DAGDirected acyclic graph MACS Manual Ability Classification System REPACQ Registre de la Paralyse C er ebrale de Qu ebec SES Socio-economic status AIM Socio-economic differences in maternal and child health are well recognized, but the role of individual-level and area-level determinants in cerebral palsy (CP) phenotypes is debated. We set out to examine (1) the association between area-level and individual-level measures of socio-economic deprivation and CP phenotype among children, including subtype, severity, and comorbidities; and (2) the direct effect of area-level deprivation not mediated through individual-level deprivation.METHOD Regional data from a provincial CP register were analyzed. The outcome of interest was CP phenotype. The area-level exposure was measured using the Pampalon Deprivation Index. Individual-level socio-economic status (SES) was determined using maternal education. We conducted multiple regression models, stratified by preterm birth, controlling for key covariates, and a mediation analysis of area-level deprivation on the association between individual SES and CP phenotype.RESULTS A socio-economic gradient in mobility was seen in our cohort, above and beyond differences in maternal and perinatal factors. The added direct effect of area-level deprivation was seen only in children whose mothers were educated to a higher level, suggesting no additional contribution of area-level deprivation in children of mothers with a lower level of education.INTERPRETATION Contextual socio-economic factors can impact the severity of CP. These findings indicate important areas for potential community-level or area-level public health intervention (i.e. neighborhood reinvestment, preventive measures), and suggest that neighborhood-level research in maternal and perinatal health should continue to be pursued.Cerebral palsy (CP) is one of the most common motor disabilities in childhood, affecting approximately 2 per 1000 children, creating an important health burden for affected children, their families, and their communities. 1 Biological risk factors for CP include placental abnormalities, major and minor birth defects, and preterm delivery. 2 Beyond these, few studies have examined the social determinants of CP. A recent systematic review 3 found an association between area-level and, to a lesser degree, individual-level socio-economic status (SES) and risk of CP. Socio-economic differences in maternal and child health are well recognized, 4 but their role in CP remains debated. It is unclear if the distribution of CP cases and case severity is associated with environmental exposures.We hypothesize that there exist both individual-level and area-level socio-economic determinants of CP severity. At an individual level, maternal SES is recognized as a robust predictor of child health. 5,6 Studies to date suggest that children with CP, born to disadvantaged families, have more severe phenotypes as defined by spastic quadriparesis, 7 poor mobility, 8-10 and cognitive impairment. 7 We hypothesize that, b...