Community Partners in Care: Leveraging Community Diversity to Improve Depression Care for Underserved Populations

Abstract

Research suggests that the quality and outcomes of depression treatment for adults can be substantially improved through “collaborative care” programs. However, there is a lack of resources required to implement such programs in vulnerable communities. Our paper examines the planning phase of the Community Partners in Care (CPIC) initiative, which addresses this problem through a unique approach in which academic institutions partner directly with a wide range of community-based and service organizations in all phases of the project fielded in two underserved communities in Los Angeles. CPIC offers a unique opportunity to understand how diverse organizations can work together to address community depression care needs and to analyze the potential strengths and tradeoffs of coordinating among such varied entities. This article focuses on intra-group dynamics that surround the process of participatory research and reports results of the first wave of process evaluation of the planning phase of the CPIC initiative. Our analysis explores two main themes: Community-Partnered Participatory Research and benefits and challenges of collaboration in diverse groups.

Introduction

Research suggests that the quality and outcomes of depression treatment for adults can be substantially improved through “collaborative care” programs, which educate and motivate both patients and providers to improve care coordination and facilitate access to evidence-based care.^{1, 2} However, the resources available to implement such quality improvement (QI) interventions are limited in public sector agencies that serve vulnerable communities, and comparable programs have not been designed for complementary outreach services by social service and other types of agencies that support these communities.

Our paper examines the planning phase of the Community Partners in Care (CPIC) initiative, which addresses this problem through a unique approach in which academic institutions in the Los Angeles area partner directly with an unusually wide range of community-based and service organizations in all phases of the project. CPIC offers an unprecedented opportunity to understand how diverse organizations collectively learn and work together to address important community needs related to depression care and to analyze the potential tradeoffs between the pooling of diverse community strengths and the challenge of coordinating among such varied entities.

In particular, we focus on intra-group dynamics that surround the process of participatory research and report results of the first wave of process evaluation of the planning phase of the CPIC initiative. Our analysis explores two themes that are directly relevant to the notion of diversity in organizations and communities:

1. Community-Partnered Participatory Research: What are the benefits and challenges associated with building a partnership to engage academics and diverse community organizations in developing a participatory research project that addresses a pressing issue in the community?

2. Collaboration in Diverse Groups: How do community and academic partners perceive the CPIC initiative? What are their experiences in the planning phase of this initiative?

First, we describe Community-Partnered Participatory Research (CPPR) and the main design components of the CPIC initiative. Second, we briefly review the literature on group diversity and group performance as it relates to the CPIC initiative. We then present the main results of process evaluation of the planning phase of the CPIC initiative, which highlight the theme of collaboration in diverse groups. In conclusion, we discuss the implications and lessons learned to date about leveraging diversity in the leadership team responsible for developing a complex, multi-stakeholder depression care QI intervention in the community context.

Community-Partnered Participatory Research

The translation of research into practice is a defining and continuing challenge of implementation science, because its success depends on the extent to which researchers and practitioners can balance internal (i.e., will the intervention work when intervention results are compared to those in control group?) and external validity (i.e., will this intervention work in other settings?) of their interventions.³ Researchers have advocated the use of Community-Based Participatory Research (CBPR) methods to improve the internal and external validity of QI interventions, which have a potential to effectively reduce ethnic disparities in health care.^4-6 The CBPR model suggests that researchers should collaborate with those affected by what they study in all phases of research design, implementation, and dissemination in an attempt to increase the understanding of the problem to be studied, to improve the quality and validity of the research itself, and to develop an effective change strategy that is likely to work in a given setting. In more intense and inclusive forms of CBPR, such close collaborations can lead to the development of “communities of practice,” cross-cutting networks of people who exchange learning and expertise, develop common goals and identities, share best practices, and solve particularly pressing problems affecting their communities.^7-9

