Paediatric obesity management remains generalised to dietary and exercise modifications with an underappreciation for the contributions of eating behaviours and appetitive traits in the development of obesity.To determine whether treatment-seeking children and adolescents with obesity cluster into phenotypes based on known eating behaviours and appetitive traits ("eating correlates") and how socio-demographic and clinical characteristics associate with different phenotypes.A cross-sectional, multi-centre questionnaire was administered between November 2015 and March 2017 examining correlates of eating in children and adolescents attending weight-management programmes in Canada. Latent profile analysis was used to cluster participants based on seven eating correlate scores obtained from questionnaires. Analysis of variance (ANOVA) was used to determine phenotype differences on socio-demographic and clinical characteristics. Multinomial logistic regression models assessed relative risk of specific characteristics associating with a disordered eating phenotype.Participants were 247 children and adolescents (45.3% male, mean BMI z-score = 3.4 ± 1.0 kg/m) from six paediatric weight management centres in Canada. Seven eating correlates clustered into three distinct phenotypes: (1) loss of control eating, emotional eating, external eating, hyperphagia, impulsivity ("Mixed-Severe"; n = 42, 17%), (2) loss of control eating, emotional eating, external eating, hyperphagia ("Mixed-Moderate"; n = 138, 55.9%), and (3) impulsivity ("Impulsive"; n = 67; 27.1%). Social functioning scores and body esteem were significantly different across groups, with the Mixed-Severe participants having the poorest social functioning and lowest body esteem. Low body esteem indicated a greater risk of being in a multi-correlate group compared to the Impulsive group, while poor social function had a greater risk of clustering in the Mixed-Severe than Impulsive phenotype.Distinct eating phenotypes were found in treatment-seeking children and adolescents with obesity. Empirical evidence is needed, but these data suggest that tailored treatment approaches could be informed by these classifications to improve weight-management outcomes.