BACKGROUND: Although several studies have documented the existence of psychopathology in obese adolescents, disagreement remains regarding the extent and nature of this psychopathology. The aim of the present study was to explore the type and frequency of psychopathology in a clinical as well as a non-clinical sample of obese adolescents, and in a normal weight control group. METHODS: The study sample consisted of a clinical study group of 30 obese adolescents, a non-clinical obese group of 30 obese adolescents, and a control group of 30 normal weight adolescents. Psychological assessment was performed using a non-structured psychiatric interview, the Child Behavior Checklist (CBCL), Children Depression Inventory (CDI), Rosenberg Self-esteem scale (SES) and the Eating Attitude Test (EAT). RESULTS: More than half of the clinical obese adolescents (16/30) had a DSM-IV diagnosis, often involving major depressive disorder (n = 10). The mean scores of anxiety-depression, social problems, social withdrawal and total problem in the CBCL scale of the clinical obese group were significantly higher than the non-clinical obese group and the normal weight control group. The mean total scores of the SES and the CDI of the clinical obese group were higher than the normal weight control group. The mean total score of EAT of the clinical obese group was significantly higher than the normal weight control group, and the mean score of EAT of the non-clinical obese group was significantly higher than the normal weight control group. CONCLUSIONS: The results support previously published reports which show a higher ratio of psychopathology (depression, behavioral problems, low-esteem) among clinical obese adolescents than among non-clinical obese adolescents. Findings provided evidence for a psychosocial at-risk population in a subgroup of obese adolescents.