Health care research shows that engaging community members and organizations in collaborative processes to design, implement, and disseminate evidence-based health interventions can enhance their relevance and impact for populations in need or at-risk.⁶ CBPR seeks to enable researchers and community members to forge a common interest in a set of health outcomes and to jointly create interventions that are sensitive to community needs and circumstances, thereby helping increase the odds of successful program implementation and sustainability.¹⁰ It calls for the use of action methods, power sharing, and researchers spending time to gain familiarity and build relationships in the community.¹¹ The CPIC initiative follows a variant of CBPR called Community-Partnered Participatory Research (CPPR), which emphasizes consistent and equal participation of community and academic partners at every step, inclusion of the academic participants as part of the community, and rigorous research to guide sustained change.¹²

While the CPPR approach promotes equal power and leadership sharing between community and academic partners, it also builds the capacity for joint planning and community implementation of evidence-based health interventions through the process of community engagement and sharing of resources and expertise.¹³ A CPPR project is typically structured according to the following format:¹⁴

1. Creation of a Steering Council (SC) that consists of community and academic leaders who jointly identify a health issue that fits community and academic priorities and then establish, support, and coordinate project’s efforts.

2. Formation of various workgroups that develop, implement, and evaluate the impact of action plans, which are periodically reviewed with the SC members.

3. Integration of the workgroups’ action plans into a unified strategy that reflects the interests and preferences of all partners and stakeholders.

4. Collaborative analyses of project data and progress.

5. Dissemination of products and findings to a wide range of audiences, ranging from local communities to national and international academic and policy stakeholders.

The CPPR approach moves beyond inviting academics to consult on research design issues or inviting community members to serve as advisors on projects and stresses that all phases of the research project should be co-led by academic and community partners. Such an approach emphasizes the value of and respect for academic and community input and allows for a more open sharing and a better understanding of diverse perspectives that project members are bringing to the table. The CPPR model benefits both academic and community partners in that it allows them to build capacity: while academic partners develop a better understanding of the community, community partners develop research skills and learn how to apply research findings to solve problems that affect their communities.¹⁴ At the same time, however, having a wide range of perspectives, interests, and personal experiences makes it more difficult for diverse groups to come up with a single decision that is acceptable to everyone. Establishing collaboration in a group where leadership is always shared among academic and community partners may take more time than in situations where either academic or community partners have the lead role while the other party only plays a consulting role.

The Community Partners in Care Initiative

Research has long shown that the quality and outcomes of depression treatment can be substantially improved for adults through collaborative evidence-based QI programs that integrate mental health treatment and support services into primary medical care settings.^{1, 2} Nonetheless, QI interventions typically reach only a fraction of eligible patients and require resources that are often limited in public-sector agencies that serve vulnerable communities. Consequently, such programs are underutilized in community-based health care settings, because they are rarely adopted by community agencies, such as social service and faith-based organizations and primary and specialty care providers. The CPPR model, however, provides a framework for respecting both community wisdom and academic expertise in the search for creative solutions to support disease management for depression in underserved communities.¹⁵

The Community Partners in Care (CPIC) study follows the CPPR model. It is headed by a SC of lead community and academic agencies that adhere to a set of participatory principles in developing, implementing, and evaluating all phases of the project. CPIC is the first group-level, randomized controlled “encouragement” trial that is intended to demonstrate the effectiveness of fully engaging community members in disseminating evidence-based healthcare and QI toolkits for depression care in underserved communities. Fielded in two underserved communities in Los Angeles (Hollywood and Metro South Los Angeles), this initiative compares two dissemination approaches to promoting the adoption and use of evidence-based QI toolkits for depression care: a Resources for Services (R4S) condition, which is a typical low-intensity strategy that provides initial training on the toolkits through a conference and limited technical assistance throughout the study, and a novel “community engagement” (CE) condition, which is a network development strategy that additionally provides training and support for a wide-ranging set of agencies (i.e., primary medical and mental health specialty clinics, substance abuse agencies, faith-based organizations, and parks and recreations programs) to collectively plan and commit to sharing resources and responsibilities for depression care.

By comparing the CE and the R4S conditions, CPIC contrasts what happens when diverse organizations are supported in “building a village”¹⁶ that surrounds community members with opportunities to learn about and engage in evidence-based depression care with what happens when similar organizations receive training, orientation, and access to documentation that supports use of the same toolkits. The study hypothesizes that the CE approach will be more likely to result in an engaged “community of practice” of agencies and providers around the depression care intervention ^17-19. The idea is that the participatory engagement and network development aspects of the CE intervention will stimulate sharing of resources and new local solutions that facilitate access to quality improvement programs and treatments across the community. The CE condition may increase agency and clinician use of the programs across the network and thereby improve client access to appropriate care, satisfaction with services, and health outcomes. Likewise, the CE approach is also expected to increase community trust in and sustained commitment to collaborative management of depression.

