Understanding Childhood Obesity in America

Linkages Between Household Income, Community Resources, and Children's Behaviors

Taylor F. Eagle, BS; Anne Sheetz,MPH; Roopa Gurm, MS; Alan C. Woodward,MD; Eva Kline-Rogers,MS, RN, NP; Robert Leibowitz, PhD; Jean DuRussel-Weston, RN,MPH, CHES; LaVaughn Palma-Davis, MA; Susan Aaronson,MA, RD; Catherine M. Fitzgerald, MA, RD; Lindsey R. Mitchell, MPH; Bruce Rogers, BS; Patricia Bruenger, BA, CCRC; Katherine A. Skala,MPH, CHES; Caren Goldberg, MD; Elizabeth A. Jackson,MD, MPH; Steven R. Erickson, PharmD; Kim A. Eagle, MD


Am Heart J. 2012;163(5):836-843. 

In This Article


The fight against childhood obesity in America is a daunting one. At first glance, the idea of simply encouraging parents and children to eat more healthily or to make healthy dietary choices and exercise more regularly seems logical and doable. However, a deeper examination into this national epidemic leads one to appreciate that the underpinnings to childhood obesity are remarkably complex, and the solutions are likely to be equally if not more complex.

Increasingly, it is clear that a child's health status involves a complex interplay between individual factors, social factors, environmental factors, and a child's ultimate selection (within their range of choices) of nutrient consumption and energy expenditure.[8,15] Individual factors include genetic predisposition. Socioeconomic factors include the built environment such as activity space, transportation, access to healthful food, recreational parks and programs, and many other items.[16–17]

In this study, we have attempted to answer 2 fundamental questions about childhood obesity. First, using data from a statewide BMI screening of children through Massachusetts, we have shown the association between low family income and higher percentage of overweight or obese youth. Second, using standardized baseline questionnaire data from the Project Healthy Schools initiative in 4 Michigan communities, we have clarified the baseline differences in childhood nutritional and activity patterns that vary across communities and are likely to be heavily influencing the prevalence of childhood obesity. In a nutshell, children attending schools in communities with lower mean annual income report behaviors of nutritional intake and physical activity, which support a greater imbalance between energy intake and expenditure, and this promotes weight gain.

Previously, we reported differences in nutritional intake, activity, and their corresponding influences on childhood obesity among middle schoolers in 2 adjacent Michigan communities of differing socioeconomic means.[13] We postulated that access to fast foods, poorer access to fresh fruits and vegetables, and poor access to recreational parks and both recreational and school-based exercise programs were very likely to play an important role in promoting childhood obesity in communities with lower mean family income.

In a subsequent report, we identified 3 independent correlates of childhood obesity among sixth grade students: less frequent vigorous exercise, more daily time spent watching television, and steady consumption of school lunches—all 3 behaviors being more prevalent in communities with poor average household income.[11]

Our findings are consistent with other reports linking lower household income with higher rates of childhood obesity. Ogden et al[18] showed nearly a doubling in childhood obesity, from 11.9% to 21.1%, across differing levels of household income. This effect was seen in both boys and girls, but the relationship was less consistent in non-Hispanic blacks and Mexican Americans. Wang et al[19] showed a similar relationship using the NHANES database. Among adolescents, those residing in the lowest tertile of per capita household income had the highest rates of overweight obese. In a study from Australia, children aged 7 to 15 years residing in the highest household income quintile had the lowest rate of overweight/obese, as seen in our study of Michigan communities.[20] Gable and Lutz[21] also showed that a relationship exists between lower family income and higher rates of childhood obesity.

