Abstract
Background
Chronic care management (CCM) and community health worker (CHW) interventions hold promise for managing complex chronic conditions such as diabetes and related comorbidities. This qualitative study examines facilitators and barriers to the implementation of an expanded CCM intervention that explicitly incorporated program staff, clinic staff, CHWs, and partnerships with community-based organizations to enhance diabetes management among Mexican-origin adults.
Method
Grounded theory was used to analyze interviews conducted in 2018 with 24 members of the CCM team, including program staff, clinic staff, and community-based program partner staff.
Results
Three themes emerged that characterize perceived facilitators and barriers to CCM implementation, based on analysis of interviews: (1) understanding roles and responsibilities across organizations, (2) building relationships across organizations, and (3) coordinating delivery of the model among different organizations. First, structured meetings and colocated workspaces enhanced understanding of CCM roles for each team member and across organizations. Barriers to understanding CCM roles were more common during the early stages of CCM implementation and amongst staff who did not participate in regular meetings. Second, regular meetings facilitated development of relationships across organizations to enhance implementation of the CCM model. In contrast, limited relationship building among some CCM team members served as a barrier to implementation. Third, CHWs and case review meetings fostered communication and coordination across the CCM model.
Conclusions
Results suggest the importance of understanding roles and building relationships among multidisciplinary teams to ensure effective communication and coordination of care.
Diabetes disparately affects Latino/a communities and contributes to comorbidities (Centers for Disease Control and Prevention, 2017). Latino/a adults with type 2 diabetes (henceforth, diabetes) have worse glycemic control and are more likely than their non-Latino/a White counterparts to have diabetes-related distress and depressive symptoms, which can adversely affect glucose control (Ali et al., 2012; Heisler et al., 2007). Among Latino/a adults (12.1%), diagnosed diabetes prevalence is highest for Mexican-origin adults (13.8%; Centers for Disease Control and Prevention, 2017).
A complex chronic condition, diabetes involves a lifelong process of monitoring and managing glucose levels, engaging in physical activity, and maintaining a nutritious diet. Diabetes is often compounded by multiple chronic conditions, including mental health issues that have historically been addressed unsuccessfully or separately from diabetes (Mezuk et al., 2013). There is increasing attention to transforming health care systems to address multiple chronic conditions (e.g., diabetes, cardiovascular disease, mental health) by integrating high-functioning care teams across organizations to deliver continuous care (Patel & Chatterji, 2015).
Chronic care management (CCM) interventions have shown promise for improving diabetes outcomes (Stellefson et al., 2013). Effective diabetes CCM models integrate clinical care and diabetes self-management support through the coordination of team-based care (Stellefson et al., 2013; Wagner, 2000). CCM models are often delivered by multidisciplinary teams, including primary care providers, other health care practitioners, and program staff (Stellefson et al., 2013; Wagner, 2000). Key elements of CCM models include an organized health care delivery system linked with complementary community resources, sustained by productive interactions between multidisciplinary care teams, and activated patients and families (Glasgow, Tracy Orleans, & Wagner, 2001; Wagner, 1998). Barr et al. (2003) describes an expanded chronic care model where health system teams include prevention and health promotion practitioners who address social determinants of health around chronic care delivered in the community and health system (Barr et al., 2003).
Growing evidence indicates the benefits of diabetes interventions led by community health workers (CHWs) or promotores, particularly in medically underserved communities and communities of color (Rothschild et al., 2014; Spencer et al., 2018; Two Feathers et al., 2005). CHWs are trusted members of affected communities who have a close understanding of the community and can serve as an intermediary between health care systems, social services, and the community (American Public Health Association, 2019). Diabetes self-management education and support (DSMES)—which can be delivered by CHWs—is defined as the “ongoing process of facilitating knowledge, skills, and ability necessary for diabetes self-care as well as activities that assist a person in implementing and sustaining the behaviors needed to manage his or her condition . . .” (Beck et al., 2017, p. 35). Social and economic inequities, stressful life conditions, and limited access to health care contribute to racial/ethnic inequities in diabetes self-management and diabetes-related outcomes (Hawkins et al., 2017; Hunt et al., 1998; Jiang et al., 2005; LeBrón et al., 2019; Schulz et al., 2005).
