PRACTICING ANTHROPOLOGY ON A COMMUNITY-BASED PUBLIC HEALTH COALITION: LESSONS FROM HEAL

Abstract

The author has served on the north Portland Healthy Eating Active Living Coalition since 2008. HEAL is a community-based health promotion program whose goal is to reduce and prevent childhood obesity in a predominantly Hispanic community. The author's role on the coalition is to conduct evaluation research and assist the coalition in identifying and addressing assets and barriers in the physical and social environment of this community so that children and their families have improved access to healthy food and recreation. Coalition work provides important opportunities for anthropologists to engage in community projects, and can help academic anthropologists develop collaborative research opportunities. Anthropologists can bring assets to such coalitions, including participatory research skills and familiarity with mediation. The time commitment and pace of community work can pose challenges for anthropologists who balance coalition work with other applied projects and academic positions. The methodology and focus of community-based work can also provide challenges for anthropologists, especially in publishing the findings and outcomes of their work. Finally, the participation of anthropologists on coalitions could be enhanced by additional graduate and postgraduate training in evaluation research, grant writing, and fiscal management.

A growing appreciation for the multiple pathways leading to obesity and associated chronic diseases has led to a move away from health promotion strategies focused solely on behavioral change and toward ecological approaches that engage communities in fostering environmental, policy, and behavioral changes. This general trend coincides with the increased popularity of, and funding for, interdisciplinary and participatory approaches to health promotion, creating new opportunities for anthropologists and other social scientists to play a role in community health promotion. A coalition framework is a good fit for such projects because partnerships require the coordination and mobilization of multiple stakeholders. Despite the challenges of funding and balancing community work with academic responsibilities, these are important opportunities for anthropologists, and anthropologists have much to contribute to coalition work.

BACKGROUND: THE HEALTHY EATING ACTIVE LIVING COALITION

In the past decade, the growing rate of overweight and obesity has become a national policy concern, garnering the attention of public health agencies, politicians, and private and public funders. Daily media reports express the problem with increasing urgency, and Michelle Obama's “Let's Move!” campaign has brought further attention to the issue. The health risks associated with obesity, including heart disease, hypertension, and type 2 diabetes, are well documented. Increasing rates among children and adolescents has been particularly significant. Based on National Health and Nutrition Examination Survey (NHANES) data, the Centers for Disease Control (CDC) reports that the prevalence of obesity among children and adolescents has risen from five percent in the period 1971–74 to 16.9 percent in the period 2007–08. The most recent data (2006–08) from the Behavioral Risk Factor Surveillance System (BRFSS) survey reveals significant disparities in obesity by race and ethnicity, with higher rates among African Americans and Hispanics than non-Hispanic whites (CDC 2011).

Oregon data also reflects disparities by race and ethnicity. Based on Body Mass Index (BMI), a majority of adult Oregonians are either overweight (BMI 25.0–29.9) or obese (BMI ≥ 30), but rates for Hispanics and other minorities are significantly higher. For example, a 2007 report indicates that 70 percent of Hispanic adults are overweight or obese, compared to 60 percent of non-Hispanic whites in Oregon (Nutrition Council of Oregon and the Oregon Coalition for Promoting Physical Activity 2007:22). The same report finds that Hispanics are less likely to meet CDC physical activity recommendations of at least 30 minutes of moderate exercise, three times per week (58 percent of Hispanics do not meet the guidelines, while 41 percent of non-Hispanic whites do not meet the guidelines). Hispanics in Oregon are also less likely to meet USDA recommendations of five or more servings of fruits and vegetables a day (21.8 percent of Hispanics meet the recommendation; 27 percent of non-Hispanic whites meet the guidelines).

