Obesity has reached epidemic proportions in the United States (U.S.), with one out of every four adults having a body mass index (BMI), ≥ 30 [1, 2]. Individuals who are obese are more likely to develop chronic diseases [3–6], and as of 2009, the annual cost of obesity in the U.S. was approximately 300 billion dollars [7]. Obesity rates vary considerably across individual states [8] and counties [9, 10]. Such geographic variation raises questions about the influence of local, community, and regional factors on obesity.

Multiple studies describe efforts to address obesity: for example for individuals [11–15], schools [16, 17], worksites [18, 19], and the built environment and community norms [20, 21]. However, these programs and initiatives to think beyond the individual level have focused more on organizations [22, 23], with less available literature on county-level experiences. Considerable variation exists not only in the types of obesity intervention programs implemented, but also in measures of their effectiveness, making it difficult to know whether and how such programs work relative to each other, and which may work best for a given population. One strategy for identifying effective health interventions is to study members of a group that are performing better than expected. Such better-than-expected performers, called positive deviants, may exhibit characteristics or approaches that could also be effective when generalized to the larger population [24]. This approach has been used to address hand hygiene compliance [25], diabetes management [26], timeliness of cardiology care [27], hospital readmission [28], and adolescent pregnancies [29]. To our knowledge, however, no study has examined county-level positive deviants for adult obesity rates.

Accordingly, our research aim was to identify positive deviant counties for obesity and using semi-structured in-depth interviews to determine any community level factors that were present across all counties that might give insights into how the lower rates of obesity were achieved by these counties. Findings from this inquiry may be useful to other localities as they consider strategies for reducing obesity rates. We identified a set of six counties with particularly low obesity rates despite socioeconomic profiles that would predict higher obesity rates and conducted in-depth, qualitative interviews with 80 individuals across the six counties. Our findings may suggest avenues for interventions to counter the ongoing challenge of obesity in the U.S.

In six U.S. counties that are positive deviants for adult obesity rates, participants consistently described their efforts to develop a nuanced understanding of the features of their communities, while also recognizing that obesity was a complex challenge. They used this strategically gleaned knowledge to establish a countywide “backbone” structure that capitalized on strengths in their communities to combat obesity and improve healthy living. Although specific programs and initiatives differed across counties, participants consistently noted the importance of breaking down silos and building partnerships, fostering community connections, and transferring ownership to community members.

Study participants emphasized the importance of thinking broadly and holistically not only about the challenge of obesity but also about their approaches to address it. They were strategic in their planning and organization and deliberate in gathering information to develop tailored, countywide efforts to improve health and combat obesity. Although the group members who contributed to the backbone structure from across the county sometimes had different, individual goals, they came together with a common focus. Additionally, these individuals described an approach that reflected a systems-level view, thinking of themselves as contributing to an organizational structure that reaches across an entire complex system, pulling together different components (sometimes even those that are quite disparate from each other) in ways that introduced communications structures and capitalized on feedback loops, in service of the health of the whole [37–39].

Within the United States, obesity has often been approached as a medical problem to be handled at the individual level, often within the doctor-patient relationship. Obesity as a diagnosis was incorporated into the International Statistical Classification of Diseases and Related Health Problems (ICD) 10 [40]. Subsequently, many strategies to combat this epidemic propose an individual approach that is focused on losing body fat through diet or exercise or a combination of the two [15, 41]. Efforts to think beyond the individual level have focused more on organizations, such as schools or workplaces [22, 23], but little literature exists on countering obesity at the county level.

Findings from this study suggest that community-level, non-medical factors influence obesity and, further, that community-level strategies can be employed to counter this epidemic in the U.S. The strategies to address obesity described by study participants do not fit a medicalized, individual-level, disease-specific approach but instead focus on tapping into the unique strengths of a given community environment (e.g. natural resources, or an active retiree community). The community and government leaders we interviewed also emphasized the potential of tapping into and weaving together existing resources in a community, essentially helping to improve individual outcomes by focusing on social and community influences that surround and influence a given individual. Findings from this study suggest that successful strategies to counter obesity might need to extend beyond individual medical care and capitalize on resources in the community.

Our study has a variety of policy implications. Identifying and tapping into existing resources across a community may be a valuable strategy to combat the expensive challenge of obesity, especially if accompanied by investment in community-wide structures to convene, energize, and support strategic partnerships. Our findings suggest that for interventions to be sustainable, community members need to take initiative and ownership. However, health departments must have sufficient resources to play the key role of providing a backbone structure across the community.

Our study results should be interpreted in light of several limitations. First, we have a small sample of six counties. However, upon completing analyses of data from these counties, we reached theoretical saturation, and no new themes emerged. Second, as with any qualitative study, there is a risk of participant self-reporting bias; however, we used a variety of strategies as part of the interview process including open-ended questions and voluntary recording of the interview in order to promote an environment to encourage participants to speak candidly. Additionally, we conducted interviews with approximately 12 government and community leaders per county. It is possible that these participants might have had different perceptions of how their counties addressed the challenges of obesity than would other community members or participants. However, since this study is, to our knowledge, the first to apply the positive deviant approach to adult obesity rates at the county level, we chose government and community leaders as a crucial first group to interview. Finally, consistent with the positive deviant methodology [42–45], we focus on counties with the phenomenon of interest: relatively low obesity rates despite being located in a state with high obesity and having socioeconomic risk factors for high obesity. Such studies typically do not include comparison groups as those counties do not have the experience that is the focus of the inquiry. Nevertheless, as an extension of this exploratory, descriptive study, future investigations using a broader sample of counties with higher and lower rates of obesity and quantitative measurement of processes and structures would be helpful for testing the hypotheses generated from the present study.

Obesity rates have steadily increased across the U.S. Our research suggests that positive deviant counties may provide a useful tool to learn about strategies and approaches that may be successful in fighting complex epidemics. Additionally, given the constraints from government resources including financial and human capital, communities can capitalize on members to promote and disseminate health messages and programs. Although fighting the challenge of obesity at a county-level is a daunting task; however, it can be aided by a focused approach that involves knowing the community and the complexity of health problems and incorporating that knowledge into the strategic plan.