The rising rate of obesity and the diminishing level of physical activity contribute to the increasing health problems among a growing proportion of the U.S. population (1). Research indicates that being overweight, physically inactive, or eating unhealthfully increases risk for certain cancers, particularly breast and colon cancers (2–7). In particular, people from minority racial or ethnic groups and other underserved populations suffer disproportionately from overweight and obesity: 70% of African American adults and 73% of Hispanic adults are overweight (8). In addition to experiencing higher rates of overweight and obesity than white Americans, African Americans, Hispanics, and other minority racial or ethnic groups are less likely to engage in healthful levels of physical activity (PA) (9) and more likely to experience a high incidence of some cancers and high related mortality rates (10). The high levels of insufficient activity and the persistent disparity in cancer rates underscore the need for better promotion of culturally appropriate strategies to increase levels of physical activity in order to prevent cancer.

The number of community programs to increase physical activity among U.S. residents is growing. Several recent intervention studies show how to plan and implement community interventions that promote physical activity among racially and ethnically diverse populations. The interventions tend to emphasize involving the target community early (i.e., during the planning stage), tailoring the physical activity programs to suit the predominant race or ethnicity of the community, and creating partnerships among diverse groups. These interventions also focus on the social determinants of health (e.g., levels of income, racial segregation, social support) and on needed structural and policy changes (11-15). However, these interventions do not always use evidence-based strategies for increasing PA.

The number of proven, evidence-based strategies to increase PA is small but expanding rapidly. Some strategies that work address the needs of individuals and the community as a whole. However, research findings do not show clearly how best to adapt and promote these strategies for use in various racial or ethnic communities. To promote evidence-based strategies for increasing PA effectively and to improve their use with people from various cultures, we need to understand what these various cultures see as the benefits of and barriers to engaging in PA. With such information, public health organizations and practitioners can create guidelines for implementing PA programs in ways that emphasize the benefits (as seen by the target community) and recommend ways of reducing or altogether avoiding the barriers of greatest concern to these communities. Tailoring evidence-based strategies to increase PA in this way could also aid in their dissemination.

To learn how best to adapt and disseminate evidence-based strategies to increase PA, the National Cancer Institute's (NCI's) Center to Reduce Cancer Health Disparities and NCI's Center for Strategic Dissemination funded four research projects in 2004 through its established Special Populations Networks (SPN) (16-19). The goal of these projects was to improve the dissemination of proven approaches to increasing PA among communities with high rates of health problems. Racially, ethnically, and geographically diverse SPN research teams conducted focus groups with members of three low-income communities: Redes en Acción worked with Hispanic women in Texas; the Asian American Network for Cancer Awareness, Research, and Training worked with Hmong parents and their children in California; and Deep South Network for Cancer Control worked with African American women and men in the Mississippi Delta. A fourth community consisted of Native Hawaiian college students who were not necessarily low income; the SPN project `Imi Hale conducted the focus groups and interviews with this group. In addition focus groups and interviews were conducted with gatekeepers (also known as key informants). These are the leaders and others who serve these communities and who can influence how community members behave.

This article discusses the background and framework for this four-site study and gives an overview of lessons learned and the common themes that emerged from all four communities. Detailed findings from each study site are available elsewhere (20-23). Here we highlight the similarities and differences among these study populations and discuss findings distilled from the research that may prove useful in adapting proven, evidence-based strategies to increase PA in order to control or prevent cancer.

Table 3 summarizes findings from the focus groups and interviews conducted at each site. An analysis of those findings follows:

Evidence-based strategies to promote PA frequently lack cultural relevance. The purpose of our study was to use a social marketing approach to find ways in which these strategies could be adapted to suit various cultures. Our findings confirm and refine what practitioners know to be true: that one size does not fit all and that tailoring proven approaches to suit each racial, ethnic, or geographic community is important to fostering behavior change. Our study also suggests that a social marketing approach to eliciting perceptions about the benefits of and barriers to participating in physical activities is viable and useful for adapting evidence-based strategies in ways that are embraced by members of specific communities. Multilevel, culturally adapted, social intervention approaches to adopting healthy lifestyles, including increasing PA, are important both for mainstream (30-32) and for underserved populations who deal with numerous barriers to active lifestyles (33-41).

