Collaborative and participatory approaches to research have received considerable attention as methods to better understand and address disparities in health and health care. Partnerships formed to undertake community-based research often include investigators from one academic institution working collaboratively with one or more community-based organizations. When several academic institutions are geographically close and/or when investigators from various institutions are working with the same communities, partnering to address the community's identified health needs can be an effective way to reduce redundancy, increase the impact of research, and improve the efficient use of resources. However, institutional differences in cultures, processes, and policies add additional layers of complexity to the process of academic-community partnership development. In this article, we describe the process of building a three-way partnership between two academic health centers—Duke University and the University of North Carolina (UNC)—and members of the Latino community in North Carolina to develop and pilot test the program Amigas Latinas Motivando el Alma (ALMA), or Latina Friends Motivating the Soul; the acronym works especially well because the Spanish word for soul is alma. We drew on the principles of participatory research, collaborative inquiry, and community-partnered research in the development of this partnership. The ALMA program was developed as a response to the needs, as expressed by newly immigrated Latinos in central North Carolina for capacity-building in mental health and for developing coping skills. The pilot program is an innovative, interactive curriculum that uses adult learning theory to teach coping skills and strategies to Latinas in order to confront and alleviate subclinical depression and anxiety in their community. Our partnership involved pilot testing and evaluating the lay health advisor program targeted for Latinas (results will appear in future publications).
Latino Immigrants in North Carolina
North Carolina has one of the fastest growing Latino immigrant communities in the United States and is representative of new trends of immigrant settlement in the South and Midwest regions of the country.1–3 The large influx of Latino residents began in the 1990s, mainly due to the then-robust economy and demand for labor in North Carolina.4,5 North Carolina's Latino population is now estimated at more than 600,000, constituting 7% of the state's total population.4 The majority (73%) of the Latino population in North Carolina is foreign-born.6 Between 1995 and 2004, over 38% of Latino immigrants came to North Carolina directly from their country of origin, and of this group, 73% came from Mexico.4 Other common countries of origin include El Salvador, Guatemala, and Honduras. According to some estimates, nearly half of North Carolina's Latino residents do not have legal authorization to reside in the United States.4
Latino immigrants typically emigrate in search of economic opportunities, given the weak economies in Mexico and Central America.1,7 Because many Latino immigrants come to the United States to work, they have a younger age distribution than the non-Latino population; 55% of North Carolina Latinos are between the ages of 18 and 44.4 Few Latino immigrants in North Carolina have more than a high school education.1 Latinos in new settlement areas, such as North Carolina, are more likely to have limited English proficiency than those residing in more established Latino immigrant communities elsewhere in the United States.1 Over half of North Carolina's Latino population have completed less than eight years of schooling.4 In part, because of low education attainment, Latino households earn less per capita than non-Latino households, and poverty rates are higher among Latinos (26%) than non-Latinos (15%) in the state.4 These social and economic realities are among the major drivers of the mental health needs and coping challenges (e.g., inability to access available mental health resources and limited social structures to support new residents) faced by new Latino immigrants in North Carolina.
Duke-UNC Health Disparities Collaboration
ALMA grew out of the joint Health Disparities Collaboration, which began as part of a larger interinstitutional effort. Hoping to spur greater cooperation between two nearby, and often rival, universities, the North Carolina GlaxoSmithKline Foundation of Research Triangle Park, North Carolina, provided the challenge, impetus, and resources for this collaborative effort through a grant to both institutions in 2005. The grant supported four projects, each directed by a principal investigator from UNC paired with a principal investigator from Duke. The goal of one of these four projects was to address health disparities. The eight project leaders received no direction as to the population they were to serve, their projects' geographic reach, or even which research topic they were to address. These co-principal investigators could choose service or research, or they could undertake a training program. They had the task of fashioning, with limited resources, a project that they could complete within a three-year time frame. An initial dinner meeting provided an introduction for the teams. The principal investigators for all the projects except the Health Disparities Collaboration already knew one another and had working relationships.
