The context of housing health hazards involving children requires a fresh look at the ethical framework for research involving humans. The ethical principles of respect for persons, beneficence, and justice may have particular meaning within the context of research that takes place in the home and community.
The ethical principle of respect for persons incorporates both the notion that individuals should be treated as autonomous agents and that persons with diminished autonomy are entitled to protections. In the context of housing health hazards research, one can question whether the community environment compromises voluntariness of consent. The environment can be viewed as potentially coercive when persons are living in hazardous conditions and when moving out of the situation (e.g., to better housing) is not a feasible option.
The principle of beneficence requires that persons are treated in an ethical manner, not only by respecting their decisions and protecting them from harm, but also by facilitating their well-being. It assumes that researchers are operating in a manner that will both maximize the potential benefits and minimize the possible harms of the research. Concern about the application of this principle can be raised because parents and community representatives may have markedly different assessments of the benefits and risks of such research than researchers and institutional review boards (IRBs).
Finally, justice addresses the questions of who should receive the benefits of research and who should bear its burden. An injustice occurs if someone is denied a benefit to which he or she is entitled or when a burden is unduly imposed. The principle of justice is typically discussed in the context of equitable selection of participants, so that no group receives a disproportionate share of the benefits or burdens of research; it has generally received less attention than the other two ethical principles. However, it is particularly salient in the context of housing health hazards research.
The distributive paradigm of justice may be challenged when considered in the context of communities that lack economic, social, and political power. Residents of these communities may have experienced a history of racism, classism, segregation, unequal access to quality health care, discrimination, and past research abuses. The resultant pervasiveness of distrust toward researchers and research institutions among many individuals and communities in which this research will take place is one aspect of the difficulty in applying the principle of justice. Community residents may believe that changes in housing and economic policies to improve low-income housing are needed, rather than more research. In addition, as discussed in Chapter 4, community residents may have a different view than researchers regarding what constitutes a risk and whether a particular research project is worthwhile.
Some ethicists have suggested using a relationship paradigm that acknowledges the notion that an individual is situated in his or her social context, environment, or community to support thinking about complex ethical issues in underserved or vulnerable communities (King et al., 1999; Fisher, 1997). A relational approach to ethical decision making draws attention to complex social, cultural, and political contextual factors. This paradigm acknowledges that researchers and participants may differ in their assessment of research risks and benefits and values each perspective. Researchers are encouraged to engage participants and community residents—and community residents are encouraged to proactively engage researchers—as partners in designing research that meets the needs and addresses the values of all stakeholders. In this paradigm, justice evolves into a principle that goes beyond the issues of participant selection to a principle that is related to past, present, and future distributions of power (King et al., 1999; Institute of Medicine, 1999). This shift reflects the view that research is part of a broader societal context and that the conduct of research often mirrors a system in which power is unequally and perhaps unfairly distributed.
Through discussion with potential and actual research participants, the respective views of researchers and community residents can be examined and a research plan can be constructed that accommodates the views and values of prospective participants, their families, and their communities
(Fisher, 1997; National Bioethics Advisory Commission, 2001). A relational approach can help address concerns regarding respect for persons, beneficence, and justice in housing health hazards research. Community involvement, through enhancing respect for persons and shaping research that addresses the needs and priorities of the communities, is one way to address fundamental ethical principles.
VALUE OF COMMUNITY INVOLVEMENT
Involving a community in housing health hazards research, especially when the involvement is in the nature of partnerships, offers many advantages. Community representatives bring different perspectives, values, and competencies, which can contribute to the research project; see Box 5-1. While investigators bring technical knowledge about the subject of study and about research methodology, representatives of the community bring knowledge of community concerns, needs, values, and priorities. They also can bring a history of activism, leadership, and coalition building and a network of community contacts. Their expertise can shape research in constructive ways by posing significant questions for study, pointing out ethical concerns, suggesting how to modify a study to reduce risks and increase acceptance of the research in the community, assuring that data collection instruments are culturally appropriate, and promoting enrollment and retention. Community input also can help researchers determine what and when compensation for participation in research is appropriate and design informed consent processes that ensure that parents understand the essential features of the research protocol. (See Chapter 6 for additional discussion of these issues.) Furthermore, community groups can help researchers anticipate and respond to risks observed in the home when the observations are not part of the research protocol. (See Chapter 7 for discussion of this issue.)
Moreover, although there is little empirical evidence on this point, community groups can use research results to advocate for changes in public policy that will allow research interventions to be sustained and disseminated. Research findings often validate long-held community concerns. Thus, community involvement may increase the likelihood that research results are translated into actions and changes that benefit the communities in which it is conducted. Communication between researchers and community members can help identify ways in which research results can be effectively communicated to and applied by the affected community, including identification of public officials whom researchers should include in their dissemination plans.
The Boston Lead in Soil Study, funded by the Environmental Protection Agency in the 1980s, was one of the first large-scale intervention studies that examined the housing-based health hazards posed to children. The study was designed to examine whether certain low-cost methods for the removal of lead-contaminated soil surrounding a home would reduce the blood lead levels of young children living in homes in disadvantaged neighborhoods.
