Abstract
Objective:
To determine the amount of variability in scores on activity and participation measures used to assess ambulatory individuals with cerebral palsy explained by strength, body composition, gait impairment and participant characteristics.
Design:
Multicentre prospective cross-sectional study.
Setting:
Seven paediatric–orthopaedic specialty hospitals.
Participants:
Three hundred and seventy-seven ambulatory individuals (241 males, 136 females) with cerebral palsy, Gross Motor Function Classification System (GMFCS) levels I–III (I = 148, II = 153, III = 76), ages 8–18 years (mean 12 years 9 months, SD 2 years 8 months).
Methods:
Participants completed assessments of GMFCS level, patient history, lower extremity muscle strength, Gross Motor Function Measure (GMFM-66), Pediatric Outcomes Data Collection Instrument (PODCI), instrumented gait analysis, 1 minute walk test, Timed Up-and-Go and body composition. Multiple linear regression and bootstrap analyses were performed for each outcome measure, stratified by GMFCS level.
Results:
The amount of variability in outcome measures explained by participant characteristics, strength, and gait impairment ranged from 11% to 50%. Gait impairment was the most common predictor variable and frequently explained the greatest variance across all outcome measures and GMFCS levels. As gait impairment increased, scores on outcome measures decreased. Strength findings were inconsistent and not a primary factor. Body composition contributed minimally (<4%) in explaining variability. Participant characteristics (cerebral palsy type, gestational age and age at walking onset), were significant predictor variables in several models.
Conclusions:
Variability in outcome measure scores is multifaceted and only partially explained by strength and gait impairment illustrating the challenges of attempting to explain variation within this heterogeneous population. Clinicians treating individuals with cerebral palsy should consider this when developing treatment paradigms.
Get full access to this article
View all access options for this article.