This study assessed the impact of a youth-targeted multilevel nutrition intervention in Baltimore City. The study used a clustered randomized design in which 7 recreation centers and 21 corner stores received interventions and 7 additional recreation centers served as comparison. The 8-month intervention aimed to increase availability and selection of healthful foods through nutrition promotion and education using point-of purchase materials such as posters and flyers in stores and interactive sessions such as taste test and cooking demonstrations. Two hundred forty-two youth--caregiver dyads residing in low-income areas of Baltimore City recruited from recreation centers were surveyed at baseline using detailed instruments that contained questions about food-related psychosocial indicators (behavioral intentions, self-efficacy, outcome expectancies, and knowledge), healthful food purchasing and preparation methods, and anthropometric measures (height and weight). The Baltimore Healthy Eating Zones intervention was associated with reductions in youth body mass index percentile (p = 0.04). In subgroup analyses among overweight and obese girls, body mass index for age percentile decreased significantly in girls assigned to the intervention group (p = 0.03) and in girls with high exposure to the intervention (p = 0.013), as opposed to those in comparison or lower exposure groups. Intervention youth significantly improved food-related outcome expectancies (p = 0 .02) and knowledge (p < 0.001). The study results suggest that the Baltimore Healthy Eating Zones multilevel intervention had a modest impact in reducing overweight or obesity among already overweight low-income African American youth living in an environment where healthful foods are less available. Additional studies are needed to determine the relative impact of health communications and environmental interventions in this population, both alone and in combination.