Community representatives participated in and contributed to the project in multiple ways. An individual who was both an employee of the health department that conducted the study and an active neighborhood resident insisted that a community advisory group be formed, that it be active in all phases of the study, and that it be a part of the decision-making process. The sense of ownership that the community held towards the project was based on the successful education and dissemination of information about both the specific potential link between the soil in the neighborhood and the high incidence of lead poisoning of the neighborhood children and the general connection between environmental hazards in and around homes and child health. The community advisory board made several significant contributions to the project:
SOURCE: Michael Weitzman, principal investigator (personal communication, December 2004).
PRINCIPLES OF COMMUNITY INVOLVEMENT
Community involvement can take many forms. The Centers for Disease Control and Prevention recommends that researchers apply and adapt with “understanding, skill, and sensitivity” concepts from literature on community participation, community mobilization, community empowerment, cultural influences, and others. Representative strategies to engage communities in research include capacity building, coalition building, and community organizing (Centers for Disease Control and Prevention, 1997c).
In the context of health interventions, it is recognized that efforts to
involve the community should address multiple levels of the social environment. For example, health behaviors are influenced by culture. Efforts to engage residents will be advanced when they feel a sense of community, see the process as worthwhile and inclusive, and believe the benefits outweigh the costs. A sense of empowerment is crucial to successful engagement efforts. Active community involvement will take time: community mobilization and self-determination frequently need nurturing. In the end, however, coalitions can be used for mobilizing, developing, and using community assets for health decision making and action (Centers for Disease Control and Prevention, 1997c).
Several authors have enumerated key principles in community engagement as it relates to community based participatory research, one form of community involvement (see below) (Centers for Disease Control and Prevention, 1997c; Israel et al., 1998; Minkler and Wallerstein, 2003). These principles might also be useful with other types of community involvement. First, researchers need to recognize the community as a unit of identity and attempt to work with existing communities. Investigators can use tools of community assessment and diagnosis to learn about the community and its economic conditions, history, norms, demographic trends, and political structure. Successful research collaboration requires researchers to understand how residents structure and maintain communities (Kone et al., 2000).
Second, community engagement and research collaboration should build on strengths and resources within the community by explicitly recognizing and supporting social structures and processes that contribute to the ability of residents to work together to improve health. Resources may include skills and assets of individuals, networks of trusting relationships, or existing organizations and institutions, such as places of worship. Because of the intensive nature of community-based participatory research, investigators who advocate for this type of community involvement move beyond these concepts to stress the importance of equal partnerships (Green and Mercer, 2001; Israel et al., 1998).
Researchers may benefit from including on the research team or consulting with individuals who are knowledgeable about or representative of the community. Such persons might be scholars in relevant social science disciplines, leaders of community-based organizations, or front-line social service or medical practitioners. Consultants familiar with a community might help identify diversity within a geographic community; a given community might have multiple groups with different demographics, social structures, and neighborhood and cultural histories.
Researchers might ask a series of questions related to the community during an initial assessment phase to help address important issues in planning a project:
Have I conducted preliminary research to ascertain the concerns of the community from which I will enroll children?
Has a community needs assessment been done to inform the research agenda?
Is the proposed research relevant to the communities of concern?
What role should community residents play in order to improve the research, facilitate informed decision making by parents, and translate and disseminate findings to the community, the broader public, and policy makers?
What capacity building is necessary to achieving meaningful involvement in the community?
How will I evaluate the effectiveness of the approach taken to community involvement?
Is there an interdisciplinary research team and advisors to guide the study and help evaluate process and outcomes?
Do I have plans to ensure that the research staff are culturally competent and sensitive to the issues in the community?
Researchers should also anticipate the questions likely to be raised by community residents (adapted from Center for Minority Health, 2004):
What is/are the exact question(s) you are attempting to answer?
How will this research benefit the targeted population?
How does this research address the stated problem?
What jobs will be available as a result of this research study and how will the community be informed about them?
What will be your method of recruiting children?
What is your proposed outcome(s)?
How will you disseminate your research findings?
How will you ensure that families can understand your questions?
How will you ensure your study is implemented in a culturally competent manner?
What written materials will be given to the parents of child subjects or potential child subjects?
What does the informed consent form say?
Are health literacy issues taken into consideration?
Is the recruitment material culturally appropriate for the targeted population?
What has the research team done to recruit the targeted population?
How culturally diverse is the research team?
The Preventing Agricultural Chemical Exposure (PACE) project was a 4-year project to reduce pesticide exposure of farmworkers in North Carolina using a community-participation framework. PACE involved a partnership among Wake Forest University School of Medicine, the University of North Carolina at Chapel Hill, and the North Carolina Farmworkers’ Project (NCFP). Many of the Latino farmworkers in this project came directly from Mexico, they were not unionized, and advocacy organizations for this population were relatively new. These characteristics suggested that pesticide safety interventions developed elsewhere in the United States might be inadequate for this population.
