The GEMS effort was to create a social environment that encouraged healthy eating and exercise, and expanded health literacy in a fun manner, and was accessible to young girls. It is easier to change health-related behaviors in the young, and the program tried to address the unique and often more acute problem of obesity in African-American young girls. The entire community and family units were incorporated into the program effort. Positive aspects of the African-American community, such as strong social support, were used by the study designers, also in line with social cognition theory. Existing support structures and social learning were combined: for example, the families in question were often not educated in how to properly read food labels, but once they were, the desire to help their daughters become healthier would hopefully reinforce the need to engage in proactive steps to improve dietary health. During Family Nights, families of the girls ate a low-calorie meal together; received suggestions about meal planning and shopping to make healthy eating affordable; and participated in collective physical activities. They were also encouraged to set realistic fitness goals for the family, just as the girls set fitness goals for themselves. To ensure compliance, when a child or family did not attend a scheduled event, a 'sorry we missed you' card was sent to the address. This also reinforced social modeling and social cognition theory, to show how the effort to get fit and maintain a healthy body weight was a family and community-based effort, even though the focus of the group was on young girls.

As a placebo, another intervention program was offered to the same population group, emphasizing self-esteem building and cultural enrichment. The control group involved "arts and crafts, self-esteem activities, creating memory books, and a workshop on African percussion instruments" (Story 2003, p.S1-57). In both experimental and control groups, all girls were assessed during an 8-week pre-intervention period, and follow-up visits took place within 2-weeks after the 12-week program. Waist, height, and waist circumference measurements were taken and girls were subject to wearing a calorie monitor for three days,...

Girls and their parents were also subjected to a self-reported questionnaire about activity habits, eating patterns, and health knowledge.
However, surprisingly, the main difference between the intervention and the control groups was that the experimental group had more health knowledge. Intervention group girls had a lower caloric intake and a lower percent of calories derived from saturated fat, and were less apt to drink soda, but they ate fewer fruit and vegetables. BMI and physical activity levels of both groups remained the same, and the intervention group girls reported higher levels of bodily self-consciousness and negative attitudes about their weight (Story 2003, p.S1-59).

Several factors can explain the results. While the study's authors point to the small, short duration of the study, and the fact that some of the heaviest girls were in the intervention group, larger social issues may be at play. Changing the personal, immediate social and familial environment of the girls does not impact larger, macro issues such as cultural preferences for certain foods, the lower cost and ease of preparation of processed foods, long-term availability of exercise facilities (as opposed to a relatively short and concentrated group of fitness sessions), and a lack of a hopeful attitude to the future, which can also affect eating habits. Interventions that change the community environment, and not merely the attitudes of a specific demographic 'macro' group or family, may have a more long-standing impact upon fighting obesity in African-American communities.