You are seeing this message because your web browser does not support basic web standards. Find out more about why this message is appearing and what you can do to make your experience on this site better.

The Harvard Family Research Project separated from the Harvard Graduate School of Education to become the Global Family Research Project as of January 1, 2017. It is no longer affiliated with Harvard University.

Terms of Use ▼

Program Description

Overview Girlfriends for KEEPS (Keys to Eating, Exercising, Playing, and Sharing) is an obesity prevention program for low-income African American girls in Minneapolis, Minnesota. Its physical activity intervention goals are for girls to (a) increase frequency of participation in sustained, moderate-to-vigorous intensity activities, (b) decrease time spent in sedentary activities, and (c) experience feelings of enjoyment, physical competence, and self-confidence in performing a range of physical activities. The dietary change intervention goals are for girls to (a) decrease consumption of high-fat foods, (b) increase consumption of fruits and vegetables, (c) decrease consumption of sweetened beverages, and (d) adopt healthy weight-related eating practices (e.g., portion-size awareness, eating only when hungry, etc.).
Start Date 2000
Scope local
Type after school
Location urban
Setting public school
Participants elementary school students (ages 8–10)
Number of Sites/Grantees 3
Number Served 26
Components The program was developed by the Girls Health Enrichment Multi-Site Studies (GEMS), a National Heart, Lung, and Blood Institute-sponsored multi-center research program. Girls participate in the program for 12 weeks. Intervention meetings are held twice a week for 1 hour after school at each site. The intervention is taught by trained African American GEMS staff. Staff training focuses on the need for, and purpose of, the intervention, and includes modeling and active rehearsal of many of the activities. Club meetings consist of fun, culturally appropriate, interactive, hands-on activities, emphasizing skill building and practice of the particular health behavior message for that week. A healthful snack, sometimes prepared by the girls, and water are offered at each meeting. Messages include information about the benefits of drinking water more often than soda, eating more fruits and vegetables, drinking low-fat milk, selecting low-fat foods for snacks, eating smaller portions of snacks, choosing smaller-sized and lower-fat entrees in fast food restaurants, increasing physical activity, watching less television, and enhancing self-esteem.

Another major component of the intervention is increasing physical activity levels with a variety and choice of activities, such as dancing (ethnic, hip hop, aerobic), double-dutch jump rope, relay races, active games, tag, and step aerobics.

To keep girls' interest and participation, incentives are built into the program for attendance, setting short-term goals, and completing activities. These include attendance beads that make a bracelet when put together at the end of the intervention, water bottles, pedometers, jump ropes, and t-shirts. Transportation home was provided by the schools’ regular buses.

The intervention's messages are reinforced by family activities, including weekly family packets sent to parents, family night events, motivational telephone calls by staff to parents to encourage them and to check their progress on meeting family goals around healthy eating and activity behaviors, and organized neighborhood walks.
Funding Level Not available
Funding Sources National Heart, Lung, and Blood Institute, National Institutes of Health, United States Department of Health and Human Services


Overview The evaluation examines program effects on participating girls, and also includes a process evaluation to monitor program implementation, help explain the outcomes, and provide meaningful data to help refine the program.
Evaluators Mary Story, Nancy E. Sherwood, John H. Himes, Marsha Davis, David R. Jacobs, Jr., Yolanda Cartwright, Mary Smyth, and James Rochon, University of Minnesota, Minneapolis and the Biostatistics Center, George Washington University
Evaluations Profiled An After-School Obesity Prevention Program for African-American Girls: The Minnesota GEMS Pilot Study
Evaluations Planned Not available
Report Availability Story, M., Sherwood, N. E., Himes, J. H., Davis, M., Jacobs, Jr., D. R., Cartwritght, Y., et al. (2003). An after-school obesity prevention program for African-American girls: The Minnesota GEMS Pilot Study [Supplement 1]. Ethnicity & Disease, 13(1), 54–64.