The project used a two-part planning framework, known as “precede-proceed”: “precede” focused on problem identification and intervention planning, and “proceed” focused on implementation and evaluation.
The precede phase included review of existing data sources and training manuals, solicitation of information, quarterly advisory meetings, and data collection from four constituencies: (1) farmworkers, (2) farmers, (3) staff of the Cooperative Extension of the U.S. Department of Agriculture, and (4) health care personnel. The four groups varied in their perceptions of farmworker pesticide exposure, risks of exposure, and effectiveness of regulations in protecting farmworkers. For example, while health care personnel believed that pesticide exposure was underreported and undertreated, farmers and Cooperative Extension staff perceived no problem with exposure. Farmworkers were concerned about pesticide exposure, but evidence indicated they were not fully aware of the risks. From these data, specific behavioral changes to be implemented in future phases were identified related to hand washing, wearing clothes that cover most exposed skin, wearing work clothes only one day before washing, and washing work clothes separately from household clothes. Specific environmental changes were also identified, including increasing farmworkers’ control of their work environment and reducing pesticides in housing. Constituent input also helped identify factors that could predispose or
APPROACHES TO COMMUNITY INVOLVEMENT
Community involvement can occur along a broad spectrum. Residents can be involved as parents of child subjects or as staff who conduct interviews, moderate focus groups, or contribute to the analysis and interpretation of findings. Alternatively, they may have a decision-making role as members of IRBs or as research partners. Greater involvement of community residents is associated with maximizing the potential for capacity building, salience of research questions and proposed solutions, and the relevance and sustainability of interventions (Hatch et al., 1993); see Box 5-2 for an example.
position farmworkers to make changes (e.g., farmworkers’ lack of knowledge of pesticide residues and existing regulations regarding pesticide exposure and perceived susceptibility to pesticide-related health problems), reinforce beneficial changes that farmworkers made (e.g., use of field safety promoters to provide education and support and use of El Terror Invisible symbol in educational materials to represent pesticide residues), and enable workers to make appropriate changes.
The design of the intervention was guided by three ideas: (1) that the intervention focus on key concepts; (2) that it be relevant to the local farming system; and (3) that it assist workers in maximizing their control over pesticide exposure. Both product and process evaluations were conducted.
The collaborative nature of the PACE project led to the development of a pesticide safety program that has been sustained and disseminated in several ways:
SOURCES: Information from Arcury et al. (1999, 2001); Quandt et al. (2001).
The mechanisms used for community involvement may be additive. And, as researchers and community members develop trust and working relationships, the level of community involvement may evolve. Investigators may use multiple strategies to maximize the communities’ capacity for current and future research partnerships and advocacy (Quandt et al., 2001; May et al., 2003). Community residents can be involved in the research process as research staff, through community consultation and review, membership on community advisory boards, and involvement in a community-based participatory research process. As discussed in Chapter 8, IRBs may also seek out input from community residents as part of their review of housing health hazards research protocols.
Community representatives may ask whether the research team has appropriate cultural diversity and whether community residents will have the opportunity to gain employment and learn job skills. Community advisory boards may ask specifically how people will be informed of possible jobs with the research project (Center for Minority Health, 2004).
As paid research staff, community members can directly benefit from the research process through the development of new skills and expertise. A variety of potential staff roles include (1) consultation on protocol development and research design; (2) consultation and development of data collection instruments; (3) recruitment and retention, including outreach to families in their homes; (4) data collection; (5) data analysis and interpretation; and (6) dissemination of findings within the community. Researchers must ensure that community members who are hired have the requisite skills for the tasks identified and are not merely those most vocal or visible in the community. The extent to which community residents who serve as research staff can influence the research process will depend on the context of the study and the willingness of those in supervisory roles to elicit and incorporate their views. A challenge is to avoid letting such staff become marginalized on the research team. In addition, researchers must guard against the possibility of corruption of local staff’s views while remaining open to views different than their own.
Community Consultation or Review
Community consultation or review enhances respect for families involved in research and communities. By deliberately bringing together researchers and community residents in dialogue, community consultation allows everyone to be informed about the potential risks, burdens, and potential benefits of research to the group and may provide more meaningful input than merely hiring community residents as staff. Foster and colleagues (1999) describe a range of activities that might be a part of community consultation: informal discussions with community members; community involvement in research planning; community participation in the evaluation of human subject protections; and negotiations of formal community agreements. Dula recommends that such consultation be “culturally sensitive, jargon free, and strictly honest about the benefits and burdens of experimentation” and be offered in a spirit of “professional humility” (Dula, 1994). Community consultation can be a key part of the “the process of engagement, dialogue, and feedback” that is engaged in before, during, and after research (Dula, 1994). While in no way replacing individual informed consent, community consultation might lead to better
communication about research-associated risks and potential benefits and might improve the informed consent process. Levine and colleagues (1991), in the context of clinical trials of new antiretroviral medications for AIDS, recommend that “prior and ongoing community consultation should be an integral part of the planning and design of clinical trials of nonvalidated therapies” (p. 10). They identify determining whether the trials are considered acceptable by the majority of prospective participants as a specific objective of community consultation.
