This study presents a 4.6-year follow-up of children treated for obesity. The primary goal of the treatment program was to prevent further weight gain. In setting this modest peal, we wanted to avoid the type of dietary restraint that has been linked to the development of eating disorders. A cognitive-behavior modification (CBM) program was designed to help the child change his or her lifestyle, to enhance self-regulation skills and specific problem-solving skills in different eating situations. Three active CBM treatment interventions were compared (group therapy, individual therapy, and summer camp) to an "advice in one session." At pretest, the mean age was ii years (SD = 2.5). Mean percentage overweight was 558 (SD = 21); all subjects were at least 20% overweight. Data from 109 children were available at the 4.6-year follow-up. representing 80.1% of the original sample. At follow-up, 72 subjects (71.6%) showed no further increase in percentage overweight. This does not mean that these obese youngsters had become thin. In this study 188 of the obese children were no longer obese (<20% overweight) and 29% of the subjects had become moderately obese (20% to 40% overweight). The obese youngsters showed a mean reduction of their overweight of - 11%; the mean overweight was still 42%. Eating behavior was evaluated in a subgroup of the obese children (n = 53) and a stabilization of abnormal eating behavior was noted within the subscales of the Dutch Eating Behavior Questionnaire. Using the Eating Disorder Inventory, 5 adolescents (9%) had an at risk score on the bulimia subscale. Taking into account the pessimistic prognosis for obese adults, the trends found here are more optimistic, at least for half of the obese children.