Although U.S. Hispanics are more likely than whites to be obese, foreign-born Hispanics in the United States are less likely than whites to be obese (Wen and Maloney 2011). Although the reasons for this “Hispanic paradox” are debated, efforts to reduce and prevent obesity in this population are especially critical (see Franzini et al. 2001). In “new immigrant gateways” (Singer 2004) such as Oregon, where a large portion of the Hispanic population is foreign born, environmental factors are likely to cause rapid increases in obesity unless effective prevention strategies are identified and implemented.

Research on the high rate of obesity in low-income and ethnically diverse populations has increasingly focused on the influence of the built environment, especially on characteristics such as food access, walkability, green space, and school environments (Booth et al. 2005). Low-income families face a significant challenge acquiring a healthy diet in a food system where energy-dense, nutrient-poor foods are the least expensive, and nutrient-rich foods (whole grains and fresh produce) are the most expensive (Drewnowski 2004). Living in neighborhoods with little access to healthy food options and limited recreation likely compounds this effect (Powell et al. 2007).

Recognizing that obesity is an environmental issue, and that the rise in childhood obesity in particular has complex social, cultural, economic and environmental causes, the Multnomah County Health Department convened the HEAL coalition in 2006 to work with the Portsmouth neighborhood in north Portland to improve opportunities for healthy eating and physical activity in the community. The county's Chronic Disease Prevention Program embraced the socioecological model of health promotion (Robinson 2008) and was committed from the beginning to participatory methodologies. Both strategies made the participation of social scientists—including a sociologist, a geographer, and an anthropologist—a good fit for the coalition's work. Based on the socioecological model of health promotion, “the coalition's activities are not only intended to change individual behavior through knowledge and opportunities to learn new skills but also to implement strategies that influence community norms and lead to the adoption of policies that make it easier for the community at large to eat well and be physically active” (Multnomah County Health Department 2007).

The socioecological model recognizes five levels of influence on health—intrapersonal, interpersonal, organizational, community, and public policy (Robinson 2008). Ideally, then, health promotion programs include intervention strategies at all levels. The popularity of the socioecological model also reflects growing dissatisfaction with behavioral change strategies, and a shift away from strategies that address individual-level risk factors, toward “those that encompass environmentally-based as well as behaviorally focused strategies of health promotion” (Stokols 1996:282). Perhaps most importantly for understanding the potential contribution of anthropology to such programs, the socioecological model is a holistic and comprehensive approach to understanding and addressing opportunities and barriers to health in a community:

The health-promotive capacity of an environment is understood, not simply in terms of the health effects of separate environmental features (e.g., air quality, seismic safety, or social climate), but more broadly as the cumulative impact of multiple environmental conditions or occupants’ physical, emotional, and social well-being, over a specific time interval. [Stokols 1996:286]

In addition to research skills appropriate for community-level interventions, anthropologists bring an important understanding of the interaction of physical and social environmental conditions and can be strong participants in such an “inherently interdisciplinary” approach (Stokols 1996:286).

In practical terms, the socioecological model requires coordination and broad participation, making a coalition framework a likely strategy for implementation. The socioecological model typically proceeds from the identification of health-promoting and health-inhibiting environmental factors. Qualitative methods are often the best fit for identifying and describing these factors and their complex interactions. Once these factors are identified, it is critical to find the “leverage points,” or areas of highest impact for intervention (Stokols 1996:290). Applied anthropologists with training in participatory research methods are particularly well suited to play a strong role in both of these steps.

In identifying assets and barriers, and designing interventions, the HEAL Coalition employs the principles of community-based participatory research (CBPR). Like the socioecological model, CBPR addresses social and physical environmental factors that influence health. CBPR, however, is explicitly a partnership approach in which community members, organizational representatives, and researchers are actively involved. Because of its partnership approach, CBPR often involves coalitions like HEAL, and coalitions find the CBPR approach useful because of its integration of knowledge and action (Israel et al. 2003). Empowerment is a key theme in CBPR, as is attention to the power imbalance between researchers and community members (Israel et al. 1998:179). Although CBPR should include “the participation, influence and control by nonacademic researchers in the process of creating knowledge and change” (Israel et al. 2001), anthropologists can play a key role in such partnerships. Many applied anthropologists have experience with participatory research and with fostering the equitable participation of community members in research and action (e.g., Cornwall and Jewkes 1995; Perez 1997). Likewise, experience with diverse cultures and familiarity with evaluation research and a range of qualitative methodologies are also assets that applied anthropologists contribute to such coalitions. Below, I describe the work of the HEAL Coalition, followed by a discussion of some of the lessons from this case that are relevant to other anthropologists working with coalitions.

HEAL COALITION WORK: A SUMMARY

Founded in 2006, with a five-year grant from the Northwest Health Foundation, the HEAL coalition came together because of the Multnomah County Health Department's [MCHD] desire to reduce childhood obesity and to address health disparities that include an excess burden of disease for low-income and Hispanic residents. In addition to staff from MCHD's Chronic Disease Prevention Program, the coalition includes representatives from Kaiser Permanente Northwest, Portland Public School nutrition services, school staff and teachers, faculty members from Portland State University [PSU], and a team of parent leaders from Clarendon Elementary School (now Cesar Chavez K–8 School). The goal of HEAL has been to work with Hispanic families in north Portland to improve access to healthy eating and physical activity to reduce and prevent childhood obesity, and to improve the health of the whole community.

Early on, the coalition chose a school-based approach to building relationships in the Portsmouth neighborhood. School communities are a logical focus for obesity prevention: children spend significant time at school; schools provide one–two meals per day, and schools’ educational setting provides an opportunity for nutrition education (Budd and Volpe 2006; Foster et al. 2008). Additionally, schools have the potential to serve as change agents in disadvantaged neighborhoods (Stone et al. 1999). As central places in neighborhoods, schools bring families together, both facilitating social ties and social support, and linking families to larger systems of support and resources (city, state, and federal). Through schools, parents can also gain experience in interacting with public agencies, build capacities for collective action, and increase skills. This can be particularly significant when working with foreign-born parents who are less familiar with U.S. systems.

HEAL work began with a focus on coalition building and parent outreach at two north Portland schools. A parent liaison working for Portland Public Schools, and a MCHD community health worker, both bilingual Hispanics, helped recruit parent volunteers and develop relationships with school staff. HEAL also helped organize walk and bike to school events, a weekly runners club in two schools, nutrition education classes for students and parents, and exercise classes for parents in the evenings. These activities have become well-integrated into the overall school program and are now sustained without the active involvement of the coalition, especially at Cesar Chavez School, where the principal is a strong supporter of physical and nutrition education (Dworkin 2010).

Although successful in introducing behavioral change programs in the school, HEAL is also committed to spearheading changes in the policy and physical environment. With funding from Metro's North Portland Neighborhood Enhancement grant program, my colleague Meg Merrick, coordinator of the Community Geography Project at PSU, and her students organized the north Portland PhotoVoice project in 2008. PhotoVoice is a participatory methodology in which community members use cameras to tell their story, identify concerns, and advocate for change (Wang 1999). Eight parent researchers from two schools used digital cameras to document the perceived barriers to health in their daily lives. Although facilitated by PSU faculty and students, parents were the researchers for the project, taking pictures, developing the findings, and presenting their concerns to policymakers at Metro, the regional government for the Portland metropolitan area.

Through the PhotoVoice project and subsequent community meetings, the coalition identified three priority areas for HEAL efforts: improving access to healthy food in the neighborhood, addressing health and safety impacts of the closure of Clarendon Elementary School, and improving the school meal program. With funding from PSU, the coalition agreed to conduct an interview-based evaluation of program activities, with the goal of exploring parent and school staff perspectives on these priority areas. A PSU graduate student and I conducted semistructured interviews with several HEAL parents, teachers, a school nurse, and a school principal in the fall of 2008. Through PhotoVoice, we learned that although north Portland has several large grocery stores, Hispanic parents find them expensive and lacking in culturally appropriate foods, and therefore travel long distances to shop in discount supermarkets. We describe this phenomenon as a “food mirage,” rather than a food desert, and argue that the potential impact on health is similar—managing the challenges of time, distance and cost means infrequent shopping trips and less fresh produce.

Although admittedly less participatory in nature, the interview study was facilitated by the rapport already created between parent volunteers and other HEAL stakeholders. This may have mitigated the distance between researchers and subjects, and enhanced the sense that participants in the study were being consulted, rather than simply “tapped” for information. Interview participants understood that their input was valued and would be used to plan future HEAL activities, and not simply used to produce academic research.

In 2010, following the interview study, participants at community meetings explored ways to improve food access in the community. Through discussion, the coalition decided to focus its efforts on the creation of a Healthy Hispanic Corner Store Network in north Portland. Healthy corner store campaigns have sprouted around the country as a strategy to improve the food environment of low-income communities (see Food Trust n.d.; Healthy Corner Store Network n.d.). Recognizing that many communities lack access to full-service grocery stores, and that many low-income and minority families rely on local convenience and corner stores for food purchases, governmental and nongovernmental groups around the country are working with existing retailers to improve access to healthy, affordable foods, including fresh produce, low-fat dairy items, and staple grains. North Portland is home to a number of small, Hispanic-owned stores that cater to Hispanic families in the area. Looking to corner store projects in cities like Philadelphia and Washington, D.C. as models, the HEAL coalition decided to approach Hispanic-owned stores in north Portland and ask them to participate in this project.

In the spring of 2010, a community meeting was held with HEAL coalition members and the owners of four north Portland stores. Community health workers from the county health department presented the idea of a healthy corner store network, using examples from similar projects around the country. I shared the findings of the interview study, and parent volunteers described the challenges of food access in north Portland and what they would like to see in Hispanic-owned stores. Store owners described some of the barriers they faced, including lack of refrigeration, air conditioning, and store space, in providing healthier and more diverse options in their stores. All four owners agreed to participate in the network and were enthusiastic about the possibility of receiving assistance in improving their businesses. The county had recently been awarded a large grant from the CDC, including funding for a countywide healthy retail initiative, so the Healthy Hispanic Corner Store Network could both serve as a pilot project for that initiative, and receive some financial support from the grant. (See Figure 1.)

Figure 1. Timeline of corner store network development.

The first step in the corner store initiative was to gather data about existing conditions in the participating stores. I designed a structured store observation tool—using similar tools developed by coalitions in Washington, D.C., and New Orleans as models—to gather data on things like the current availability of fresh produce, healthy snacks, staple goods, and low-fat dairy in the stores (Custer 2009; Washington D.C. Healthy Corner Store Program 2008). The tool also elicited information about the store environment, ability to accept SNAP or WIC purchases, and access to public transportation. The observation form was translated into Spanish, and eight parent volunteers were recruited and trained to conduct the observations in teams of two at each of the four stores. For parents to play a meaningful role in discussions and decisions emerging from the store observations, it was important for them to participate as researchers. Also, we felt that Hispanic store owners would be more comfortable with parents (many of whom they knew as customers) conducting the observations, as opposed to either representatives of county government or academic researchers. There were also a few open-ended questions on the observation tool, which yielded unexpected responses from parents, such as the desire for fresh fish in the stores. This participatory methodology was not without challenges, however, as we found a lack of consistency across the teams in recording such things as prices and weights or quantities of items. In fact, a PSU graduate student and a community health worker did follow-up observations of the stores to recheck the data collected. (See Figures 2–3.)

Figure 2. Sample section from store observation form.

Figure 3. Sample store observation findings, presented in Spanish to store owners and parents.

Following the structured observations, I created a structured interview guide to collect information from store owners about their business. The guide included questions about customers, store inventory, barriers that store owners face, and what they would most like assistance with. The guide was translated into Spanish, and interviews were conducted by a community health worker with MCHD. The interviews revealed that all owners would like assistance with equipment, especially refrigeration units, and with identifying local suppliers of fresh produce. In March 2011, the coalition convened a meeting with parents and store owners to discuss the findings of both the structured observation and the interviews, and to launch the healthy corner store initiative. The county has created bilingual signage that stores can use to identify themselves as participants in the healthy retail initiative, and to identify healthy choice options on shelves. In the coming months, the coalition will work with store owners to seek grant funds for refrigeration and other equipment. We are currently working with the store owners to create a buyers’ network to help them purchase local produce at wholesale prices, and have connected them with a cooperative of Hispanic farmers in the region that can provide culturally desirable items such as chili peppers and tomatillos.

LESSONS LEARNED

Achieving Meaningful Community Participation

Many of the lessons and challenges of the HEAL coalition's work stem from the commitment to involve community members, especially Hispanic parents, in every stage of the process, from the identification of research questions, to research design, data collection, analysis and action. We have been most successful in achieving meaningful participation in areas of data collection, analysis, and action, and somewhat less successful in the areas of research design. That is, parent leaders and other community partners have often played a key role in collecting data, discussing findings at community meetings, and determining priorities for action. Community members typically have played a smaller role in the initial stages of research, especially in developing research tools and in study design. This pattern, however, may be somewhat justified. That is, it is not necessary, or even desirable, for all members to play the same role on the coalition, so long as everyone has a role that is significant. As Kaye notes, “Grassroots leaders and members have had the experience of being ‘tokens’ on coalitions rather than having roles with real power and substance” (2001:274). The focus of research is often driven by a study of what has worked in other cities, by recognition of what funding agencies will support, and by an understanding of the strengths and weaknesses of various research methodologies, such as whether a structured observation or a survey instrument will be most useful in a particular phase of the program. Academics and public health officials can provide this kind of technical assistance, without eliminating opportunities for community members to be meaningful partners in other phases of the work (Wolff 2001b). Participation has worked best when academics on the coalition have designed the research tools, with parents and other coalition members collecting the data. Although the identification of research questions admittedly begins with the priorities of the funders and the county health department, input from community members at coalition meetings plays a significant role in determining the specific issues to be explored.

This strategy means that researchers must resist drawing conclusions or identifying priorities that should come from the community, a difficult challenge for academics who are used to a more traditional cycle of research design, data collection, and analysis. It is vital, however, to respect community process and to act in a facilitating role. In assisting with research design and developing research tools, it is also important that the data and methods must be driven by community needs, and always connected to action. If the data to be collected does not help determine immediate steps for action, it is probably not worth collecting, or as one evaluation expert recently advised, “if it's nice to know, it's gotta go.” There is a discipline to this approach that can be challenging for academics, even those with experience in applied research. This focus can also jeopardize publication opportunities for academics, because such action-oriented research designs are not always viewed as scientifically valid (see Viswanathan et al. 2004 for a discussion of the limitations of CBPR).

The participation of a bilingual community health worker and a bilingual parent liaison working at one of the public schools was key to reducing the barriers to active parent participation. Particularly in the early years of the coalition, these members invested much time and energy into developing relationships and trust in the community, which has been the basis for ongoing participation and genuine engagement (Uyeda et al. 2009). They have built strong relationships with families and store owners in the community, and send invitations in Spanish, frequently following up with face-to-face invitations and phone calls. Also important, all community meetings are held in Spanish, and childcare and food are provided. Meeting times, typically in the evenings, are selected to accommodate the parents’ schedules, and this helped bring both men and women from the school community to meetings. One challenge, not unusual for this type of coalition, has been trying to expand participation beyond this core group of parent leaders (see Wolff 2001a). As we move toward broader policy and environmental change efforts, it will be important to reach a larger population. The healthy retail initiative, for example, will require marketing efforts that reach not only Hispanics but also other families in the area.

The more community members are involved in the process and contribute their time and energy, the more they will expect to see concrete results, increasing the risks associated with raised expectations. This is compounded by the fact that the iterative and community-based approach of such coalitions means that the process is usually time consuming and progress is slow. Store owners hoping to see concrete benefits, for example, may become discouraged when months pass without program implementation. Parents who contribute limited free time to the coalition can likewise become discouraged by modest successes and slow change. There is also the risk of burnout and loss of volunteers, so the desire for meaningful participation must be balanced with reasonable expectations of community-members’ time and energy. Coalition members must also be able to see progress and feel that they personally are benefiting from the coalition's efforts (Wolff 2001b).

Community–University Partnerships

Additional lessons emerge from the opportunities and challenges of community–university partnerships. Academic and community partners on such coalitions have different needs, expectations, and goals. As noted above, when data collection is driven by community needs and an action approach, publication goals must be secondary (see Strand et al. 2003). The long-term commitment required of such partnerships means that such projects do not fit easily in the promotion and tenure timelines. There are opportunities for researchers to publish about the outcomes of partnerships and their role. However, such activities are not always valued as strongly as more traditional research, especially in promotion and tenure decisions (Israel et al. 2001). Action research explicitly rejects the objectivity of the researcher. Although this is a powerful model for promoting positive change in marginalized communities, it is not always equally valued or considered rigorous by other academics (see Israel et al. 1998; Small 1995). With data collection driven by decision-making needs and project development, and because of the heavy reliance on qualitative methods, it is often difficult to prove intervention success (Israel et al. 1998).

For such partnerships to be successful and ongoing, they must be mutually beneficial to all parties, meaning that there should be a balance between community needs and university requirements (Calleson et al. 2005; Nyden and Wiewel 1992). Universities can and should be more inclusive of their definition of “scholarly activity.” Although dissemination of findings to our academic peers is very important, other kinds of publications, presentations, and outreach should also be considered a legitimate part of a scholarly portfolio (Israel et al. 2001). Likewise, universities and departments should expand their view of “reputable journals” to include those that publish the results of participatory research (Israel et al. 2001:191). Nyden and Wiewel (1992:50) argue that “collaborative research that seeks solutions to social problems and negotiates the difficult line between what the discipline needs and what a community needs, should be put on an equal footing with exclusively discipline-oriented research.”

There are, however, data collection strategies that can maximize the benefits both to the coalition, and opportunities to publish findings in academic venues. In designing a community-based research project, it is important to think about not only what data you need to make immediate decisions but also how to tell your coalition's story effectively as the process unfolds. This is important for reporting to funders, for sharing your model with other coalitions, for seeking community support, and for eventual publication in academic venues. It is important to plan ahead for what information and documentation to collect and at what points. For example, in the corner store project, store observations and photographs taken at the beginning of the project will help guide intervention strategies, but will also allow us to document change later on. We may never be able to document that obesity rates have declined in the neighborhood (because the project goal was to improve food availability and access), much less that store improvements are responsible, but we can show that the characteristics of the stores themselves have changed, and qualitatively describe changes in the neighborhood food environment. We have also learned that photography can be a powerful tool in telling that story.

Data Collection

Israel and colleagues argue that in CBPR, “methodological flexibility is essential,” meaning that methods used should fit the purpose of the research and the needs of the community involved (2001:189). The ability to use a wide variety of methods also works well. One method may identify broad priorities, while another can be used to collect detailed information needed to plan a specific intervention. The HEAL coalition, for example, has used PhotoVoice, asset mapping, interviews, structured observations, and surveys. In our coalition, a clear division of labor has generally worked well: the anthropologist designs data collection tools, a bicultural community health worker recruits parent participants and store owners, community members collect data, and the entire coalition participates in discussing the findings and agreeing on next steps.

Given the explicit rejection of objectivity, and the blurring of the lines between researcher and subject, such research projects do not always fit easily within the guidelines of university institutional review boards (IRB). Because community partners do not typically have their own review processes, IRBs often serve as the only ethical review process available. Although IRBs should exercise some flexibility in such cases, researchers also have a duty not to assume that the IRB is the only step in assuring ethical compliance. Rather, community partners should be consulted in the research design and in any possibly sensitive issues, even before the IRB review. This means that IRB should allow preliminary contact with study participants before the research project begins, and should understand that community partners are not traditional “subjects” (see Shore et al. 2010).

Funding

As with any community–university partnership, funding to sustain the work of the coalition is an ongoing challenge, which is also reflected in the literature on community-based health promotion (Johnson et al. 2009:71; Wolff 2001b). Funding mechanisms still prioritize projects that “use traditional research designs which emphasize researcher control and randomized control designs” (Israel et al. 2001:188). Israel and colleagues argued a decade ago that calls for proposals needed to better incorporate the principles of CBPR, and that grant reviewers needed to have experience in participatory methods and not just in the relevant content area. The last decade has indeed seen an increase in funding opportunities for community-based projects, especially from private foundations like the Robert Wood Johnson Foundation and the Northwest Health Foundation, which has provided funding for HEAL work. The National Institutes of Health has also increased funding for community-based research, such as a recent RFP for “Community-Based Partnerships for Childhood Obesity Prevention and Control: Research to Inform Policy” (PA–09–140).

Although funding opportunities have expanded, “Expectations of funders have been unrealistic in terms of the time and resources needed to accomplish CBPR projects” (Krieger et al. 2002:378). Moreover, funding does not always compensate every partner involved, or does not provide adequate compensation for all partners. This is especially the case when funding is available to support either the interventions themselves, or the research needed to plan the intervention, but not both. This can leave either the academic researchers or the community partners without adequate funding to sustain the partnership. Without adequate support, academic partners have to fit research activities and community meetings into an already full schedule of teaching and research, causing further delays and slowing progress. Although my own university strongly supports community-engaged research, pressure to increase total research funding makes participation in small-budget projects increasingly difficult.

Presentation of Findings

Anthropologists and other academic partners can play a key role in telling the coalition's story, and sharing coalition successes with a broader audience including the media, policymakers, funders, and academic peers. It is also important that anthropologists have an opportunity to publish about the participation in community-based research to contribute to their own professional development. Nonetheless, community members and coalition partners can and should be included as coauthors and copresenters whenever possible and appropriate, reflecting the shared nature of the research process and the findings.

An additional lesson involves the unique sensitivities that arise from publishing about the process of community-based work. Inevitably, different perspectives, agendas, and opinions emerge during the work of coalitions. Publishing findings that are potentially sensitive for one or more coalition members can damage relationships necessary for problem solving in the future. For example, a publication that included parents’ detailed concerns about school meals, findings from our interview study, jeopardized our relationship with the public school's nutrition program (Everett et al. 2009). Although the parents’ concerns were valid, and represented a key priority emerging from the community, reporting those findings at a time when the public schools were trying to be responsive to the issues raised was not in the best interest of the coalition, and demonstrates the difficulties of balancing the needs of all parties. Consulting with members of the nutrition program before publication would have been a better way to balance the needs of all stakeholders.

In addition to publications, there are many other ways that coalitions tell their story. Local newspapers reported on HEAL efforts (Odom 2009). Media coverage can be an important way for coalition members to receive recognition for their efforts, and to inform a larger population about coalition goals. Several members of HEAL also participated in a series of workshops hosted by the Northwest Health Foundation. The workshops included training on conducting evaluation research and on using findings to tell the coalition's story. We also had the opportunity to interact with other coalitions around the state of Oregon and to learn from each other's experiences. At the end of the workshop, we produced a one page bulletin to report on HEAL's progress in the healthy corner store initiative. The bulletin was distributed to coalition members, policymakers, and officials, as well as funders. It also provided the coalition with the opportunity to assess our progress by identifying key milestones of our project. (See Figure 1.) Exchanging our assessment tools and progress bulletins with other coalitions also helped us learn from each other and develop models for future work.

