=11), II (n=12), or III (n=12). We assessed body functions and structures, activity, and participation at baseline and at 2 months and 6 months postbotulinum toxin type A (BoNT-A) injection. Repeatedmeasures analysis of variance evaluated change, and linear regression assessed relationships. Baseline score relationships were moderate to strong but, despite similar directions of change at 2 months, change score relationships between measures of body functions and structures (spasticity and timed walk), activity (Gross Motor Function Measure and Pediatric Evaluation of Disability Inventory), and participation (Pediatric Outcomes Data Collection Instrument) at 2 months and 6 months were poor to fair (r<0.40). Predictor combinations accounted for <69% of variation in activity and participation change scores. Predictors often pertained to baseline score, GMFCS level, or age. Relationships between changes at different ICF levels are complex, and activity and participation gains post-BoNT-A are likely to be influenced by the child and environment factors.The World Health Organization's International Classification of Functioning, Disability and Health (ICF) model 1 is commonly used as a framework for thinking about evaluation, goals, and interventions. This is reflected in the move beyond the traditional assessment and intervention focus on body functions and structures to an emphasis on goals and outcomes related to activity (carrying out tasks) and participation (involvement in life situations). What remains unclear is the extent to which outcomes at various ICF levels are associated. While the ICF model 1 illustrates the multidirectionality of relationships, this new framework has yet to be empirically tested. 2 Lack of understanding of outcome relationships among ICF component variables can make it challenging to use the ICF framework for goal planning and intervention; e.g. if knee strength increases (body functions and structures), to what extent might stair climbing abilities (activity) improve, and what might happen to the child's abilities to go shopping with friends at a nearby shopping mall (participation)?ICF relationships have commonly been studied in the context of cross-sectional measurement studies, exploring associations between ICF level outcomes at a single time. Work in children with cerebral palsy (CP) has verified the complexity of associations. [3][4][5] Studies that evaluate the effectiveness of rehabilitation interventions typically incorporate several outcome measures of body functions and stru...