In this paper, we examine the initial stages of forming these “communities of practice” around depression care improvement by focusing on the CPIC SC and the experiences of its members in planning the wider community demonstration trial. Our analysis highlights the benefits and challenges of bringing together an unusually broad assortment of agencies that often do not have any direct experience in providing depression-related services (i.e., churches, parks and recreation facilities) to plan a large-scale, complex intervention using a collaborative approach to decision-making, where community and academic partners work together to make all decisions jointly.

Diversity within the CPIC Initiative

As described above, CPIC not only engages an unusually wide range of agencies to build partnerships, pool existing resources, and collaborate with each other to solve a pressing mental health problem in underserved communities, but also has a very diverse leadership team - the SC, which consists of:

• Academics from University of California - Los Angeles, RAND Corporation, and University of Southern California, including physicians, psychiatrists, psychologists, statisticians, sociologists, and anthropologists, among others;

• Representatives of grassroots and advocacy organizations for organizing around local community health issues (i.e., Healthy African American Families); for improving the economic self-reliance, civil rights, and empowerment of minority groups (i.e., Los Angeles Urban League); and for representing the interests of individuals with mental illness and their families (National Alliance for the Mentally Ill – Urban Los Angeles)

• Representatives of faith-based organizations that provide healthcare for uninsured and low-income individuals (i.e., the QueensCare Health and Faith Partnership);

• Representatives of non-profit community-based healthcare organizations that provide substance abuse, mental health, and related services (i.e., Behavioral Health Services, Los Angeles County Department of Mental Health);

Diversity, as a group level phenomenon, refers to differences that exist between group members on any attribute that may potentially lead to the perception that one group member is somehow different from another.²⁰ While talking about group diversity, we consider not only observable or demographic characteristics of the SC members (social category diversity), but also their value- and attitude-based differences (value diversity), as well as experience- and knowledge-based differences (informational/functional diversity), both of which are not so easily observable.^20-22 Research on group diversity yields contradicting results. Diverse groups can draw on more varied sources of information and have the potential to generate innovative ideas. Nonetheless, heterogeneity in groups is also associated with interpersonal conflicts and inefficiency in terms of time required to make a decision.^{23, 24}

Social psychologists have consistently found that people tend to prefer similarity over difference in their interactions with others; a similar finding in sociology is termed homophily.²⁵ People categorize each other either into in-group (groups that they themselves belong to) or out-group (groups they do not belong to themselves) members, often based on visible demographic characteristics such as race and gender, and tend to trust and like those who belong to their in-groups more.²⁶ Consequently, the more homogeneous a work group is, the more likely it is to be characterized by high levels of group cohesion and commitment and low levels of relationship or personal conflicts.^{20, 22, 27} This line of research seems to suggest that homogeneous groups are likely to be more effective, compared to heterogeneous groups, because of the visible similarity between group members that makes them more likely to trust and rely on each other and reach decisions more quickly and easily.^{23, 28, 29}

Similarly, value congruence or homogeneity in groups is likely to reduce the amount of conflict within the group and therefore make the group more efficient. When group members have similar values, or fundamental beliefs about the desirability of certain behaviors, they are likely to agree on group goals, tasks, and procedures. As such, value congruence protects groups from both personal conflicts and task related conflicts, because group members tend to identify with each other and interpret existing problems in similar ways.^{22, 27} Groups do not need to spend excessive time trying to search for compromises about the most pressing goals or the best way of achieving them.

