The objective is to estimate the prevalence of binge-eating disorder (BED) and subclinical BED in children and adolescents. Relevant articles were searched in the databases of PubMED and PsycINFO. Articles were included if they measured BED, subclinical BED, binge eating and/or loss of control (LOC) eating in samples of up to 20 years of age or with a mean age below 20 years. Subclinical BED covers participants with < 4 OBEs but ≥ 1 OBE pr. month and studies measuring subclinical DSM-IV/DSM-5 BED, but where all criteria were not met. All study types and measuring methods were accepted, but studies were excluded if they did not assess and exclude cases of recurrent compensatory behaviors. Meta-analyses were used to obtain an overall estimate of the prevalence of BED and subclinical BED, while stratified meta-analyses were used to assess sources of heterogeneity. 39 studies measuring BED, subclinical BED and/or a low frequency of binge eating were included. Two meta-analyses resulted in an overall estimated prevalence of 1.32% BED and 3.0% subclinical BED in children and adolescents. The results were influenced by high heterogeneity. Potential sources to heterogeneity in the BED result were weight of participants and sample types as well as level of risk of bias in the included studies. BED seems to be as frequent in children and adolescents as anorexia nervosa and bulimia nervosa. Hence, treatment of BED and BED symptoms in younger populations should be prioritized on the same terms as anorexia and bulimia nervosa.