Obligatory Exercise and Eating Attitudes—A Pilot Study on Polish Adolescents

Background: This pilot study examined differences in levels of obligatory exercise between two groups of adolescents: psychiatric patients diagnosed with eating disorders and youth engaged in qualified sports. Methods: The study included 45 adolescent psychiatric patients with eating disorders and 45 students training in sports departments (track and field, judo, gymnastics), aged 14–20 years. The sample comprised 50 women, 28 men, 10 non-binary participants, and two individuals who declined to identify their gender. The following instruments were used: the Eating Attitudes Test (EAT-26), Exercise Dependence Questionnaire (EDQ), Obligatory Exercise Questionnaire (OEQ), and Inventory of Physical Activity Objectives (IPAO). Between-group differences were analyzed using the Mann–Whitney U test. Most scales demonstrated acceptable internal consistency (Cronbach’s α > 0.70), although several scales showed lower reliability estimates. Results: Adolescents diagnosed with eating disorders showed significantly higher levels of problematic eating behaviours (rg = 0.94), exercise fixation (rg = 0.71), exercise commitment (rg = 0.33), withdrawal symptoms (rg = 0.52), exercise for weight control (rg = 0.83), and insight into problems (rg = 0.77) than athletes. In contrast, athletes reported higher levels of exercise for social (rg = 0.74) and health-related reasons (rg = 0.40). Effect sizes ranged from moderate to very large. These findings indicate marked differences in the psychological motives and compulsive characteristics of exercise between adolescents with eating disorders and athletes. Conclusions: This pilot study suggests that adolescents with eating disorders differ from athletes primarily in the psychological characteristics and motives underlying exercise rather than in exercise volume alone. These findings may help inform the assessment of problematic exercise in clinical practice, although they require confirmation in larger, adequately controlled studies.