In contrast, some diversity theorists argue that heterogeneity based on functional knowledge leads to higher quality group decisions, because diverse groups are more likely than their homogeneous counterparts to have different types of knowledge, skills, experiences, and abilities that tend to improve the quality of the group’s decisions. Heterogeneous groups can also generate more innovative solutions to existing problems through thorough discussions between people with diverging, and possibly non-traditional, perspectives.^{20, 30} As noted previously, however, functional diversity in terms of education and knowledge is also associated with intra-group task conflicts, which slow down the decision-making process and cause inefficient use of time in diverse groups.^{22, 27, 30} Therefore, the benefits associated with heterogeneity in groups appear to manifest only if groups are able to manage their internal diversity effectively.²²

Meta analyses of the diversity literature tend to suggest that the impact of group diversity on performance is not consistent; it varies depending on the type of diversity, measures of performance, and kinds of conflict and communication between group members.³¹ Moreover, research on innovation and creativity in groups suggests that while diverse groups have a potential to generate new and creative ideas that benefit from the multiplicity of experiences and backgrounds of their members, heterogeneous groups may have a hard time implementing these ideas, because they typically demonstrate less capacity for teamwork than homogeneous groups. In other words, group diversity may have a positive effect on generating creative ideas, but a negative impact on making decisions and developing action plans.^{23, 24, 32} We explore the applicability of these ideas to the CPPR approach through the results of a self-assessment of the group and participatory processes within the CPIC Steering Council (SC).

Steering Council Self-Assessment Process

In line with the CPIC study’s principles as a CPPR project, the SC developed self-assessment procedures intended to:

• Continually track the participatory and group processes of the SC,

• Provide the SC with opportunities to reflect and improve how it works for all participants, and

• Document lessons learned for conducting this type of community-partnered research.

The self-assessment itself was designed in a participatory manner by academic and community partners in the study’s Implementation Evaluation Workgroup (IEW), with input from the full SC. The mixed methods design utilizing 32 short questionnaire items that were adapted from Schultz et al.’s survey instrument,³³ optional open-ended responses to each questionnaire item, and group feedback and discussion sessions was intended to incorporate advantages of both quantitative and qualitative methods, while also taking into account the limited evaluation resources of the study. SC members filled out the questionnaire online; all community and academic partners had access to a computer and were comfortable with on-line instruments. None reported any technical difficulties.

The first round of self-assessment took place in March-April of 2009. All 28 members (10 community and 18 academic) of the SC were invited to fill out an online questionnaire about their individual experiences and perspectives. The IEW compiled the results of this survey based on the 22 responses received (6 from community and 16 from academic partners) and presented them to the SC during two one-hour sessions. Two academic and two community IEW members shared the responsibility for drafting the preliminary analysis, presenting survey results, and leading the group discussion that accompanied the presentation. The purpose of these group discussions was to stimulate dialogue on issues that arose in the survey in order to facilitate a multifaceted interpretation of the survey results.

The experiences of the IEW survey analysis team showed that it functioned as a microcosm of the SC, reflecting the dual challenges diverse teams face - that of producing high quality results yet operating within time constraints. The team members differed not only in demographic characteristics and values, but in their depth of knowledge and experience with the practice of evaluation. The academic members took the initiative to conduct the data analysis and write up the preliminary report, then invited the community members to share feedback on interpreting and presenting the data. As two of our community members jointly noted regarding academic members: “They gave us a role that we could succeed in.” Although only IEW members collaborated intensively on the preliminary analysis and post-discussion write-up, the evaluation process adhered to CPPR principles in that all SC members had the opportunity to participate in the evaluation process and comment on the final report.

Steering Council Self-Assessment First Round Results

Results of the survey and group discussions revealed the following six themes:

1. Equitable participation and shared roles

The most popular theme in the discussion was the importance of equitable participation and sharing among academic and community partners. Overall, all SC members either agreed or strongly agreed that responsibilities for shaping the vision and overall project objectives are appropriately shared among community and academic partners. However, SC members are still learning how to collaborate and share responsibilities in an effective way at this early stage of the project. As one academic partner noted: “We have been learning how to do this [share leadership roles], and sometimes the way forward is surprising, such as a community leader coming up with a technical design idea, or an academic leader having a sense for how to approach the community during recruitment sessions in the community.” Such an unorthodox approach to decision-making clearly has benefits for creating innovative solutions to existing problems, which have a potential to better implement QI programs intended to help communities cope with depression.