The theory and practice of community consultation has its limitations. Most community consultation guidelines on protecting communities from research harms have focused on indigenous communities, which are generally geographically bounded, socially distinct, and have traditions of self-identification and political autonomy. To the extent that other communities do not have these characteristics, community consultation may be problematic (Weijer and Emanuel, 2000). Indeed, it may be difficult to define the community in question or to identify appropriate representatives. Community consultation may produce locally variable findings in geographically dispersed populations and highlight difficult-to-reconcile perspectives caused by nested social identities (Foster et al., 1999). More formalized processes, though more involved, may address some of these concerns
Community Advisory Boards
Community advisory boards have been proposed as one mechanism for meaningful community input into the design of research projects (Quinn, 2004; Morin et al., 2003). Advisory boards offer an opportunity to adopt a relationships paradigm that enables researchers to anticipate and address the context in which communities understand risks and benefits and individuals give consent (King et al., 1999; Fisher, 1997). Community advisory boards can be created to advise a specific study or have oversight of multiple studies by an investigator or institution; see Box 5-3 for an example. They often advise researchers by assisting in the development of materials, representing participant concerns, disseminating information about the study and its risks and benefits to the community, providing a set of recommendations to help potential participants decide whether or not to participate in a study, and acting as advocates for rights of human subjects. Advocacy might include creating documents and resources that support participants’ rights and promote community perspectives, refocusing the research question on the needs of vulnerable groups, and modifying or stopping trials that are considered or become unethical (Cox et al., 1998; Morin et al., 2003; Quinn, 2004).
The goal of Seattle Partners for Healthy Communities (SPHC) is to “support and evaluate community-based health promotion programs that are sensitive to the cultural needs of minority communities in Central and Southeast Seattle.” SPHC uses a participatory action research approach that stresses “participation, equal power, and joint planning.” SPHC developed a community-researcher model that included a community board of residents, activists, community-based organization staff, health professionals, and researchers. This board developed collaborative principles that (1) ensure community participation; (2) respect community values, contributions, and perspectives; and (3) ensure community benefit from the research. Responsibilities of the board include selecting and monitoring projects, as well as approving grant applications, budgets, and hiring decisions. Each project has its own steering committee, which designs the project and evaluation, oversees the project budget and implementation, reviews findings, and helps disseminate results. The board provides feedback and offers suggestions to steering committees on a quarterly basis.
One SPHC project is the Seattle-King County Healthy Homes Project (SKCHH), designed to “improve asthma-related health status by addressing poor housing and indoor environmental quality.” The first phase of SKCHH was a randomized controlled trial of an outreach and education intervention to reduce exposure to allergens, irritants, and toxins in homes. The intervention included home environmental assessment by a community health worker, who developed an action plan with participants to reduce exposures. Over the next year, outreach workers made seven additional visits to provide education, social support, and resources to reduce exposure, and advocacy for improved housing conditions.
The project steering committee included representatives from national and
Community-Based Participatory Research
Community-based participatory research is an intensive approach that allows communities affected by research to participate in all phases of the research process. Under the principles of community-based participatory research, research must address the concerns, needs, and priorities of the communities where it is conducted and lead to actions and changes that benefit the community. Community stakeholders are active partners in all aspects of the study, from the formulation of research questions to the application of findings. They use their knowledge and experience in the community to specify issues to be studied, develop research questions in culturally sensitive ways, and use study results to help support relevant program and policy development (Israel et al., 2003). Community-based participatory research has been used in research on a range of topics, in-
community organizations—including the American Lung Association, Center for Multicultural Health, Community Coalition for Environmental Justice, Seattle Housing Authority, Community Health Centers—the health department, and university faculty. The steering committee promotes power sharing between the community and researchers, oversees project implementation, and reviews evaluation findings. In addition, a parent advisory group was convened to obtain input directly from affected community residents. Members were recruited through contacts in public housing, health, and environmental justice communities. Each member is reimbursed for meeting time ($10/2-hour meeting) and is provided with dinner, child care, and transportation. The group meets once or twice a year to provide advice on project implementation and evaluation, as well as their perceptions of the collaborative process.
Issues raised by community partners have included: tensions between working to improve housing quality while minimizing risk of eviction for tenants; sensitivity of entering homes and evaluating levels of asthma triggers (e.g., presence of cockroaches, level of household cleanliness); recruitment difficulties related to mistrust of government by people of color, especially immigrants; the length, cultural appropriateness, and validity of surveys; use of control groups that receive no benefit and randomization of study participants; and the high complexity and unacceptability of some intervention protocols. These concerns were addressed openly at steering committee meetings. Resolutions included protocols developed to instruct tenants on their rights, show them how to approach their landlords, and make referrals to the tenants union; simplification or elimination of other protocols; removal or revision potentially offensive or overly intrusive questions on the survey and the provision of a lower-intensity version of the intervention to the control group.
SOURCES: Information from Kone et al. (2000); Krieger et al. (2002a, 2002b); Metzler et al. (2003); and Krieger et al. (2005).
cluding cancer screening, international public health, nutrition, genetics, tobacco control, as well as research on housing health hazards (Viswanathan et al., 2004; Ammerman et al., 2003; Corbie-Smith et al., 2003; Israel et al., 2003; Lam et al., 2003; Lewis et al., 2004; Metzler et al., 2003; Minkler and Wallerstein, 2003; Naylor et al., 2002; Parker et al., 2003a; Riley et al., 2001; Sloane et al., 2003; Weijer and Emanuel, 2000; Mercer et al., 2004).
Community-based participatory research views community members as equal partners with researchers (Viswanathan et al., 2004; Minkler and Wallerstein, 2003). Such a partnership requires a long-term process of mutual learning, capacity building, and sharing of resources. This approach builds on the strengths and resources within communities. It involves learning on the part of both researchers and community representatives, recognizing the expertise of each individual; see Box 5-4. Scientists need to learn the values and perspectives of the community and the organizations and
Community Action Against Asthma (CAAA) is a project of the Michigan Center for Environment and Children’s Health, which was funded by the National Institute of Environmental Health Sciences and the Environmental Protection Agency. The housing-based intervention of the project stemmed from research on the effect of indoor and outdoor air quality on childhood asthma and was adapted from the Seattle-King County Healthy Homes Project (see Box 5-3). The intervention consisted of at least 12 home visits over 2 years by a “community environmental specialist” who provided families with education and materials related to the reduction of exposure to asthma triggers. In addition to the housing-based intervention, there was a complementary neighborhood intervention in which community organizers worked with community residents to reduce physical and psychosocial environmental asthma stressors on a neighborhood level. Recruitment occurred in public schools, where screening questionnaires were sent to parents of children aged 6-10 to identify those with symptoms of asthma; 328 families agreed to participate. It was a randomized study design with delay, so that half the families received the household intervention immediately after baseline data collection and the other half received the intervention during the second year.
Of the families that began the intervention, 81 percent of the children were African American, 10 percent were Latino; 46 percent of the families reported incomes of less than $10,000 a year; 47 percent of the children fell into the moderate to severe asthma category; and 82 percent of the children had been previously diagnosed with asthma.
The adoption, implementation, and evaluation of CAAA was conducted in accordance with the principles of community-based participatory research. One focal point of the project was the establishment of a steering committee that was actively involved in all major phases of research. The steering committee was made up of representatives from community-based organizations, the local health department, an academic institution, the integrated health care system, and one non-institution-affiliated member. Sample accomplishments included:
resources within the community. Community representatives need to learn about the problems to be studied, research methodologies, and organizational skills. Partners need to learn from each other and be willing to teach each other. This process requires patience, openness to new knowledge, interpersonal skills, and respect for people from different backgrounds. Community representatives can be empowered through data, the communication of study findings, and health literacy materials or workshops to take
There were also several challenges encountered in the implementation of community-based participatory research principles, including:
The research and community partners of CAAA derived several lessons from each effort to adhere to the principles of community-based participatory research. For example, they found that:
SOURCES: Information from Lewis et al. (2004); Parker et al. (2003a, 2004).
action on their own behalf. The process is cyclical and iterative, requiring time and commitment. Communities may change over time, and therefore the relationship between researchers and communities will also evolve. Partnerships ideally are developed for the long term, not just to obtain a single grant or project.
Community-based participatory research advocates sharing of decision-making power and resources between researchers and community part-
ners. Most often this takes the form of community advisory committees, although there are also examples of steering committees in which community partners had equal power. For example, at the Columbia Center for Children’s Environmental Health, the community partner serves as a co-principal investigator and is codirector for the community outreach and translational research component of the project.
Even with this approach there can be considerable variability in the extent of community influence on the research process. An evidence-based review conducted for the Agency for Healthcare Research and Quality found that community partners were involved in intervention design and implementation in 93 percent of intervention studies and involved in participant recruitment and retention in 83 percent of intervention and nonintervention studies (Viswanathan et al., 2004). Community partners were involved with helping to set research priorities in 47 percent of all studies. There was little mention of community involvement in data interpretation or manuscript preparation. In some cases, limiting the community’s involvement in data interpretation may be appropriate to avoid actual or perceived lack of objectivity. At the same time, community partners might be able to help the researchers understand why certain results have occurred.
Community participation varied greatly across projects carried out through the Centers for Children’s Environmental Health and Disease Prevention Research, all of which were required to incorporate a community-based participatory research component (Israel et al., 2005). At most centers, community partners made valuable contributions in research design, development of data collection instruments, and recruitment and retention activities. For example, in one study community partners who insisted that all child subjects receive some direct benefit from the research convinced investigators to change a control group to a delayed treatment group. Community partners could also suggest new topics for data collection and ensure that instruments were culturally appropriate. Community partners were active in developing appropriate and effective strategies for recruitment and retention of child subjects. In some cases, community partners were involved with data collection through the hiring and training of local residents. Moreover, community partners played important roles in disseminating the findings of the research to the community in various ways, as well as providing general information to the community about the topic of study. However, community partners played little role in defining the research topic and questions and in data analysis and interpretation.
Community involvement in housing research has the potential to produce tangible research benefits in terms of better recruitment, better data, better analysis, and more relevant and sustainable interventions. However, there is also the potential for manipulation of individuals and communities, especially when involvement is limited and decision-making power of com-
munity members is absent (Hatch et al., 1993). When considering involvement of the community, it is important for researchers to clearly define the relevant community and carefully consider the extent of community involvement that enhances the research.
EVIDENCE ABOUT COMMUNITY INVOLVEMENT
While some consider achieving meaningful community involvement “a prerequisite for effective and efficient studies of children’s environmental exposure” (Needham and Sexton, 2000, p. 615), most evidence regarding the benefit of community involvement relates to process measures rather than outcomes. These findings document both the degree of involvement community members had in the research process, as well as the intermediate results of that involvement (e.g., better recruitment and retention of study participants, greater participant satisfaction, more willingness to engage in further research studies, increased trust in researchers). The details are most commonly found in qualitative reports that are usually published separately from the main intervention results. This information is critical to determining whether the principles of community-based research are being followed (Israel et al., 1998) and may be helpful in improving the validity and reliability of data (Kagawa-Singer, 2000).
Greater participation may help investigators to determine how best to understand, approach, and address housing health hazards in a particular community. A systematic review of community-based participatory research interventions (Viswanathan et al., 2004) found the most common outcomes were enhanced program quality (11 of 12 studies) and recruitment efforts (8 of 12 studies). Other outcomes included improved research methods, dissemination, and descriptive measures.
One example of these enhanced research processes comes from The East Side Village Health Worker Partnership in Detroit (Schulz et al., 2002). The partnership addressed social determinants of health problems, such as childhood asthma, using interventions and community change strategies at the individual, organizational, and community level. As a consequence of collaboration with the community, the research team produced a context-specific conceptual framework that guided development of a community survey. The researchers also concluded that the regular partnership meetings enhanced their understanding and interpretation of research findings.
Similarly, the Cameron Park Project of Cameron Park, Texas, used participatory action research in order to obtain valid and reliable information from community residents for an environmental health education program assessment (May et al., 2003). The communities were colonias, isolated, unincorporated rural settlements near the U.S.-Mexico border with marginalized Hispanic populations. The program used promotoras, or in-
digenous outreach workers, to address concerns that resident wariness would affect participation rates and data quality. The promotoras were involved in different areas of research design (e.g., designing the interview protocol), implementation (e.g., sampling and data collection), and evaluation (e.g., analyzing and interpreting data). These community members provided specialized expertise that improved data quality, accuracy, and interpretation.
Community involvement may also increase residents’ trust of researchers (Corbie-Smith et al., 2004; Schulz et al., 1997). High levels of distrust of researchers are well documented, especially among minority populations (Corbie-Smith et al., 2002). In the Cameron Park Project, for example, using community residents in both decision-making and implementation roles helped to bridge researcher and community cultures in order to overcome resident guardedness towards outsiders (May et al., 2003). Increasing community participation in the research process may help establish and maintain “reciprocal relations,” a two-way flow of resources that benefits the investigator, individual study participants, and larger community (Corbie-Smith et al., 2004). For example, increased trust was one of the intermediate outcomes identified in urban research centers, funded by the Centers for Disease Control and Prevention (Metzler et al., 2003), which required community collaboration. All members of the collaboration invested significant energy and resources into building trust. This increased trust was assisted by the creation of collaboratively developed principles and procedures, as well as the successful pursuit of research funds.
Less information is available on whether community involvement strategies lead to improvements in health outcomes or other project-level outcomes. One systematic review of community-based participatory research studies was unable to conduct a comparison of the relative health effects of the interventions, because methods and outcomes differed across projects (Viswanathan et al., 2004). Another review of health-focused collaborations found evidence of community-wide behavior change (e.g., decreased substance abuse, increased physical activity, more safer sex) as a result of the partnerships (Roussos and Fawcett, 2000); however, the authors concluded that there was insufficient evidence to determine the impact of collaborative activities on population-level morbidity and mortality.
There are limits to what is known or can be known about the outcomes of community involvement. First, the published literature about community involvement in environmental health research is most commonly about community-based participatory research strategies. Second, most research on community collaboration has been descriptive rather than comparative (Thompson et al., 2003; Zakus and Lysack, 1998), and few studies have investigated whether greater community involvement in research is related to better project outcomes (Roussos and Fawcett, 2000; Viswanathan et al.,
2004). Additional outcomes research is needed to better understand the effects of various levels of community involvement in research on the subsequent health and well-being of communities and families who participated in research.
ISSUES IN COMMUNITY INVOLVEMENT
Involving communities and community residents in research poses challenges that must be considered in the design of the research project.
Definition of Community
Communities may be characterized in a variety of ways (Weijer and Emanuel, 2000). The term community is used to refer to a variety of populations comprised by persons who have, or are perceived to have, commonalities that can be identified. Weijer and Emanuel (2000) propose a typology of morally relevant characteristics of communities involved in research. These include a common culture and history, legitimate political authority, geographical localization, a common economy and resources, and self-identification as a community. At one end of the spectrum are communities with a defined political and cultural identity, such as Native American tribes; other communities, such as those defined by a disease, may be much less cohesive.
Because the word is used in a variety of contexts and with different meanings, it needs to be defined carefully. Some groups are cohesive in the sense that they self-identify as a community. Some groups may be defined by geographical boundaries, while other groups may be physically dispersed. Some groups may be characterized by a common culture or ethnicity, while other groups may be culturally or ethnically diverse. Still others may be defined by a particular physical condition. Before they undertake a project, researchers need to identify the community that is relevant for that project.
Researchers on housing health hazards involving children must recognize that a community that appears homogeneous may actually be quite heterogeneous. For instance, residents of a low-income neighborhood may appear to be homogeneous—for example, African American and Latino. However, the perceived African Americans may be of African or Caribbean heritage, and the Latinos may have emigrated from a variety of Spanish-speaking countries with differing cultures, history, and traditions.
The continuing legacy of institutionalized racism, housing segregation, or “residential apartheid,” as sociologist and environmental justice advocate Robert Bullard writes, tends to cluster residents of color—whose backgrounds are of specific ethnic, racial, or national origin (regardless of in-
come and class)—in contiguous, overlapping, or densely populated geographic areas (Bullard, 1994). For example, a multifamily building may be located in a low-income area but may have residents of differing class, income levels, and educational attainment. To the extent that persons of higher educational and economic levels live in housing containing health hazards, it would be advisable for researchers to try to include such children as study subjects.
The projects of the Centers for Children’s Environmental Health and Disease Prevention Research took a variety of approaches to defining community. Most used “geographic boundaries and common characteristics,” which included “lack of resources and disproportionate burden of health problems they face” (Israel et al., 2005). Some centers involved “communities of identity,” defined by ethnic background. In two cases, centers decided that diverse stakeholders in a large geographical area needed to be involved, for example “both farm workers and representatives from agricultural industry organizations.” One center defined the community in terms of having a child with a specific condition.
Identification of Community Representatives
Once “community” is defined, it may still be unclear which residents are representative of the identified community. There may be diverse views regarding the research. Some communities may have a formal governmental structure and a recognized political authority, as Native American tribes. Other communities may lack political authority but have clearly defined leaders, as in religious communities. However, leadership may be contested: even when there is legitimate political authority, critics may charge that the leaders are not acting in the best interests of the community with regard to the proposed research project.
In communities with a diversity of leaders and views, there is a risk that researchers will selectively work with those who are favorable to the research project. Moreover, there are ethical concerns that the community representatives with whom the researchers are working may not legitimately represent the interests and views of those who have not been consulted about the research (Juengst, 2000). At the extreme, researchers may engage in “forum shopping,” which has been characterized as “the ‘morally problematic’ practice of seeking out population spokespeople and research participants whose positive response to a research plan can be predicted in advance” (Juengst, 2000). In many communities, there are multiple sources of leadership and authority. Careful consideration must be given to what aspect of the community a particular person or organization will represent and how researchers can hear the range of views in a community. Researchers need to be imaginative in trying to identify different voices in a commu-
nity and ensure that residents are represented. While the specific community representatives to be consulted depend on the protocol and the communities where the studies will be carried out, researchers should consider reaching out to multiple nongovernment organizations such as places of worship, social service agencies, tenant advocacy groups, and other advocacy groups.
Goals of Community Involvement
Weijer and Emanuel (2000) specify various ways in which communities might be protected in research, including consultation in protocol development, disclosure of information about the research to the community, informed consent by the community before individuals are approached about participating in the research, involvement in the conduct of the research, and involvement in dissemination and publication of results. They argue that there are “particular characteristics of communities…necessary for the implementation of specific protections” (Weijer and Emanuel, 2000, p. 1143). In their view, only cohesive communities with legitimate political authority can make binding decisions to approve or reject a study on behalf of the community. No one set of protections or procedures for community involvement will be appropriate in all situations and types of research. Similarly, others argue that community approval “seems workable only with small groups that have a well-defined leadership structure” (Reilly, 1998, p. 684). However, Foster et al. (1999, p. 1719) argue that “community concerns can be incorporated into existing review mechanisms without necessarily giving communities the power to veto research proposals.” Foster provides two examples in which broadly based consultation in communities that lacked political organization and established moral leaders identified risks in the research that were perceived as important in the communities. Moreover, the researchers were able to address those concerns by revising the protocols. Foster suggests that the aim of community discussions should be “to identify possible risks and concerns, not to formally approve or disapprove individual research protocols” (Foster et al., 1999, p. 1723).
Although the above analyses of the challenges in community involvement in research were based on genetics research, much of the analysis is pertinent to research on housing health hazards involving children. Testimony of community representatives before the committee, as well as the experiences of Committee members who carry out such research, confirm that community discussions can identify concerns and risks that researchers (and IRBs) would not have identified or weighed as much. Although community representatives need not be given decision-making power over protocols, and community consensus is not a precondition to conduct research, it
is possible for community concerns to be addressed by researchers in ways that do not compromise, and may enhance, the scientific value of a study. Indeed, responding to community concerns may reduce the risks of the study and enhance its benefits. These kinds of discussions can be carried out in many ways, depending on the project, the community, and the relationship between the researchers and the community. However, the level of risk and the potential for direct benefit to the child subjects should be considered when determining how to involve the community, with projects that present more than minimal risk and no prospect of direct benefit expected to engage the community in a meaningful way, consistent with the principles of community-based participatory research (see Chapter 8 for a discussion of research that presents more than minimal risk and no prospect of direct benefit). In all cases, however, the committee believes that the most important issue is to ensure that the community is informed, their concerns about the research are articulated and heard by the researchers, and that researchers respond to those concerns.
Community involvement requires infrastructure and the investment of researchers’ and residents’ time, which are not typically supported by research grants. Because grants usually do not cover the true effort required in community involvement, researchers and community partners often have to “donate” considerable time, as well as resources, for travel and related expenses. At a minimum, community residents or organizations involved with a research project should be reimbursed for expenses; ideally, they would also be compensated for their time. However, researchers have to balance the need to provide appropriate compensation with the possibility that the level of compensation will affect, or be perceived as affecting, individuals’ or organizations’ views about the research. The appropriate level and method of payment will depend on the specific project. Several authors have advocated for funders to take a more active role in supporting and encouraging community involvement in research (Green and Mercer, 2001; Minkler et al., 2003; Israel et al., 2005).
Policies at research institutions may make it difficult for researchers to provide tangible support to community partners. Overhead charges imposed by research institutions on subcontracts to community groups may be substantial and may discourage full partnerships. Plans to hire research staff from the community may conflict with institutional human resource policies, which may require traditional job descriptions and educational degrees. However, many community-based organizations and social services groups employ staff from the community with diverse skills and professional backgrounds in public health, urban planning, social work, and
health education. The Columbia Center for Children’s Environmental Health contracts with a community-based environmental justice organization to provide specific outreach and organizing efforts in order to translate center findings into policy and educational interventions (Peggy Shepard, personal communication, November 2004).
Different communication styles may make partnerships difficult, even when stakeholders can overcome a history of mistrust. In some communities, English may not be the predominant language, and few researchers may speak the prevalent language in the community. Even if translators are available and documents are translated, residents who do not speak English fluently may not be able to participate actively in meetings. Other aspects of communication style may also present unintentional barriers to full participation by community partners. Researchers commonly use e-mail to communicate, but community partners may prefer face-to-face communication or may lack access to the Internet.
The characteristics of the community, the kinds of resources available, and the history of community organizing will have implications for the type of involvement in research that is feasible and appropriate. There will be times when researchers have to make decisions to retain the scientific rigor of their project over feedback from the community. A single set of guidelines for community participation will not be appropriate for the disparate types of communities that may be involved in different types of research projects (Israel et al., 2003; Weijer and Emanuel, 2000). In several research projects on housing health hazards involving children, community participation has been incorporated in the planning, design, and implementation of the research project. However, no single model of community involvement will work in all situations and in all types of research (Israel et al., 2005; Minkler and Wallerstein, 2003; Weijer and Emanuel, 2000; Israel et al., 1998). A number of commentators argue that the characteristics of community-based participatory research are an ideal but may not be attainable in every case (Israel et al., 1998; Reilly and Page, 1998).
CONCLUSIONS AND RECOMMENDATIONS
Community involvement in research on housing health hazards with children has the potential to lead to greater understanding of community perspectives of the risk and benefits of research, improve informed consent, increase study enrollment, enhance data validity and quality, and build trust for research. When involving communities in the design or conduct of research, researchers must resolve questions about who represents communities and their subgroups, how conflicting viewpoints in the community are voiced, and whether the process for selecting community partners is appropriate.
Community involvement, though time and resource intensive, is a necessary and useful component of housing health hazard research with the potential to enhance trust and increase the relevance of research to affected communities. Thus, attention to the issues raised by the community and consideration of the most appropriate method of community involvement for a given research project is warranted. As funders and researchers devote more attention to community involvement, the mechanisms and effective approaches to community involvement will become further refined.
Recommendation 5.1: Researchers carrying out research on housing health hazards involving children should describe in their protocols and IRB submissions how they have involved and will continue to involve the affected community in the research project, justify the lack of such involvement, and report how they have responded to any community concerns.
Recommendation 5.2: Institutional review boards should require appropriate community involvement in housing health hazards research involving children and require that investigators’ protocols are responsive to any community concerns.
Recommendation 5.3: Federal agencies (e.g., U.S. Department of Housing and Urban Development, the Environmental Protection Agency, National Institutes of Health, Centers for Disease Control and Prevention), private foundations, and other funders of research on housing health hazards involving children should require researchers to have appropriate community involvement in the research. Funders should provide adequate funding to involve affected communities and should sponsor research to evaluate the outcomes of community involvement.