During this first dinner meeting, we, the principal investigators of the Health Disparities Collaboration (G.C.S., S.Y.), described our work to date and shared our approaches to working with populations facing health disparities. The commonality of our approach—to engage potential populations and communities to ask what would be both welcome and culturally appropriate—opened the door to a series of informal inquiries with communities within a 75-mile radius of Duke/UNC to ask about needs and, where indicated, approaches that might be effective. At this first meeting, we discussed communities where we each already enjoyed strong relationships with the local population. Later, we (G.C.S., S.Y., M.G., T.S.) contacted local residents who had worked with us in community research or service programs to inquire about unaddressed needs. We also contacted our colleagues at Duke or UNC who were engaged in community-based work or who had expertise with specific populations that face health disparities. We asked which local programs were successful, and why, and whether they could benefit from an additional service or research effort. We also asked other local and university contacts which issues were affecting potential project communities, recognizing that a project may be welcome but that local conditions may make initiating and sustaining the program untenable. We then asked these contacts for further introductions to local informal or formal leaders so that we could ask those leaders not only about the health issues affecting their neighbors and families but also whether, in their opinions, their communities might want to help us craft a program to address community-identified needs. Given the reach of both universities into local communities, we also inquired about whether a community-engaged research project might be well-received in nearby communities which had previously hosted research projects associated with one or both institutions. After each round of inquiries, we came back together to compare and discuss our field reports, the data we gained on vulnerable health populations in North Carolina and the Research Triangle region, and our own prior or ongoing work with minority and disadvantaged populations in area communities. The structure of these meetings varied from formal meetings with several contacts held at one of the institutions or at a community setting to less formal one-on-one meetings or telephone calls.
Principles of Collaboration
In defining the scope and focus of the health disparities research project, we relied on the principles of participatory research, collaborative inquiry, and community-partnered research.8 Framing our collaboration in this manner allowed us consistency both in our interinstitutional working relationships and in our academic-community partnership. Participatory methods, collaborative inquiry, community-partnered research, and other research partnerships that use participatory approaches to research have the common goal of increasing the value and impact of the research product.9 These approaches to research acknowledge the unique strengths and insights that each community and academic partner brings to framing health problems and designing solutions.10–12 Participatory methods assure community members a voice and a role in program development. Researchers conducting participatory research consider and include cultural and community context within the research.13 Key elements in participatory research partnerships are joint definition of the health condition to be studied, shared learning about issues highlighted as concerns by the community, shared decision making, mutual ownership of the products and processes of research, and commitment to sustainability and action for social change.9,14
Although we were investigators from different institutions, we found we shared these principles in our approaches to academic-community collaborations, and we employed these principles not only in our work with our community partners but also in the development of our own interinstitutional academic partnership. As collaborators, the power of the academic partnership we could bring to bear was evident from the project's onset, as together we called on local partners, other university collaborators, and mental health professionals working with Latino patients to support us in defining the challenges to address and potential solutions to these challenges. Each partner was able to open different doors in communities as well as different doors within her own institution. Between us, we discovered a large cadre of community and academic partners. Although we initially expected to work in more geographically distant communities on this new venture, the promise of the project that arose from these initial inquiries had too much potential value to neglect communities closest to our institutions. The decision to test our new partnership in communities where we each had vested relationships and services was a testament to the growing trust between the co-principal investigators (G.C.S. and S.Y.) and their teams (M.L., M.G., I.O., T.S., G.P., C.B.).
Identifying a Health Topic of Interest
Based on available epidemiologic data on the health and health care needs expressed by local lay leaders and Latino community organizations in the target region, and on our prior research and service experiences, we chose to work with the emerging Latino community near our universities in North Carolina. Through the informal focus groups and individual meetings that we convened with formal and lay community leaders, we learned that although many Latinos suffered from depression, social isolation, and anxiety, their mental health needs were largely unmet. We learned that trauma crossing the border (e.g., rape, child abuse, robbery), sadness in leaving strong, intergenerational family systems, and fear fed by the hostility toward undocumented immigrants in the United States all reinforced the population's cultural habit of suppressing expressions of mental health needs. The lack of culturally competent, Spanish-language mental health services for the largely uninsured Latino population in North Carolina and the Research Triangle region, coupled with cultural norms that discourage sharing personal problems outside the family, served only to deepen the isolation, anxiety, and depression that formal and lay Latino community members described. Community leaders in our local area indicated that their members would welcome mental health training and support, especially if this training and support were community based. The community leaders also offered that women were more likely than men to accept mental health services.