While some CCM models include CHWs (Wagner, 2000), few complement the intervention with expertise of local community-based organizations to support patients with complex social and economic needs that affect health (Stellefson et al., 2013). Salud y Vida 2.0 (SyV 2.0) is an expanded CCM model that integrated a multisectoral team with CHWs and community-based organizations. SyV 2.0 was tailored to individuals with persistent uncontrolled diabetes despite participating in SyV 1.0, a CHW and DSMES program, for at least 6 months.
Despite the success of CCM interventions, an implementation evaluation of expanded CCM models that explicitly incorporate CHWs and partnerships between health care systems and community-based organizations is lacking in this literature. Research examining factors that facilitate or impede intervention implementation is warranted. Few studies (Collinsworth et al., 2014; McElmurry et al., 2009; Raaijmakers et al., 2013) have examined these factors from the perspective of the implementation team. Program staff are critical stakeholders in implementing and sustaining effective CCM models. The integration of CCM models in community-based settings offers a unique approach to enhancing diabetes management among Mexican-origin adults with diabetes. To address this gap, we examined factors that facilitated and impeded the implementation of an expanded CCM model (SyV 2.0), from the perspective of members across the implementation team. SyV 2.0 integrated multiple health care providers (e.g., pharmacists, behavioral health) with community-based staff to address the complex social and health needs of participants with diabetes and other comorbidities.
Method
We conducted a qualitative study of factors that facilitated and challenged implementation of an expanded CCM intervention for Mexican-origin adults with uncontrolled diabetes. Implementation staff (n = 24) participated in semistructured in-depth interviews. This study was reviewed by the New England Independent Review Board (Reference No. #120170278) and determined to be exempt from institutional review board review in 2017.
Intervention Setting
Located on the northern bank of the Rio Grande River separating the United States from Mexico, the Rio Grande Valley is home to 1.2 million residents (U.S. Census Bureau, 2013). Approximately 31% of Mexican-origin adults in the Rio Grande Valley have diabetes (Fisher-Hoch et al., 2012). This region is characterized by several assets, including innovative partnerships to promote community well-being and the strength of binational ties.
This region also has the highest concentration of colonias in Texas, which are unincorporated settlements along the Texas–Mexico border lacking basic living necessities (e.g., drinking water, paved roads, safe and sanitary housing). Given limited health care access and concentrated poverty, the Lower Rio Grande Valley presents opportunities to intervene on health inequities affecting Mexican-origin communities (Barton et al., 2015; Davila et al., 2014).
The eligibility criteria for the SyV 2.0 CCM intervention included SyV 1.0 participants who resided in Cameron or Willacy Counties, Texas, who did not achieve an HbA1c (hemoglobin A1c) value <9.0% in their first 6 months of the SyV 1.0 program and were patients at one of two local health clinics with an HbA1c ≥8.0%. In SyV 1.0, adults with uncontrolled diabetes were assigned a CHW who provided education and social support during home visits and phone calls for 15 months and were enrolled in a 6-week DSMES course. Among SyV 2.0 participants (n = 176; M age = 51.5 years), most identified as female (70.5%) and Latino/a (92.1%), primarily spoke Spanish (67.7%), and had a monthly household income of ≤$1,000 (74.1%). More than half of the participants did not graduate from high school (59.1%), were unemployed (61.4%), and uninsured (69.4%).
Intervention
The University of Texas Health Science Center at Houston, School of Public Health (UTHealth SPH) and its partners implemented the SyV 2.0 evaluation in September 2016 to June 2018. SyV 2.0 is an expanded CCM model (Barr et al., 2003; Wagner, 1998) that integrates health care services with home and community-based wraparound services delivered by CHWs and community partners to provide a continuum of care for residents with diabetes. SyV 2.0 is informed by the transtheoretical model (Prochaska & Velicer, 1997) and social cognitive theory (Bandura, 1986). The SyV 2.0 theory of change is that comprehensive and coordinated community and clinical services delivered to adults with uncontrolled diabetes will lead to improved physical and mental health outcomes.
