Abstract: Objective To investigate the correlation between gross motor function and risk factors in children with cerebral palsy (CP). Methods We included the children aged 2-12 years old who were diagnosed with CP and registered in Chengdu Federation of Disabled Persons in 20 districts of Chengdu. Gross motor function was estimated by gross motor function classification system (CMFCS). Risk factors of the CP children were investigated retrospectively. Univariate and multivariate logistic regression analyses were used to analysis the correlation of the CP risk factors and GMFCS grades in the children with CP. Results A total of 333 children were identified with a male to female ratio of 1.15∶1,an average age of （6.32±2.74） years old, and dominated in 6-<12 age group (197 cases,59.16%) and spastic CP (228 cases,68.47%).Thirty risk factors for CP were included, among which the majors were listed as premature delivery （185 cases，55.56%）,low body mass （150 cases，45.05%）,and neonatal hypoxic-ischemic encephalopathy （104 cases，31.23%）.No risk factors of CP were found correlating with GMFCS grades by univariate logistic regression analysis. Stepwise multiple logistic regression analysis showed no risk factors of CP in the regression equation. Conclusion The risk factors of CP can not be used to predict the severity of gross motor function of children with CP.