Conducting community-based research in mental health coping skills among Latinas was completely new work for both the UNC and Duke research teams. To determine whether this was appropriate or feasible, we convened a meeting with Latino-serving-agencies staff and Latino community advocates, as well as meetings with colleagues from both academic institutions. These community and university contacts, who were either known to one of us or referred by a local contact, included a health program director from a local, community-based service agency for Latinos in Durham; the director of a care management group that provided access to medical care and community resources to Latino clients in Durham; and the coordinator for the Chatham Health Initiative, a group working to increase access to medical care at the Chatham County Hospital.
The mental health professionals, both academic and community, urged us to undertake a project to increase self-care coping skills for Latinas. These professionals agreed that we would need a mental health professional to work directly with the participants; this professional would also provide appropriate mental health backup or expertise, screen prospective study participants, review intervention materials, and, if needed, give referrals to patients or direct them to appropriate mental health services. This original group of academic and community mental health leaders also suggested both (1) including the Latinas as study participants in order to test the effectiveness of the training curricula that we created and used to teach them, and (2) using the participants' training experiences to build their skills in order to develop a team of promotoras who could offer outreach to others. On the basis of their advice, we decided to build our program based on a lay health advisor model (described below). In the end, we chose to develop a community-based training program around mental health and coping skills for promotoras, women who have formal or informal leadership positions within their communities and who are in a position to reach out to their Latina peers.
Despite the fact that the field of mental health research was new to both the Duke and UNC teams, we recognized that we had strong local relationships in both communities and that each university's history of providing health services and research to minority and underserved populations was a strategic asset. Duke's Division of Community Health had been operating neighborhood and school-based community clinics in Durham that provide services mainly for indigent patients, including a large population of uninsured Latino patients. In addition, Duke's Division of Community Health had been operating a free health navigation / advocacy / patient support program, Local Access to Coordinated Healthcare (LATCH), since 2002. LATCH had an enrollment of over 12,500 uninsured Durham residents at the onset of ALMA, 96% of whom were Latinos without health insurance.15 To build and sustain LATCH, Duke built strong relationships with Latino-serving organizations throughout Durham County. Likewise, UNC had built strong relationships within Chatham County over the decades preceding the ALMA project. UNC's involvement in Chatham County included lay health advisor programs employing Latinos and community-based research programs. These previous partnerships afforded a base of UNC providers and researchers who could guide the Health Disparities Collaborative team in developing a place in the community for ALMA. Team members from both schools had served patient populations clinically and were able to draw on these experiences to craft the ALMA project.
Engaging Community and Academic Advisors
One of our first tasks was to convene a group of advisors to guide and support our efforts. As we met with community organizations and advocates and contacted academic colleagues engaged with North Carolina Latinos and with mental health, we brought their advice and experience back to the team. The group then determined whether the skills and experience of each contact were important to our project's success. Initially, we created both an Academic Advisory Committee (AAC) and a Community Advisory Committee (CAC).
We were intent on building a cadre of mental health professionals in the AAC who provided service to this population and who could advise us on how to provide mental health prevention and support services in a community context. Because mental health was a new area of research for members of our team, we wanted to ensure that we would have the active engagement of a number of mental health experts who were well versed in Latino mental health needs and cultural issues. We also sought to engage AAC members who had working knowledge of the communities under consideration—Durham and Chatham counties. The 13 members of the AAC included experts from both Duke and UNC and represented the Department of Psychiatry, the Department of Medical Psychiatry, the Department of Medicine, the UNC School of Public Health, the UNC School of Nursing, the UNC Center for Global Initiatives, and Duke Hospital. In addition, the AAC's members represented local organizations, including Auger Communications and Wake County Human Services. Finally, an artist and visiting professor from Guilford College and a self-employed contractor were also members.
As mental health professionals constituted the AAC, community experts composed the CAC. In building our CAC, we looked to those agencies working directly in the communities we were considering, agencies that had been recommended by either academic colleagues or, more importantly, Latino community advocates. The 13-member CAC included representatives from Catholic Charities, area county health departments, local nonprofit Latino service organizations, local Latino advocacy organizations, and a local nonprofit Latino mental health agency. Several of the community agency representatives had previous experience with lay health educator or peer educator programs. In addition, all of the community representatives, having worked with the local Latino population, enjoyed previously established relationships with the populations we wanted to serve and acted as gatekeepers in our target areas, providing entree and introductions for the ALMA program.
