Objective To evaluate whether race or ethnicity was independently associated with parental refusal of consent for their child's participation in a multisite pediatric critical care clinical trial. Study design We performed a secondary analyses of data from Randomized Evaluation of Sedation Titration for Respiratory Failur e (RESTORE), a 31-center cluster randomized trial of sedation management in critically ill children with acute respiratory failure supported on mechanical ventilation. Multivariable logistic regression modeling estimated associations between patient race and ethnicity and parental refusal of study consent. Result Among the 3438 children meeting enrollment criteria and approached for consent, 2954 had documented race/ethnicity of non-Hispanic White (White), non-Hispanic Black (Black), or Hispanic of any race. Inability to approach for consent was more common for parents of Black (19.5%) compared with White (11.7%) or Hispanic children (13.2%). Among those offered consent, parents of Black (29.5%) and Hispanic children (25.9%) more frequently refused consent than parents of White children (18.2%, P  < .0167 for each). Compared with parents of White children, parents of Black (OR 2.15, 95% CI 1.56-2.95, P  < .001) and Hispanic (OR 1.44, 95% CI 1.10-1.88, P  = .01) children were more likely to refuse consent. Parents of children offered participation in the intervention arm were more likely to refuse consent than parents in the control arm (OR 2.15, 95% CI 1.37-3.36, P  < .001). Conclusions Parents of Black and Hispanic children were less likely to be approached for, and more frequently declined consent for, their child's participation in a multisite critical care clinical trial. Ameliorating this racial disparity may improve the validity and generalizability of study findings. Trial registration ClinicalTrials.gov : NCT00814099.