Evaluation Mary Story, Ph.D.
University of Minnesota
Division of Epidemiology
1300 South Second Street, Suite 300
Minneapolis, MN 55454
Tel: 612-624-8801
Fax: 612-624-9328
Program Mary Story, Ph.D.
University of Minnesota
Division of Epidemiology
1300 South Second Street, Suite 300
Minneapolis, MN 55454
Tel: 612-624-8801
Fax: 612-624-9328
Profile Updated May 3, 2005

Evaluation: An After-School Obesity Prevention Program for African-American Girls: The Minnesota GEMS Pilot Study

Evaluation Description

Evaluation Purpose To examine both the program's process and effectiveness in achieving its intended outcomes, including participant decreases in body mass index (BMI), improvements in diet, increases in physical activity, and improvements in psycho-social measures.
Evaluation Design Experimental and Non-Experimental: To assess program outcomes, 54 potential program participants were recruited from the three schools serving as intervention sites based on the following criteria: (a) being an 8- to 10-year-old African American girl, (b) having a BMI greater than or equal to the 25th percentile for age and sex, (c) being able to participate in physical education classes at school, (d) being fluent in English and having a primary caregiver fluent in English, and (e) not having been held back more than one grade in school. All girls completed baseline measures, and were then randomly assigned to either a treatment group (n = 26), who participated in the Girlfriends for KEEPS program, or a control group (n = 28), who participated in a non-nutrition/non-physical activity program focused on promoting positive self-esteem and cultural enrichment (participants attended three monthly Saturday morning meetings, which included arts and crafts, self-esteem activities, creating memory books, and a workshop on African percussion instruments). Most data were collected at a baseline clinic visit (pretest) and at a 12-week follow-up clinic visit (posttest) following the end of the program year. Retention among the 54 girls was 98%, as only one girl did not return for the 12-week follow-up visit. All analyses adjusted for pretest values of the outcomes of interest.

For all 54 girls at baseline the average age was approximately 9 years. About 80% were pubertal, though all were pre-menarcheal. The mean BMI of the girls was 20.7 kg/m2, with a mean percent body fat of 31%. Approximately 83% of parents were African American, 6% were biracial, and 11% were Caucasian. The majority of households were low-income, with 54% of parents reporting annual incomes of less than $30,000. Approximately 44% of homes were female-headed households. The average BMI for parents was 32.8 kg/m2, with the majority (92%) of parents qualifying as overweight (BMI = 25–29.9 kg/m2) or obese (BMI = 30 kg/m2).

There were no significant between-group differences for parental baseline variables. For youth baseline variables, the treatment group had a significantly higher proportion of girls with high cholesterol (p = .04) and were significantly heavier (p = .04). Treatment group youth, however, were also marginally significantly (p = .08) taller than control girls, which meant that there were no significant pretest differences between the two groups in BMI.

Implementation/process related data were also collected from program staff, parents, and participating youth, and through site observations.
Data Collection Methods Document Review: Checklists completed by Girlfriends for KEEPS staff were collected, and included measures of program attendance and whether the day's activities were completed.

Interviews/Focus Groups: Focus groups were conducted with parents in the treatment groups to study program implementation and participants' experiences in each program

Observation: Each of the 24 sessions at the three Girlfriends for KEEPS sites was observed by to study program implementation. Observers also documented attendance at family events.

Surveys/Questionnaires: Girls in both the treatment and control groups and their parents completed surveys at both pretest and posttest. These surveys measured demographic characteristics (age and race of girls and parents, parent education, total household income, household composition, and home ownership), as well as a number of assessments and outcomes.

In addition, dietary intake was assessed using questionnaires asking girls in both groups to recall what they ate in the past 24 hours both at baseline and at follow-up. Measures from the dietary intake recalls included total energy intake (kcal/day), percent of energy derived from fat, and number of servings per day of fruit, juice, vegetables, water, and sweetened beverages.

Post-intervention evaluation surveys were also administered to parents and girls in order to understand participants' perceptions of the intervention. In addition, parents completed evaluation forms for the intervention's family events.

Test/Assessments: Assessments collected during clinic visits included BMI, which was computed using measurements taken of each girl's weight, height, and waist circumference; parental height and weight; percent body fat; sexual maturation (assessed through direct observation of breast and pubic hair development by centrally trained female staff using five standard stages of pubertal development); insulin, glucose, and lipid levels (assessed through an overnight fasting blood sample drawn from the girls); physical activity (assessed using a device called the Computer Science Application (CSA) that girls are instructed to wear continuously for 3 days, and which measures physical activity levels).