Initially, the CAC and AAC met separately to offer advice and guide the intervention; however, we soon realized that having the groups meet jointly was advantageous. Representatives from academia and the community were then able to hear one another's concerns, which led to more synergy within the group and more integrated feedback to program staff.
The members of the AAC and CAC have been continually and actively involved in all phases of the development, implementation, and evaluation of ALMA. During and following the first joint advisory committee meeting, members of our robust group of experts provided guidance and direction on specific areas of the curriculum and promotora recruitment and screening. We asked for input into the curriculum, and from the first joint AAC/CAC meeting, we received extensive input and creative ideas. For example, the aforementioned artist, an AAC member who had authored books documenting migration journeys for Latinos, helped us develop components of the curriculum that emphasized active learning using visual, kinesthetic, affective, and cognitive learning styles. In addition, CAC members assisted in the implementation of the sessions by donating meeting space to host ALMA training sessions, helping to secure culturally sensitive materials, cofacilitating sessions focused on their areas of expertise, and recruiting study participants. Findings from the project, even in its earliest form, were presented to the joint AAC/CAC. Together, the group reviewed the findings and engaged in problem solving.
Determining Program Goals and Objectives
Prior research has demonstrated the efficacy of peer education via lay health advisors.13,16–24 In addition to the evidence in support of this approach, one of our guiding principles was the development of a program that would be sustainable even after funding for the project ended. Our aim in this initial collaboration was to test the feasibility of recruiting and training promotoras who would be able to provide mental health services, teach coping skills, and share stress-reduction strategies with their peers. The promotoras' skills would continue to be of service to their communities at the culmination of the intervention. We also hoped to describe the impact of the promotoras on their peers' (1) knowledge of area mental health services, (2) communication about emotional health with health providers, and (3) decrease in depressive symptoms, anxiety, and improvement in coping skills.
Promotoras participated in a 10-week training program, comprising 10 modules each lasting about two hours and each attended by 12 to 30 women. Members of the CAC and other local experts with experience teaching in the Latino community cofacilitated specific segments of the training (e.g., domestic violence and navigating the school system). Some of the sessions were explicitly aimed at increasing immigrants' knowledge of area resources (i.e., health facilities, school policies) to support their needs. Most sessions began with a vivencia: a thematic story, video, or song to open dialogue about emotions and how they relate to stress and coping. One or more experiential group activities per session built listening skills, awareness of community resources, and additional stress management techniques. During training, promotoras received a comfort basket containing tangible items for reinforcing healthy coping strategies: a journal, aromatherapy candle, body lotion, bath wash, and meditation CDs. All meetings closed with a family-style meal, which afforded opportunities for bonding and fellowship. Weekly homework assignments provided opportunities for promotoras to practice their coping skills through personal expression and creativity. The ALMA curriculum is rich in hands-on, skill-building activities meant to offer techniques for navigating life as an immigrant in the United States.
Negotiating Administrative Structures and Learning Institutional Cultures
Perhaps the most intriguing, and sometimes entertaining—or frustrating—part of the ALMA project was working with and navigating two different administrative structures and academic cultures. Working across two different institutions, the team often found administrative or policy areas where one university had rigid rules, while the other university permitted latitude. Each institution had administrative policies that were easier or more difficult to navigate, and we often found humor in the disparity between the two approaches. Generally, after discovering these differences, we determined which institution's policies were less amenable to change and then adapted our approach to meet that more rigid university's criteria. Table 1 lists some key areas in which the teams found differences between the two universities, and it delineates the collaborative solutions we reached in order to move the project forward.
One unexpected hurdle was the distribution of funding. Originally, before the budget was finalized, grant dollars were evenly allocated between Duke and UNC; however, as the work began, the ALMA team determined that Duke would hire project personnel and pay the promotora stipends as Duke had more flexibility in incentives for participants and was able to create new positions more rapidly. Funds were redistributed accordingly, but we made all hiring decisions jointly (e.g., during the recruitment for the ALMA licensed clinical social worker, faculty and staff from both UNC and Duke conducted group and individual interviews with prospective applicants).
