Obese children are at an increased risk of becoming obese adults, with an elevated risk of morbidity and mortality in comparison to non-obese children. The development of powerful interventions that modify childhood obesity and maintain these effects throughout development may decrease the prevalence of adult obesity, and improve the health of these children as adults. We have made initial progress in developing effective treatments for obese children, and using these methods have shown significant maintenance of treatment effects over ten year intervals, with the average child being followed from ages 10-20. In these studies we have identified two factors that appear important in treatment, targeting both parents and children and use of exercise programs. In addition, we have documented a variety of retrospective factors that were related to long term follow-up, including dietary fat, self-monitoring weight, parent and friend support for eating and exercise. The proposed research is designed to initiate a new randomized, controlled study to evaluate the effect of problem solving interventions for childhood obesity treatment. Problem solving interventions have been successfully used to treat a variety of child and adult problems, and the initial research on obesity has been promising. The proposed study randomizes families with obese children and parents to groups in which both the parent and child receive problem solving training, the child alone receives problem solving training, or a control group in which neither parent nor child receive problem solving training. Families will be provided a standardized six month intervention, with follow-up meetings at 12, 24 and 36 months. Problem solving will be assessed using laboratory and at home measurements. Family support for eating and exercise change and intake and expenditure measures will also be collected.