Assessments collected through the youth surveys included the following: Healthy Choice Behavioral Intentions (a 12-item measure assessing behavioral intentions for choosing healthy food; Stevens et al., 1999), Self-Efficacy for Healthy Eating (a 9-item self-efficacy measure, including items such as “How hard would it be for you to…” eat more of particular foods and less of others), Diet Knowledge (a 6-item measure assessing knowledge of healthy diet practices; Stevens et al., 1999), Fruit and Vegetable Snack Accessibility (a 2-item measure assessing fruit and vegetable snack accessibility in the home), Parent Encouragement for Healthy Eating (a 5-item measure assessing ways that parents encourage healthy eating), Physical Activity Self-Concept (a 9-item measure assessing physical performance self-concept; Harter, 1982), Physical Activity Preference (a 37-item measure assessing preferences for physical versus sedentary activities), Physical Activity Outcome Expectancies (a 17-item measure assessing girls' positive expectancies for engaging in physical activity), Self-Efficacy for Physical Activity (a 9-item measure of self-efficacy for physical activity), Physical Activity Home Environment (a 5-item measure assessing home environmental factors related to physical activity, e.g. “It is safe to play outside near where I live”), Body Satisfaction (a measure adapted from participants' reactions to eight body sizes that measured the degree of discrepancy between which picture looks most like oneself and which looks the way they would like to look; Stunkard, Sorenson, & Schulsinger, 1983), and Weight Control Behaviors (two scales assessing moderately unhealthy weight control behaviors and unhealthy weight control behaviors, derived from the elementary school version of the McKnight Risk Factors Survey; Shisslak et al., 1999).

The youth survey also included the GEMS Activity Questionnaire (Treuth et al., 2004), which evaluates both previous-day and usual activities. It includes a checklist of 28 activities typically performed by African American girls, along with pictures of activities. For each activity, girls check off whether they had engaged in it yesterday, the duration of the activity, whether they “usually” engage in it, and the frequency of engagement.

Assessments collected through the parent surveys included the following: Availability of Lower-Fat and Higher-Fat Foods (a 29-item measure divided into two subscales measuring the home availability of higher-fat foods, and lower-fat alternatives), Low-Fat Food Practices (a 25-item measure assessing the frequency of preparing and serving lower-fat foods in the home, adapted from Kristal's Food Habit Behavior Scale; Kristal, Shattuck, Henry, & Fowler, 1990), Motivation for Healthy Eating (a 5-item measure assessing motivation for health eating, e.g., “How interested are you in drinking less regular soda pop?”), Self-Efficacy for Healthy Food Preparation (a 10-item measure assessing self-efficacy for healthy food preparation, e.g., “How hard would it be for you to have fresh fruit on the kitchen counter, or somewhere your daughter could easily see it?”), Food Availability (a 31-item measure divided into three subscales measuring vegetable, fruit, and sweetened beverage availability in the home), Parental Dietary Intake (a measure estimating the percentage of parental energy intake derived from fat, and a measure estimating intakes of both fruit and vegetables; Thompson et al., 1998; Thompson et al., 2000), Motivation for Physical Activity (a 2-item measure assessing parents' interest in spending more time being physically active in general and with their daughters), Self-Efficacy for Physical Activity With Daugther (a 5-item measure assessing parents' self-efficacy in engaging in physical activity with their daughters, e.g., “How hard would it be for you to get your daughter to be physically active instead of watching TV?”), Parental Support of Daughters' Activity Levels (a 6-item measure assessing parents' level of support for their daughters' activity levels, e.g., “I try to get my daughter to play outside when the weather is nice”), and TV Watching (a 4-item measure assessing parents' reports of their daughter's TV watching on weekends and weekdays).

Harter, S. (1982). The perceived competence scale for children. Child Development, 53, 87–97.

Kristal, A., Shattuck, A., Henry, H., & Fowler, A. (1990). Rapid assessment of dietary intake of fat, fiber, and saturated fat: Validity of an instrument suitable for community intervention research and nutritional surveillance. American Journal of Health Promotion, 4, 288–295.

Shisslak, C. M., Renger, R., Sharpe, T., Crago, M., McKnight, K. M., Gray, N., et al. (1999). Development and evaluation of the McKnight Risk Factor Survey for assessing potential risk and protective factors for disordered eating in preadolescent and adolescent girls. International Journal of Eating Disorders, 25, 195–214.