Another interinstitutional challenge we faced was ensuring institutional review board (IRB) approval at both institutions. IRB forms and procedures at UNC and Duke differed, as did the level of completeness and areas of scrutiny required before a program was initiated. We had to develop an informed consent process that not only met both institutions' IRB requirements but also remained feasible for use with research participants who were documented and undocumented immigrants whose privacy and safety needed to be maintained. The IRB process took nearly five months (significantly more time than we anticipated), in part because of institutional differences in UNC's and Duke's requirements. For example, Duke's risk management office became involved earlier and more directly than did UNC's during project development. In addition, the use of one another's forms also made the IRB process more difficult at times.
Another institutional difference is the process of teaching and training faculty, staff, and students before they work in communities as representatives of either academic center. Duke has a highly developed training system that is designed to ensure rigor and uniformity in content. Whether for clinical service activities, such as medical screening at a health fair, or for health education events, Duke University Health System faculty, staff, and students must complete basic online training and pass an online assessment regarding expectations of conduct and cultural competency while working or volunteering in community settings. These events or activities themselves receive formal advance review and approval that are linked to risk management and secure coverage for the event. Further, at Duke, separate online IRB training is required for conducting human subjects research, and a module on conducting community-engaged research is offered through Duke's online IRB training. UNC's approach is less comprehensive in scope, focusing solely on community-based research as a subset of other human subjects research. Interestingly, we found that neither system adequately prepares trainees for the nuances of negotiating interactions with vulnerable populations. Because of challenging interactions with trainees from both institutions, we chose an approach that incorporates greater oversight of and feedback for any trainee engaged in the ALMA project.
Contextual Challenges to Partnership Development
Some of the challenges to partnership development existed in the larger context of county and state politics and policies. The ALMA project was conceived at a time when the wheels fell off the mental health care system in North Carolina. As mental health reform in North Carolina failed, there were (and still are) serious repercussions for patients across the state and for state and local program leaders. When North Carolina began implementing its mental health, developmental disabilities, and substance abuse service system reform in 2001, local area programs and county mental health agencies became managers of services rather than direct providers of care for low-income residents. The local mental health authorities were given responsibility for managing finances, authorizing services, and contracting with direct service providers. While the local, public mental health agencies lost their capacity to directly provide services, ALMA was seeking to work collaboratively with the population to develop a local prevention and early intervention strategy that would (1) build coping skills and resiliency among community members and (2) address language barriers as well as the cultural perceptions of discussing mental health issues. Although ALMA was a pilot project, the stakes were high because the area had limited mental health services for patients who were not Medicaid beneficiaries or who did not have severe mental health needs. Within the local (and national) environment, where debate over immigration raged, the capacity to develop sustainable, self-help, community-embedded services became more urgent.
In addition, race relations in central North Carolina overall became strained as a result of the recent growth in the Latino population.25 Beginning in the late 1990s, some residents in Kirby County voiced their opinions about an increase in criminal activity and overcrowding in schools and clinics, both of which they blamed on the growing number of undocumented Latino immigrants. The chief county commissioner responded to community concerns by writing a letter to the Immigration and Naturalization Service asking for greater enforcement in the county. At the height of the tension, David Duke, former Louisiana state legislator and grand wizard of the Ku Klux Klan, was invited to speak at a rally of county residents who claimed their community was being “overrun” by undocumented immigrants from Mexico.26 The social and political environment created by these events effectively legitimized discrimination against Latinos, and the area saw an increase in open hostility toward the newcomers, increasing the need for mental health services and making the success of the ALMA program even more important.25 Ethnographic studies of other new immigrant destinations in the South have noted antiimmigrant sentiment and racial discrimination toward Latino immigrants, similar to those in North Carolina.7,27
Our interinstitutional and community partnership, bolstered by our advisors and partners in academia and the community, allowed us to stretch boundaries. Although differing institutional cultures and policies had the potential to create conflict, we applied the differences in institutional approaches to benefit the partnership. We were able to leverage the strengths of one institution to overcome weaknesses or obstacles in another, viewing the two institutions ultimately as complementary rather than competitive. The institutional barriers and operating roadblocks we met were hardly insurmountable, and we took the approach of interested learners as we dealt with each institution's requirements and gatekeepers. In many ways, having the partnership tested during its formative phase, and working under time pressures to deliver a research project, further strengthened the team. The lessons we learned from what we taught the promotoras in ALMA were lessons we applied to ourselves: We listened to one another, supported one another during frustrating points in the process, found ways to celebrate together, were open and respectful in our disagreements, were quick to share credit and accept blame, and were open about problems we encountered. Our regular team debriefings, in person or by conference call, leading up to and following every training session, became a place where we could explore problems, solutions, and new ideas. These were, indeed, some of the skills we were hoping to build in our study participants.