What then explains the higher frequency of childhood obesity in lower income communities? First, access to fast food restaurants and relative poor access to stores selling fruits and vegetables are at the top of the list. Studies in Seattle, New Orleans, and Massachusetts all support a strong correlation between greater obesity rates in lower income neighborhoods with a high density of fast-food restaurants.[22–24] Fast food typically provides more energy-dense food at a low price when compared with other alternatives. Fast-food corporations fully appreciate this "opportunity," and for example, in East Los Angeles, 63% of fast-food restaurants are within walking distance of a school.[15,25] In New Orleans, obesity increases in direct proportion to density of fast-food restaurants and convenience stores and inversely with density of supermarkets.[25]

Second, studies suggest a distinct inverse relationship between availability of recreational parks and programs and average community household income.[16,26–30] Low-income neighborhoods and schools generally have fewer immediate physical and programmatic opportunities for regular interval physical activity.[31–33] Increased TV time and reduced weekly physical exercise are the result. Our studies and others suggest that these effects directly influence the development of childhood obesity.[34–38]

Third, the association between household income and childhood obesity also relates to parents.[39] Single-parent households, on average, have lower income. Similarly, educational level of parent(s) directly relate to emphasis on healthy behaviors in and outside the home. Two-parent homes, with highly educated parents, are more likely to stress better nutrition and physical activity than single-parent homes where the parent has a lower level of education.[40,41]

What are the implications of our research? Where are the opportunities to stem the rising tide of childhood obesity? We believe that our response needs to be multidimensional, sustained, and vigorous. First, by focusing on children's belief and attitudes, we have the opportunity to create for each child a yearning for a healthier lifestyle. In Project Healthy Schools, we have shown that through a series of educational activities, by improving school-based nutritional and beverage opportunities, and by encouraging frequent exercise, we can improve a child's understanding of how their choices affect their health.[12] We have also seen an improvement in their self-reported behaviors and physiologic markers of health such as cholesterol and triglyceride levels as a result.[12,13] Second, based on this research and that of others, we need to critically engage multiple community leaders, businesses, nonprofits, and municipal/county/state agencies in the process of understanding and improving the built environment. Until healthy food choices are available to children and until safe and accessible facilities and programs that promote regular physical exercise are made available, it is hard to imagine that intense education alone will have much impact on childhood health. We certainly must engage the parents wherever we can, given their critical role in this problem.

In this regard, the community efforts that have been implemented in 1 Massachusetts community deserve comment (see Figure 1). Its public schools are strong supporters of wellness throughout the system. They offer physical education in schools from kindergarten through grade 12. The physical education programs offer diverse, age-appropriate physical activities and incorporate wellness topics including good nutrition in their curriculum. The high school has a fitness room available to students and staff before/after school and during the day for students/staff during free periods. Middle schools offer before and after school programs emphasizing the 5–2-1–0 concept: 5 fruits/vegetables per day, <2 hours of screen time per day, 1 hour of exercise per day, and 0 high sugar–containing drinks per day. Elementary schools also offer before or after school physical activities including walking clubs, yoga, and others. Nonfood birthday celebrations are encouraged, too. Food services encourage healthy choices in school lunch and beverage offerings. Regular soda and candy are not available to students. Whole grain and whole wheat breads are the rule, not the exception. At the community level, bike paths are available and maintained, organized bike safety is provided by the police department, many students still walk to school, and community-wide weight watcher and yoga programs are available.

It is clear, based on examples such as the one described above, that community-wide action can have a dramatic impact on childhood health. In addition, the degree to which local, regional, state, and national legislation supports healthier and affordable choices should hasten the community transformation that is needed. Similarly, the critical involvement of parents in both demonstrating and encouraging a healthy lifestyle cannot be overstated.

Childhood obesity is our nation's number one health issue. Its solution is not easy. Each community offers a unique challenge, opportunity, and set of resources. By working together, sharing ideas and resources, and through directed legislation, the time is ripe to meet this challenge head on.


Our analysis focused entirely on the association between community's average household incomes, the prevalence of overweight/obese among middle schoolers and how socioeconomic status may directly impact childhood behaviors. We were not able to further define the interaction between both race and gender with prevalence of overweight/obese. In addition, studies like that of Rundle et al show that low-income but walkable neighborhoods have less overweight/obese adults than in low-income nonwalkable neighborhoods.[42] We were not able to assess this community characteristic in our study. Finally, the analysis of childhood behaviors reflected data on just 999 sixth graders. Although we believe that the sample is representative of the communities involved, the confidence limits around the identified associations are broad.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: