LESSONS FROM THE AMERICAN CANCER SOCIETY DISPARITIES RESEARCH PORTFOLIO: STRATEGIES TO MOBILIZE COMMUNITY ACTION
Clearly, the research summarized in this volume offers very important approaches for changing individual health behavior, supporting better decisions, enhancing quality of care, and promoting quality of life. These are the kinds of strategies that must we must bring to scale. However, taking these interventions to scale is largely dependent on the availability of resources and, hence, the political will that our health care system must be rededicated and redesigned to eliminate cancer disparities.
As such, the most important lessons we can draw from these examples involve ways to raise public awareness and build a national consensus for broad-based social policies to eliminate disparities. The studies in this volume demonstrate the kinds of strategies necessary to mobilize and guide a nationwide grassroots campaign to address cancer health disparities in a way that is comprehensive and sustainable. Reading through this volume study after study, it becomes apparent that researchers have arrived at powerful new techniques for activating communities and transforming delivery systems. What remained in writing this closing topic was simply to recognize and label this potential. The lessons learned fall into eight distinct areas, outlined below:
Lesson 1: Acknowledging individuals’ multiple identities, and addressing stigma and discrimination, is empowering. Perhaps the most fundamental cause of health disparities in our society is that we view access to care and healthy living conditions as a privilege rather than as a human right. Public acceptance of inequitable access to health care is tied into politically motivated images and stereotypes that serve to justify disenfranchisement and inequitable treatment (Ryan, 1976). The prevailing social construction of ethnic, sexual, or religious minority groups often includes explanation and justification for why they are to blame for their poor health, unwilling to take care of themselves, or otherwise unworthy of society’s benefits. These stigmatizing attitudes and beliefs are amenable to change by promoting greater familiarity and providing countervailing examples.
Several studies in this volume help bring to the fore the concerns and the potential of groups that are often ignored or demeaned. For example, Christopher and colleagues discuss the history of their work in partnership with the Apsaalooke community. Research of this sort would be impossible without validation of the communities’ concerns and ownership of the research process. However, many underserved communities are not geographically centered or as highly organized as the Apsaalooke. Levels of stigma and disorganization can impede evaluation of needs and tailoring of programs. Boehmer and Bowen discuss how sexual minority women’s social networks and relationships can influence their health behavior. This work offers a rich and nuanced perspective of a social group that has often been subject to stigma and ignorance in the health care system and beyond. These examples show how social science research can help to organize segments of the community, and help give voice and legitimacy to their concerns.
No group in this country suffers from worse health disparities than African American men. However, it is overly simplistic to presume that one-size-fits-all solutions are appropriate for these men. For instance, Griffith and colleagues discuss efforts to tailor interventions according to men’s multiple identities, to determine competing priorities that interfere with health behavior as well as potential avenues for change. It is noteworthy that settings that may be suitable for reaching some segments of the community (such as workplaces or fraternal organizations) would entirely omit other segments.
Regardless of the specific health behavioral targets or intervention strategies, studies such as these emphasize the efforts of marginalized groups to pursue better health and quality of life. They also highlight some of the historical and social factors that have impeded success. Individual projects such as these serve to express the particular struggles and challenges faced by each group. Taken together, these studies help to put a human face on cancer health disparities statistics. Examples such as these must be communicated to the broader public, to contribute to a greater sense of commonality with disenfranchised groups. It becomes much harder to deny access to health and quality of life to people who are seen as "us" rather than "them."
Lesson 2: Raising public awareness of disparities is a vital tool for mobilizing community support. The introductory topics of this volume present overwhelming evidence concerning the extent, nature, and consequences of cancer health disparities. These data are absolutely vital for analyzing the extent and nature of disparities. However, what is not clear is whether and to what extent individuals who are subject to these disparities are aware that they exist. Although there have been decades-long campaigns focused on health risk behaviors, there is virtually no widespread or sustained attempt to raise consumer awareness of disparities in cancer care. The next generation of research to reduce cancer health disparities must include efforts to build public awareness of health disparities, including risk for receiving inaccurate information and substandard care.
Of course, there have been arguments against releasing information grading physician or hospital outcomes to the general public. The concern is that ratings do not take into account the variation of cases across providers and settings, and that the public is not sophisticated enough to understand this information. As several studies in this volume demonstrate, these problems may be readily addressed. For example, Latini and colleagues show how attention to health literacy helped men to understand choices related to prostate cancer, while Makoul and colleagues discuss approaches to communicating information about colonoscopy. We have the tools that we need to help many patients understand quantitative concepts such as relative risk and likelihood of benefit. These same concepts can be applied to helping consumers evaluate information about their local health care system. Providing consumers with question lists and other tools can also help them to determine whether that they know what to ask their providers and that they are getting more complete information. Research related to health literacy must be broadened to help individuals and families become more effective and better informed health care consumers.
In addition to supporting health literacy interventions to inform consumers, lay health advisors (LHAs) and community health advocates could be trained to help patients understand their choices in care. In the work of Erwin and colleagues ,LHAs employed to encourage individual behavior change are in a great position to also help consumers obtain and interpret information about the treatment facilities and providers available in their communities. Just as we train LHAs to access and provide accurate information about cancer treatment, screening, nutrition, and other topics, we can ensure that they provide appropriate information about local health choices. We need to develop support systems in communities that can provide guidance and support for active health care consumers.
Community organizers and change agents for social justice would necessarily share many of the same characteristics that we look for in community LHAs. People in this role must have the skills to share their own stories and elicit these stories from others. They must understand the tacit concerns and assumptions that members of their communities would bring to discussions of health inequities, to anticipate and address these issues. They must be able to establish the trust necessary to encourage community members to take action at the individual level (e.g., insisting on appropriate time and attention from providers) or at the collective level (e.g., participating in meetings to share information about disparities that they have experienced).
Lesson 3: Health care delivery and research must join together to create learning systems. Patient navigation also provides a powerful tool for addressing system sources of health disparities. In programs such as the one described by Hendren and colleagues,navigators do more than address individual motivation and lack of knowledge. Through their work, navigators necessarily encounter problems in health systems that impede access to screening and care. Information from navigation could be used to determine where to employ quality improvement strategies to reduce disparities, and to evaluate the efficacy of these systemic strategies. Ideally, navigation should not only link individuals to services one at a time; it should contribute to systemic solutions that make navigation less necessary. Future studies of patient navigation must be conducted in conjunction with systems-level interventions to reduce health disparities.
The interventions in this volume are already providing a safety net for cancer patients who may be subject to fragmented or poorly coordinated care. It is noteworthy that the Sisters Informing SistersSM intervention by Sheppard and colleagues is helping to ensure that women receive adjuvant treatment that ought to be the standard of care. Although it is necessary to assist these women, it is also important to address the reasons why they are falling through the cracks. Data from projects like Sisters Informing Sisters™ shed light on systemic problems, and can contribute to systemic solutions by identifying the reasons for gaps in care. Individual-level interventions to ensure that patients receive a widely accepted standard of care are only necessary because the health delivery system is failing to do its job. Future individual-level interventions to foster access to care must also include systems research to identify and close gaps in care.
Adams and Chien’s discussion of the use of linked datasets to examine the impact of policy demonstrates another important lesson. Datasets must be available to track the impact of policy changes on health disparities. What is perhaps most remarkable about Adams’ work is the lengths that she needed to go to link datasets so that she could examine the impact of major changes in public health policy on system uptake and performance. For purposes of accountability, stakeholders and researchers should be involved in the implementation of all major policies that affect public health and health disparities. Plans to ensure the rigorous evaluation of performance should be in place before programs are implemented. Involvement of health services researchers in the planning process should help to ensure that data are captured in ways that facilitate implications of programs for health disparities. Evaluation of public health policy should be a priority from the outset, with mechanisms to gather the necessary information and link databases designed into implementation plans from the outset.
Lesson 4: Determine points of contact with communities where interventions can be introduced. The intervention studies described in this volume were each successful in finding natural and acceptable ways to intersect with people’s lives. For instance, reaching young adolescents through Boy Scout programs is a very important example. Clearly, offering these young men incentives and encouragement to engage in healthy behavior is a natural. It is possible to consider taking this to another level in several ways. Scouts and other youth leaders can be empowered to serve as champions for preventive health advocates in their communities.1
Very valuable information about community-specific cancer statistics and the quality of local health care can also be provided in venues such as beauty salons or on line.Again, interventions in these settings must be designed to ensure that individuals receive accurate information about health disparities and local choices in care. These settings can also facilitate consumers sharing information with one another about their experiences in care. In fact, it is highly likely that consumers in these settings already exchange information about the quality of doctors, hospitals, screening programs, and the like. Our interventions can support this process by ensuring that information is complete, accurate, and understood in the proper context. The next generation of research must foster the exchange of information about available health care choices, to help people make decisions about where to go for care.
Lesson 5: Communities should experience immediate benefits from involvement in research. In addition to specific strategies to promote informed decision making, motivate change, and build necessary skills and supports, many of the interventions in this volume have incorporated "nonspecific" components that are important to the success of health behavioral interventions. These same nonspecific factors are also important to the efforts to promote community support and involvement for social justice efforts. It is necessary to carry out interventions in ways that help people to build their social networks, learn something of value, receive recognition for their accomplishments, or simply have a good time—strategies employed by many of the interventions in this volume. As with any other community interventions, grassroots efforts to eliminate disparities must provide a positive, meaningful experience for participants.
In addition, participants in our interventions may be interested in health careers, including public health research. Our research programs should be involved in facilitating health careers by linking to training programs such as the NCIs CURE opportunities (National Cancer Institute [NCI], 2011). Programs that employ LHAs and other community health workers have also provided entrée to health careers for people in underserved communities. Future community-based research efforts should be linked to existing and new training opportunities, to broaden health career participation for people in low-income and underrepresented minority communities.
Lesson 6: Efforts to eliminate cancer disparities must engage the family. Several of the interventions described in this volume have included spouses, partners, and other family members of "target" populations. Focus on the family is intrinsically more complicated and challenging than working with individuals, particularly in underserved communities. At a minimum, involving family members entails logistical challenges, particularly in light of partners’ or family caregivers’ multiple roles. Family members may also face their own unaddressed health issues. A family history of conflict or communication problems adds even more challenges. In addition, delivery and financing of health services is often organized around individuals rather than families, and providers and settings may not be trained to engage families in care. Nonetheless, the advantages of working with families make it necessary to find ways to overcome these challenges, such as described in the work of Ayala and colleagues.More generally, implementing health promotion in ways that engage families is more consistent with the kinds of lifestyle changes that people must make in key areas such as diet, physical activity, sexual risks, and tobacco and substance use. Similarly, health decisions are often made in the family context. Empowering consumers means assisting families in making use of information about health care. The role of the family in reducing cancer health disparities should be a major focus of research across the cancer continuum.
Lesson 7: Eliminating disparities means enhancing quality of life. The research by both Ashing-Giwa and colleagues and Moadel and colleagues offers an interesting and important twist on the question of health disparities by focusing on diminished quality-of-life outcomes. This is a critical focus, especially if we are concerned about consumer empowerment. Recent research in the area of quality-of-life appraisal and response shift (Rapkin & Schwartz, 2004) suggests that people vary widely in the personal criteria and standards of comparison that they use to evaluate their own quality of life. Consumer empowerment means ensuring that people are evaluating their health and well-being and the quality of care available to them using appropriately high standards and expectations. In fact, helping to raise people’s standards and expectations for their own health or the quality of care that they receive may lead to expressions of dissatisfaction and reduced ratings of quality of life. This brings the discussion of our future research agenda full circle. Making people aware of disparities and inequities in their lives is a necessary part of the empowerment process, and such awareness may contribute to a diminished level of quality of life, increased medical mistrust, and a reduced sense of control. From an ethical point of view, it seems more acceptable to make people aware of the disparities that they face than to skirt the issue or encourage them to access care that we know is substandard. Nonetheless, pursuing a consumer empowerment agenda means that interventions must strive to ensure that people have ways to act on their own behalf to improve their access to quality. For example, in some communities, there may simply be no good options for quality cancer care for anyone, or for people with the wrong type of health insurance. Raising awareness of this fact makes it possible for health consumers and their families to take action on their own behalf. Research on empowerment-oriented interventions must be concerned with raising individuals’ standards for quality of life, and ensuring that avenues to take action on their own behalf are available.
Lesson 8: Theoretical constructs and frameworks that guide health behavioral and decision-making interventions are highly relevant to community mobilization and organizing. Several authors in this volume have discussed the importance of theory in designing interventions. Theoretical aspects included intrapersonal, interpersonal, and social contextual factors that could impact health behavior. Very similar theoretical constructs can be brought to bear to understand factors that would help raise individual awareness of the health disparities that affect them, their families, and their communities. For example, the PEN-3 framework includes constructs to identify factors that affect desired behaviors. Individuals necessarily hold positive, existential, and negative perceptions about the forces that affect the quality of health care available in their communities, and their ability to take actions. Enabling factors might include providing them with evidence about the comparative quality of care, or strategies that have been implemented in other communities to reduce disparities. Community organizing to promote awareness of others with similar concerns and a shared sense of struggle may help to nurture and sustain grassroots action.
Another important concept is the notion of intervention tailoring. Several of the interventions discussed in this volume build in procedures to assess the needs and priorities of individuals at the point of intervention delivery, to ensure that approaches are most responsive to individual circumstances. Tailoring is also relevant to community organizing. Determining and addressing individuals’ sources of greatest concern, or the most daunting barriers to involvement, must be a critical component to efforts to empower communities. It is certainly possible to imagine community organizers using a motivational interviewing approach to better understand how to engage individuals in taking action at a variety of levels. A number of the interventions described in this volume have incorporated multiple components, to support individuals through a progression of changes. A similar logic can be applied to community-organizing efforts.
As Griffith and colleagues discuss, tailoring must address both surface and deep-level realities and concerns. As with behavioral interventions, initiatives to expand social justice in health care must be appealing, accessible, and practical in order to encourage community participation. At a deeper level, interventions must take into account communities’ narratives of political struggle, transitions, and efforts to achieve quality of life. Interventions may need to elicit these narratives to guide change or to help communities develop alternatives to narratives of hopelessness and inevitable defeat. This can be an advantage of rooting interventions in geographically defined settings.Clearly, in many communities, spirituality and religious identity may be an important part of these narratives, just as they are in other struggles for civil rights and social justice.
TRANSFORMING CANCER HEALTH DISPARITIES RESEARCH TO PLACE SOCIAL JUSTICE AT THE CORE
As the lessons gleaned from the studies in this volume amply demonstrate, concepts and techniques that have been used successfully to bring people together for interventions and motivate behavioral change can be readily applied to efforts to empower health care consumers, build local capacity for delivering and adapting effective programs and services, and hold the health care system accountable for effective prevention and outcomes of cancer care. Nonetheless, it is a very different thing to conduct a study to help people quit smoking, or get a mammogram compared to a study to help people hold their local doctors and hospitals accountable for high-quality care. We have come to accept the aims of research to promote health behavior as legitimate, scientific questions. Is the same true for research to promote equitable treatment in health care and social justice, or is this tantamount to hijacking science for a political agenda? Should scarce research resources be spent on organizing communities and building self-advocacy skills to demand better health care in poor and underserved communities, rather than continuing to incrementally find ways to help individuals modify behaviors that we know lead to cancer?
Of course, in some ways this is a false dichotomy. For example, efforts to promote screening or smoking cessation could serve as an entrée to broader consumer education and empowerment. Even so, it is important to address a key ethical question. The justification for encouraging individuals to take a particular test or eat particular foods is that scientific research has shown that it is good for them. We do not have this same sort of justification when it comes to social justice. We cannot say that providing consumers with information about the poor quality of cancer care available in their community will be helpful. Even though we may have ample evidence that individuals in some communities are at high risk for receiving substandard care, we do not (yet) know whether there is any direct benefit if we raise awareness and help them to act on this information. Indeed, such information might be construed as harmful if it scares people away from necessary treatment.
However, there is another ethical dimension to this discussion. Much of our research is predicated on findings that strongly link cancer disparities to conditions of poverty and inequitable distribution of health resources. Is it ethical to withhold (or fail to actively communicate) this information? No matter how effective our efforts to promote health behavior, improve patientprovider communication or even enhance conditions in specific neighborhoods or communities, we researchers know that the findings of all of the individual research projects sponsored by the American Cancer Society (ACS), the NCI, and other major funders are highly unlikely to alter cancer disparities without major investment to get these programs into practice. Shouldn’t research participants also be made aware of this, as part of providing informed consent? Should the lack of social justice in health care be part of the risk-benefits equation in research participation (Powell, Rapkin, & Weiss, 2009)? More generally, even without guarantee of benefit, shouldn’t individuals have the right to know about the quality of health care available to them, to make informed choices in care? As Rappaport (1981) argued 30 years ago, this discussion reflects the tension between medical paternalism and patient empowerment, between a system driven by attention to caring for patient needs versus helping them to exercise and attain their human rights.
It is no coincidence that this portfolio of ACS-supported studies on health disparities uses methods and concepts that are so consonant with the social justice agenda. We have arrived at a point in the development of our understanding of science and public health at which we can no longer rest with the assumption that "normal science," predicated on a needs-based, individual-level, medical-model approach to overcoming health disparities, is viable. Rather, we are in the midst of a paradigm shift that involves a rethinking of the ways in which research can contribute to the elimination of cancer health disparities and other pervasive inequities in health care. Rather than being anti-scientific, the social justice agenda must propel this paradigm shift.
Recently, the NCI and other agencies have called upon the research community to consider a new generation of research designs and methods, to foster dissemination of evidence-based interventions. For example, the NCI has produced recommendations based on an ongoing series of "Dialogues on Dissemination" (NCI, 2010). Specific suggestions include support for community organizations involved in dissemination, incorporation of surveillance systems to monitor intervention implementation, and uptake of evidence-based practices. Key strategies include the need to involve representative stakeholders from all who might potentially benefit from dissemination (Barton-Villagrana, Bedney, & Miller, 2002; Miller & Shinn, 2005; Yoshikawa et al., 2003), to establish lasting partnerships between investigators and community collaborators (Altman, 1994; Lamb, Greenlick, & McCarty, 1998), to encourage a culture of evidence-based practice with community organizations (Foster-Fishman, Berkowitz, Lounsbury, Jacobsen, & Allen, 2001; Guerra & Knox, 2008), to create models of collaboration based on mutual self-interest (Wandersman, 2003; Stokols et al., 2005), and to make it easier for communities to institutionalize evidence-based strategies (Goodman, McLeroy, Steckler, & Holye, 1993;Goodman & Steckler, 1989). Recommended areas of study for dissemination research include factors that affect adoptions of innovation in practice, studies of motivational factors that affect implementation and uptake at multiple levels, research on the effective use of change agents, and contextual factors that influence and support implementation. The NCI has emphasized that research designs must move beyond exclusively individual outcomes to also address systems- and population-level changes, with priority given to targeted dissemination efforts able to reach middle-to-late adopters of evidence-based practices (Von Eschenbach, 2003).
There is great potential for dissemination research to foster social justice in the service of eliminating health disparities. To realize this potential, dissemination research must eschew the traditional medical-model view of dissemination, which places a premium on direct replication of behavioral interventions with high fidelity to core elements. There is a very strong scientific and ethical rationale for moving away from top-down technology transfer (Bull, Gillette, Glasgow, & Estabrooks, 2003; Glasgow, 2003; Glasgow, Goldstein, Ockene, & Pronk, 2004; Kerner, 2002). Due to an almost complete lack of attention to external validity in research, we cannot claim with any certainty that an intervention found to be effective in one setting will be equally efficacious in another setting (Bonfill Cosp, Marzo Castillejo, Pladevall Vila, Marti, & Emparanza, 2001; Green, 2001; Green & Glasgow, 2006; Greenhalgh, Robert, Macfarlane, Bate, & Kyriakadou, 2004). Thus, it is generally not warranted to offer behavioral interventions to communities as well-established solutions or best practices. Rather, the most rigorous way to gain any clarity on the question of "what to disseminate" is to try out different interventions in different places (Brownson et al., 2007; Green & Ottoson, 2004). This would allow us to gain an understanding of "program-context interaction" through what Green and Glasgow (2006) refer to as practice-based evidence. Green and Ottoson (2004) discuss dissemination as a joint process of problem-solving, involving key stakeholders in making decisions about proper approaches to implement in different contexts. Similarly, Sandler (2007) argues that it is necessary to engage communities in critical thinking and reflection about what programs they need and how they are working.
The key is that we cannot expect to overcome disparities by applying any particular solution in a given community. Rather, research must foster a shared process of critical reflection and problem-solving over time, using scientific methods to progressively arrive at optimal solutions for specific settings. We need to stop funding research designed to be "definitive trials," intended to establish universal "intervention effects." Rather, the science that we need to overcome disparities will be advanced by garnering a deep understanding of how to use behavioral science principles, methods, and interventions to support community problem-solving.
In addition to being scientifically and ethically superior, community-engaged approaches to dissemination of evidence-based practice provide a strong platform for advancing social justice. There is no better way to foster this scientific direction than to expand the criteria that reviewers consider in evaluating research proposals and journal articles. Lessons learned from the portfolio of ACS-supported research presented in this volume point to criteria that will serve to ensure that the next generation of cancer health disparities interventions advances social justice. Consider how emphasis on the following review criteria would start to transform the science of cancer health disparities:
1. Does the research foster community ownership of research methods as a problem-solving strategy to stimulate ongoing community dialogue about disparities and health promotion based on research?
2. Are intervention outcomes understood in terms of improvement in indicators of overall performance, including reach, efficiency, sustainability, and fit with the local service ecology, not merely efficacy?
3. Do interventions designs incorporate quality improvement and tailoring strategies to maximize intervention performance in the local community?
4. Does research empower health consumers by increasing the communities’ understanding of local cancer health disparities and by providing action steps to address those disparities?
5. Does the research project increase the accountability of local providers and make it possible for providers to earn the greater trust of the community through effective communications and improved services?
6. Does the research provide information needed by policy makers to guide allocation of resources?
7. Does the research project offer opportunities for leadership, skill development, and access to health careers for local community members?
8. Does the intervention make provisions to include people experiencing multiple physical, psychological, and/or social problems that may interfere with intervention outcomes?
9. Does the research champion the value of diversity, and enhance the social identities of groups that are subject to marginalization, disenfranchisement, and stigma?
10. Does the intervention strive to affect all relevant outcomes, including noncancer-specific outcomes such as better access to primary care and improved overall quality of life?
11. Does the research contribute to a transparent, ongoing relationship between care providers, community stakeholders, and academic researchers that extends beyond the life of any one study?
12. Will planned analyses place findings in context, to permit comparison with other studies and to determine the external validity of results?
Each of these criteria may be operationally defined and evaluated in the context of grant review, and in the evaluation of studies for inclusion in meta-analysis and the like. Of course, there is considerable need to determine standards for how research projects can incorporate these criteria. Nonetheless, thinking of quality research in these dimensions (as opposed to applying the randomized control trial as the single standard of excellence) would go a long way in fostering research to address the sources of social injustice that underlie cancer health disparities.