Working with a socially, politically, and economically vulnerable population, our attention was, and had to be, directed toward protecting the safety and privacy of our study participants. Interacting with the ALMA study participants, in as intimate an arena as mental health, with sessions conducted in Spanish on weekends in noninstitutional settings, required us to rely on one another's judgment and to communicate effectively, especially when confronted with unexpected situations and adversity. We used the trust we had earned when difficult situations did arise, such as when, outside of the training program, money collected by one promotora for a group event was lost or stolen. This type of situation required not just communication between the two teams but also a joint decision on how to handle the potential breach of trust within the group. In this situation, we reached out to our CAC and asked a licensed clinical social worker, who had spoken in one of the sessions, for advice. She advised confronting the issue even though the problem did not take place in the training program, and she successfully facilitated a session devoted to addressing the lost money and to developing and communicating expectations within the group.
The investment of our community partners, the strong interpersonal relationships built in preparing this project, and the confidence of our institutional leaders in appointing us as leaders of this project helped sustain our patience and good will. Together, we learned to
* Reach out to colleagues and community lay and formal leaders before settling on a challenge to address; we did this early on by discussing project possibilities with colleagues who worked with the community through Duke's LATCH program, as well as with local Latino faith and community leaders. They, and others, helped us identify mental health as a pressing Latino health issue.
* Be respectful of the burden of research on a community, most particularly a vulnerable community, and ascertain whether university involvement is desired, not simply tolerated. Our study team held initial, informal project meetings with community gatekeepers to ensure that this research project was truly acceptable to the community.
* Convene people from the population we hoped to serve and ask them not only to discuss their needs but also to describe the types of interventions they would find helpful and think would be successful.
* Keep community members close throughout the process and rely on their wisdom and guidance when confronted with difficult situations in the intervention.
* Agree to use “straight talk” among the university partners. The principal investigators (G.C.S., S.Y.) agreed in advance to work out differences between themselves and to jointly air differences of opinion regarding directions and tactics with their research team.
* View the partnership as a learning collaborative and assume university partners are working from honorable motives. Though the learning curve might have, at times, slowed the progress of the project, it made for richer learning experiences and greater bonds between the Duke-UNC team members.
* Expect that the way the investigators and staff treat one another will be reflected in the way the study participants relate to one another and to the project. In ALMA, the friendly and collaborative nature of the Duke-UNC study team spilled over to establish a positive environment at the training sessions. The community participants in the ALMA project were effusive in their comments about the staff and the project, expressing emotion, affection, and gratitude during training sessions and graduation celebrations. Promotoras became a social network for one another, supporting each other through their joys (pregnancies and new jobs) and sorrows (family tragedies and immigration raids).
* Bring partners together on a regular basis to share the data and products from the project, to hear the tough and exciting parts of the work to date, and to reflect together on the strengths and weaknesses of the work.
* Find ways to thank those who work with the project: community organizations, advocates, lay leaders, formal community leaders, institutional partners, and one another.
In the process of building this three-way academic-community partnership we found that relying on shared principles and values of community engagement facilitated crossing our own institutional boundaries. These principles helped us successfully address mental health needs in the Latino community. Explicitly setting forth these principles and giving attention to process has been important in other academic-community partnerships; so, too, we found this approach vital for identifying and navigating cross-institutional differences. This deliberate approach not only created a strong, successful collaboration between academic institutions and researchers but, more importantly, became the foundation of a productive partnership with members of the Latino community.
“ALMA—Amigas Latinas Motivando el Alma” or “Latina Friends Motivating the Soul” is a Duke University-University of North Carolina- GlaxoSmithKline Health Disparities Initiative. It is supported by a grant from the North Carolina GlaxoSmithKline Foundation of Research Triangle Park, North Carolina.
This research was also supported by grants from the National Center for Research Resources, a component of the National Institutes of Health (NIH) (UL1RR024128 and UL1RR0257470), and the NIH Roadmap for Medical Research.
The research presented here has been presented previously at multiple national scholarly meetings and conferences.
The research presented here is solely the responsibility of the authors and does not necessarily represent the official views of the National Center for Research Resources or the National Institutes of Health.