Graduate and Postgraduate Training

Graduate anthropology training already includes many strengths that serve us well in participating in coalition work, including competence in working with diverse communities. Programs with strengths in applied anthropology additionally develop skills in the integration of research and practice. The ability to provide students with practicum experiences in their communities is an increasingly vital piece of graduate education, especially for students in master's programs. Beyond practical experience and applied curriculum, however, students will benefit from developing specific skills that will be assets for coalitions. Language training is essential for those interested in working with immigrant populations, an area of increasing importance and opportunity for anthropologists, especially in “new immigrant gateways” like Oregon that have rapidly increasing numbers of foreign-born residents. Experience and training in grant writing is another skill that will provide graduates with opportunities in government agencies and NGOs, and will be critical to successfully funding coalition work. Graduate and postgraduate training in evaluation research, community-based participatory research, survey administration, and fiscal management are also important for developing a successful “tool kit.”

CONCLUSIONS

Anthropologists are an increasingly familiar presence on public health coalitions, and the opportunities for anthropologists to work on public health promotion will likely expand as the popularity of community-based approaches continues to grow. This article has identified a number of lessons, based on my experience working with the HEAL coalition, which may be relevant to anthropologists working on similar projects.

Fostering meaningful participation should be one of the most central contributions anthropologists can make to coalition work. A commitment to meaningful participation entails a long and slow process, but enhances the likelihood of sustainable changes and fosters community empowerment. Meaningful participation requires frequent community meetings, a willingness to take time to train volunteers in data collection, and discipline in avoiding drawing conclusions or making recommendations without community consultation. It is possible, however, for academics to play a greater role in some phases of the process, most notably in designing research instruments, while still maintaining meaningful participation. Although participation is critical, it is also advisable to avoid overburdening community members, and to look for both short and long-term gains to encourage sustained participation.

Coalition research must be driven by community needs and should be focused on driving action steps and reporting results. Nonetheless, coalitions can be strategic about anticipating data needs that will help the coalition tell its story to decision makers and facilitate eventual publication in academic venues. Universities likewise need to expand their understanding of “scholarly activity” to include coalition work, and to encourage faculty to communicate the results of this work in a wide range of venues. Untenured faculty need to be cautious about creating a scholarly agenda based only on partnership work. Despite the value of this work, academic partners have little control over the pace and direction of these endeavors (and funded research may be a priority), making this a risky strategy for promotion and tenure.

Anthropologists should consider a mixed methods approach to community-based research, and choose methods that best fit the needs of the coalition. This is another key contribution that anthropologists can make to coalition work, especially given the range of qualitative methods that most applied anthropologists are familiar with. Given the time challenges and need to put research into action, smaller sample sizes are generally necessary, although this may make publication in academic venues more challenging. Graduate and postgraduate training should include more opportunities to develop skills in grant writing, survey research, and fiscal management.

Our most pressing health issues have complex social causes and will require environmental, policy, and behavioral changes. Anthropologists and other social scientists have a key role to play in understanding the social determinants of disease, and in tackling these challenges alongside communities. Despite the challenges, these are important opportunities for applied anthropologists and for the discipline.

NOTE

Acknowledgments

Acknowledgments.  The author gratefully acknowledges Ben Escalante and the Multnomah County Health Department's Community Wellness and Prevention Program for the opportunity to work on this project. The author also thanks Scott Ellis for his assistance on data collection. A Faculty Enhancement Grant from Portland State University provided support for the author's participation in the coalition. The author especially thanks the members of the HEAL coalition. This article reflects the views of the author alone, and she is solely responsible for its contents.