2. Problem-solving and goal development

Survey results showed that all SC members enjoy working together to solve problems, and the vast majority (84%) thinks that there is appropriate sharing of responsibilities for developing goals and tasks. They seem to value differences that exist between members and are willing to benefit from them: “…as different council members enter the project, they have to learn both … the large ideas at [the project’s] core and the way that the Council as a body and as individuals think about the project. Sometimes there are differences across Council members, which should be expected, but that is all part of the shaping, and even differences are opportunities to achieve greater clarity.” While this comment suggests that collaboration between diverse people can help clarify the project’s goals and vision, it also shows that SC members are in the forming phase of group development where they are still defining tasks, learning about expectations, and discovering “ground rules.”^{34, 35}

3. Shared leadership roles

Although all SC members agreed that leadership is shared, one respondent noted that there is a difference between theory and practice: while “leadership roles are shared (or at least offered equally to all), members may not choose to accept the responsibility; so a few people do much more than others.” In the discussion, it was noted how the academic principal investigator of the project actively attempts to engage community partners in leadership roles to ensure that the representation at the SC level is fair and balanced, but often it is difficult for community partners to devote extensive time to this initiative. This theme was particularly important for existing community partners, who were concerned with the fact that not many community partners are willing to take the personal initiative to accept leadership responsibility. The same group of active community partners volunteers to serve on multiple committees, which means that leadership roles are not equally shared among community partners.

One potential explanation for this phenomenon is that the SC is still in its forming stage with new community partners continuing to join the project. As certain community partners become more involved, they may be invited to join the SC. Empirical research on group processes suggests that new group members are expected to go through socialization process first, before they are encouraged to actively engage in the group decision-making and take on leadership roles.³⁰ This also seems to be the case with the CPIC initiative. As one relatively new community partner noted about her experiences with the project, “this is a real steep learning curve for me. I’ve never been involved in anything like this. So while I make comments, I’m still really learning how this whole research piece fits together. There will be a time when, yeah, I would love to step up and take more leadership responsibility, but I still feel I’m in that learning process…I’d like to come and just spend some time with you and learn more about how the whole process develops. I think it’s fascinating. I’m here because I really want to learn more. I think this is a key to a lot of what we do in the community.”

4. Resource sharing

As suggested by another SC member, the notion of leadership sharing is intricately intertwined with the notion of resource sharing. In the survey, participants showed some disagreement about resource distribution: most (43%) had a neutral position, only 1/3 had a positive opinion, and the rest had a negative opinion about resource sharing. Some discussion participants suggested that resources are “not evenly distributed among the community partners: some get more than others.” Group discussion that followed the presentation of survey results revealed that some SC members think that resources should not be “equally” divided; rather they should be distributed proportionally to the extent partner organizations participate in the study. Being an active leader on the project does require additional resources: “Some people are giving enormously to the project without getting much in the way of resources. This is a project where a lot of people are giving volunteer time—maybe more than people are aware of—to the project because they believe in the project, which is really pretty inspiring. But that’s tough on people on the project. So one thing we could decide as a group is, if we need more leadership of certain kinds, can we free up some resources for people that don’t currently have a lot of resources for the leadership roles they’d like to play in the project?…As our group shared leadership style grows, I think we should feel free to redirect the budget, especially at the margins, where we can, to allow that.”

5. Mutual trust and respect

Survey results showed high levels of trust among SC members, with nearly three-quarters of respondents agreeing or agreeing strongly that members of the SC trust each other. In addition, having an honest discussion about resource and leadership distribution within the project at such an early stage of the initiative appeared to help both academic and community partners further reinforce their trust in each other and increase confidence in the group’s ability to build capacity and make a difference in the community. As one community member noted, “It’s funny for me to see how we are starting to work together. It’s also helpful to know that some of the things that I may feel, that it isn’t just the community thing; it’s also on the academic side.” Another community partner added, “I would say this is an extremely comfortable group. I don’t have any problems speaking in this group. Everyone’s been warm and wonderful.” While being comfortable expressing personal ideas in such a diverse group is necessary for having productive discussions and group development, it does not necessarily help the group make decisions rapidly, as we discuss below.