Stevens, J., Cornell, C. E., Story, M., French, S. A., Levin, S., Becenti, A., et al. (1999). Development of a questionnaire to assess knowledge, attitudes, and behaviors in American Indian children. American Journal of Clinical Nutrition, 69(Suppl.), 773S–781S.

Stunkard, A., Sorenson, T., & Schulsinger, F. (1983). Use of the Danish Adoption Register for the study of obesity and thinness. In Kety, S., Rowland, L., Sidman, R., & Matthysse, S. (Eds.), Genetics of neurological and psychiatric disorders. New York: Raven Press.

Thompson, F. E., Kipnis, V., Subar, A. F., Schatzkin, A., Potischman, N., Kahle, L., et al. (1998). Performance of a short instrument to estimate usual dietary intake of percent calories from fat. Bethesda, MD: National Cancer Institute, Information Management Services, Inc.

Thompson, F. E., Subar, A. F., Radimer, K., Smith, A. F., Midthune, D., Rosenfeld, S., et al. (2000). Performance of two cognitively enhanced fruit and vegetable short assessment forms. Cleveland, OH: National Cancer Institute, Cleveland State University.

Treuth, M. S., Sherwood, N. E., Baranowski, T., Butte, N. F., Jacobs, D. R., Jr., McClanahan, B., et al. (2004). Physical activity self-report and accelerometry measures from the girls health enrichment multi-site studies. Preventive Medicine, 38(Suppl.), 43–49.
Data Collection Timeframe Not available

Formative/Process Findings

Parent/Community Involvement Family nights in the treatment program were well attended, with 88% of parents and 95% of girls attending at least one family night. Over 80% of the parents and 90% of the girls reported enjoying the family nights.

Program staff completed motivational telephone calls with 86% of treatment group parents.

Over 80% of parents indicated that they had bought some of the low-fat snack foods that the program had sent home with the girls.

The organized neighborhood health hikes were not well attended by participants' families, though the evaluators note that this may have been partially due to inclement weather conditions.
Recruitment/Participation On average, treatment group girls attended 21 of the 24 sessions.
Satisfaction The majority of treatment group girls (92%) reported liking the program “a lot.”

The majority of treatment group parents reported that both they (96%) and their daughters (88%) were very satisfied with the program. All parents reported that they would recommend it to other parents.

Summative/Outcome Findings

Family At the 12-week follow-up, parents of treatment girls reported significantly less availability of higher-fat foods in their homes (p = .001), more low-fat food practices (p = .009), and lower energy intake from fat in their own diets (p = .03), compared to parents of control girls. No significant between-group differences were observed for the other parent-reported diet and activity measures, although most measures changed in the direction expected for the treatment group parents.

About half of parents reported that the family nights helped “a lot” in making changes in eating or activity for their family, while 38% said the motivational telephone calls helped “a lot” in making changes in eating or activity.
Prevention There were no significant differences in BMI at the 12-week follow-up between the treatment and control groups.

There was a trend for waist circumference to be 1.4 cm higher in the treatment, compared to the control, group (p = .08) at posttest.

For dietary intake measures at posttest, treatment group girls had lower caloric intake, lower percent of calories derived from fat, and more servings of water per day compared to control group girls. Treatment group girls also, however, had lower fruit and vegetable servings per day, and higher sweetened beverage servings per day than control group girls. None of these differences were statistically significant.

Physical activity measures demonstrated consistently greater activity levels in the treatment group compared to the control group at posttest. CSA counts per minute, minutes of moderate to vigorous physical activity between 12pm and 6pm, and self-report of usual physical activity all increased more among girls in the treatment group, though none of these differences reached statistical significance.

At posttest, treatment girls reported significantly higher scores on the healthy choice behavioral intentions (p = .001), diet knowledge (p = .001), and preferences for physical activity (p = .04) scales than did control girls. Treatment girls were also significantly more likely than control girls to report a preference for larger body size (p =.01), and were more likely to report engaging in both moderate (p = .004), and unhealthy behaviors related to weight concern (p = .04). No between-group differences in the prevalence of dieting were observed.


© 2016 Presidents and Fellows of Harvard College
Published by Harvard Family Research Project