Supported exercise improves controlled eating and weight through its effects on psychosocial factors: extending a systematic research program toward treatment development.

BACKGROUND Behavioral weight-loss treatments have been overwhelmingly unsuccessful. Many inadequately address both behavioral theory and extant research--especially in regard to the lack of viability of simply educating individuals on improved eating and exercise behaviors. OBJECTIVE The aim was to synthesize research on associations of changes in exercise behaviors, psychosocial factors, eating behaviors, and weight; and then conduct further direct testing to inform the development of an improved treatment approach. METHODS A systematic program of health behavior-change research based on social cognitive theory, and extensions of that theory applied to exercise and weight loss, was first reviewed. Then, to extend this research toward treatment development and application, a field-based study of obese adults was conducted. Treatments incorporated a consistent component of cognitive-behaviorally supported exercise during 26 weeks that was paired with either standard nutrition education (n = 183) or cognitive-behavioral methods for controlled eating that emphasized self-regulatory methods such as goal setting and caloric tracking, cognitive restructuring, and eating cue awareness (n = 247). RESULTS Both treatment conditions were associated with improved self-efficacy, self-regulation, mood, exercise, fruit and vegetable consumption, weight, and waist circumference; with improvements in self-regulation for eating, fruit and vegetable consumption, weight, and waist circumference significantly greater in the cognitive-behavioral nutrition condition. Changes in exercise- and eating-related self-efficacy and self-regulation were associated with changes in exercise and eating (R(2) = 0.40 and 0.17, respectively), with mood change increasing the explanatory power to R(2) = 0.43 and 0.20. Improved self-efficacy and self-regulation for exercise carried over to self-efficacy and self-regulation for controlled eating (β= 0.53 and 0.68, respectively). CONCLUSIONS Development and longitudinal testing of a new and different approach to behavioral treatment for sustained weight loss that emphasizes exercise program-induced psychosocial changes preceding the facilitation of improved eating and weight loss should be guided by our present research.