6. Ability to make decisions and “listening to each other.”

Some SC members raised concerns about the group’s ability to make decisions, which is a challenge, diversity literature frequently argues, shared by most heterogeneous groups.^{23, 24} One survey participant noted: “Sometimes I think there’s too much listening and not enough decision-making.” This comment spurred a lively discussion about the need to make decisions more efficiently. As one community partner suggested, “I don’t necessarily think we could do this work without the listening…If everybody had joined at the exact same time, I think that everybody would be up to par…But I do think at this stage of the game where we are now, we need to get concrete decisions made right now, move down, and then we end up at the end of the meeting with a whole lot more of a laundry list of things that need to be done…”

SC members also suggested options for how to make decision-making more efficient. One strategy that was offered focused on utilizing workgroups more effectively. The CPPR approach assumes that many decisions are made within various task-specific workgroups consisting of both academic and community partners. The SC attempts to delegate a certain amount of decision-making power to workgroups, which are responsible for more logistical and technical decisions and for raising items with more strategic and cross-group implications for discussion during SC meetings. However, as mentioned in the feedback session, many logistical and technical decisions were also being raised for discussion at SC meetings, in some cases because the necessary community or academic input was not available at the workgroup level. According to one SC member, the problem is to ensure delegation and participation: “It does sound like we need to ensure that the committees [workgroups] have as much participation as possible, so that those decisions can be made there, if possible. That’s something we have to work on.” This would require that workgroups also take responsibility for presenting pros and cons of making a certain decision to help SC members make decisions more effectively. Such a strategy where decisions are being made within the project by the designated group, which consists of both academic and community partners, balances the capacity of diverse groups to produce high quality decisions with the need to meet the timeline and goals of such a complex project.

Conclusion

The first wave of results from the CPIC Steering Council’s self-assessment survey and feedback session highlight the tradeoffs between the benefits and drawbacks of diversity within a community-partnered, participatory initiative. The SC appears to have achieved a relatively high degree of equitable sharing of participation in leadership and other roles, mutual trust and respect, and open communication that allow for such a diverse group of stakeholders to commit to the collective goals of the CPIC initiative. However, the open-ended survey comments and feedback group discussions also indicate that a number of differences of opinions and perceptions and areas of improvement in participatory process remain at this stage. Thus, the SC’s diversity demands that a great deal of time and effort be devoted to managing group dynamics.

The survey results highlight that these tradeoffs are made even more problematic by the complexity of the collective project, which for CPIC includes the multi-level study design (e.g., the SC, local community councils in the CE dissemination condition, and agency site-level implementation of evidence-based depression care) together with a community-partnered, participatory process at each level. Yet the complexity and diversity of the CPIC initiative are not unrelated. Indeed, the failures of traditional approaches of disseminating evidence-based health interventions in terms of intervention uptake, implementation, and sustainability over time within community settings of care are the impetus to attempt such a participatory approach among a broad range of stakeholders.

The SC self-assessment process itself demonstrates a number of challenges in attempting to manage the tradeoffs of diversity within a complex, participatory effort like the CPIC study. These challenges included issues such as how to maintain consistent measurement of group process over time within a formative evaluation context in which survey items are allowed to change in response to group feedback. Here, the mixed methods design, which includes consistent collection of qualitative data through the feedback sessions, is considered a strength of the CPIC self-assessment process. However, the chief challenge, which was raised explicitly in the feedback sessions, has centered on balancing the time, effort, and resources required for meaningful self-assessment analysis and reflection on group process with the need for the SC to focus on pending tasks and decisions.

Consistent with the literature,^{23, 24, 32} our findings show that while group diversity within the SC allows its members to entertain creative ideas for developing culturally appropriate solutions to pressing community problems, it makes it more time consuming for the SC to come up with actionable decisions. The need to have thorough group discussions to better understand the positions of multiple stakeholders makes it more challenging for the SC to make and then implement decisions. Delegation of technical and more logistical decision-making to the appropriate workgroup seems to be one of the possible solutions to this problem. While workgroups are diverse and consist of both academic and community partners, they are much smaller than the SC itself and therefore are likely to come up with actionable decisions faster. As SC members become accustomed to working with one another, they are also likely to become more comfortable delegating decision-making power to relevant workgroups, which will increase the SC’s decision-making efficiency.