Our research indicates that the strategies that appeal most to these four populations are those aimed at improving social support and increasing access to places for PA. Desirable components of physical activity programs are those that involve group activities (all sites) or partners (Latina women and African Americans), incorporate cultural components (Native Hawaiians and Hmong), are family-centric or encourage family and community support (all). Community members and gatekeepers also placed high value on increasing access to places for PA in their communities, particularly if these facilities are easy to get to, accommodate family and work schedules, and are available at low or no cost.

Participants in our study believed that it was important to be active and wanted their activities to be culturally relevant. Messages to encourage increased PA and associated activities need to be 1) tailored to the unique cultural and sociodemographic characteristics of each community and 2) sensitive to individual factors such as time and financial constraints. Ideally, activities should be group-focused, hands-on, purposeful, and inclusive. They should engage not only families and friends, but also extended families, church, and community — major social forces within the culture of these four communities. Messages to these groups should also stress that setting aside time to be active is important. Many participants assumed that the recommended amount of activity could be done through one's daily activities. It is also critical that messages emphasize 1) that the recommended levels of PA can be achieved through culturally relevant activities and 2) that everyone needs to purposefully set aside time for these activities.

On the basis of our findings, interventions would benefit from efforts to decrease physical or environmental barriers to places where people can be physically active. Interventions need to incorporate existing community facilities into their strategies, but these facilities may need to be modified so that they are easy to reach, safe, and available for various kinds of PA. Partnerships with community leaders to overcome social and physical barriers to PA (e.g., concerns about crime, lack of infrastructure, inadequate transportation, lack of access to places where people can be physically active) should be an integral part of any intervention (14). Community leaders, for example, can help recruit community and, possibly, state partners in health- and nonhealth-related divisions of local government (e.g., transportation, education, commerce) that can increase opportunities for PA within the community. Mobilizing relevant individuals and organizations (such as CHARPS in Mississippi and the Native Hawaiian elders recruited for our study) to advocate for outlets for PA is critical to sustaining increases in PA.

Interestingly, many participants' reported barriers to PA were not culturally specific; they are common among underserved populations of diverse heritage (38-40,42). Lack of time, lack of transportation, lack of economic resources to pay for programs and membership fees, literacy problems, unsafe neighborhoods, and lack of access to places to engage in PA were common to the Hmong, Latina, African American, and Native Hawaiian groups. These groups were also mainly low income, and with the exception of the Native Hawaiians, not well educated. Having a low income and a low level of education (regardless of race, ethnicity, or culture) poses major challenges to improving fitness (43). Policy changes at the community level are critical, since interventions that target the cultural, physical, social, and economic environment of a community will be more effective than interventions that target individuals in reducing the barriers to physical activity that many low-income populations face.

The use of social marketing methods in this study is consistent with the community-based participatory research (CBPR) approach. This approach involves communities in all phases of research that affects them and recognizes the community as an equal research partner (44). This collaborative process — involving both community members and leaders — is integral to CBPR and will help produce useful results for the participating communities and their partners.

During this study we learned the perceived benefits of, and barriers to, strategies to increase PA among groups whose perspectives were not previously explored. Results are not, however, statistically representative of a larger population because of nonrandom recruiting techniques, small sample sizes, and use of qualitative rather than quantitative research methods (45). Another weakness of this study is the limited discussion about point-of-place prompts (i.e., signs to encourage activities such as taking the stairs rather than the elevator). Other studies show that point-of-place prompts are effective in a Latino community (46) but not in a suburban African American community (47). Given these limited and mixed findings, future qualitative research on this strategy is warranted. Our study focused on the perceived benefits and barriers associated with proven PA strategies, which may have precluded our generating ideas for other strategies for increasing PA. However, the strategies we did discuss are evidence-based and have a high chance of succeeding. In future qualitative research, participants might first brainstorm to come up with strategies and then discuss in-depth any suggested strategies for which there is evidence to indicate they were successful in other communities.