The 12-month SyV 2.0 intervention included care coordination by a team of providers (e.g., CHWs, pharmacists) and referrals to behavioral health counseling by master’s-level trained counselors, medication therapy management services delivered by pharmacists for participants with low medication adherence, culturally sensitive peer-led support groups, community-based lifestyle programs that promoted healthy eating (e.g., La Cocina Alegre cooking course), and coordination of enhanced CCM activities by program staff (Figure 1). CHWs, fluent in Spanish and English, provided screening, home-based wraparound services, and motivational interviewing.
Salud y Vida 2.0 expanded chronic care management and community health worker intervention.
Interview Participants
Interview participants (n = 24) included the following: organizational-level program staff who coordinated intervention implementation (n = 8), community-based program partner staff (n = 10, including CHWs), and clinic staff (n = 6) involved in intervention implementation.
Data Collection
The goal of the interviews was to understand the context, facilitators, and barriers to expanded CCM program implementation and operational-level workflow (Kegler et al., 2019; Shelton et al., 2017). Evaluator (author VLP) developed the interview guide in collaboration with UTHealth SPH (authors BR, LMB, MRS), based on knowledge of intervention components, the RE-AIM implementation framework (Glasgow, McKay, et al., 2001), and literature gaps (Garney et al., 2018). The interview guide included open-ended questions and probes related to participants’ roles in the delivery of SyV 2.0 (Table 1). Staff responsible for implementing SyV 2.0 were invited to participate in interviews conducted in June to July 2018. Four evaluators, experienced and trained qualitative researchers who were not involved in the day-to-day implementation of the intervention, conducted semistructured interviews during site visits or by phone (Garney et al., 2018). Interviews lasted 30 to 60 minutes and were conducted in participants’ preferred language (n = 20 English interviews; n = 4 Spanish interviews).
Expanded Chronic Care Model Staff Interview Questions.
Note. EMR = electronic medical record; BMI = body mass index.
Data Analysis
Interviews were digitally audio-recorded, professionally transcribed, and entered into NVivo (QSR International Version 12), a qualitative data analysis software. Spanish language interviews were translated and then analyzed. Our systematic data analysis involved four steps: (1) Two trained qualitative researchers who did not conduct the interviews reviewed transcripts to develop a mutually agreed on codebook using a grounded theory approach (Charmaz, 2012; Glaser & Strauss, 2017). The codebook included deductive codes based on development of the interview guide and key research questions, as well as inductive codes that emerged from interviews (Asada et al., 2017; Griffith et al., 2013). (2) Two researchers (authors AMWL and another evaluator) independently coded the transcripts. The coders and qualitative analysis team discussed and reconciled discrepancies between codes and applied revisions to the codebook. Differences were reconciled through discussion until consensus was reached. (3) To assess factors that facilitated and challenged program implementation, we examined the following codes: leadership meetings and communication, staff and provider meetings, trainings, integrated behavioral health model transition and buy-in, roles and responsibilities, relationships, workflow, provider–provider interactions, and provider–patient interactions. To improve trustworthiness of findings, researchers debriefed with interviewers, reviewed codes, and discussed emerging themes and interpretations (Garney et al., 2018; Morrow, 2005).
Results
Participant Characteristics
The final sample included 24 participants whose average age was 45.1 years with 8.7 years of experience in their role; the majority identified as female (75%) and Latino/a (91.7%; Table 2).
Sociodemographic Characteristics of Interview Participants, Rio Grande Valley, 2018 (n = 24).
Perceived Facilitators and Barriers to CCM Intervention Implementation
The results are organized by three emergent themes related to implementing the CCM model: (1) understanding roles and responsibilities across organizations; (2) building relationships across organizations; and (3) coordinating care among organizations (Table 3, Figure 2).
Major Themes and Relevant Quotes.
Note. CCM = chronic care management; CHW = community health worker.
Categories and themes related to delivery of expanded chronic care management (CCM) model.
Theme 1: Understanding Roles and Responsibilities Across Organizations
Understanding roles and responsibilities across organizations emerged as a facilitator and barrier to implementation of the CCM intervention, yielding two subthemes: (1) structured meetings and work environment enhanced understanding of CCM roles across organizations and (2) barriers to understanding CCM roles were more common during early implementation and among staff who did not participate in regular meetings.
Subtheme 1.1: Structured meetings and colocated workspaces enhanced understanding of roles across organizations
Program staff and community-based program partner interviewees characterized structured in-person communication among staff across organizations (e.g., case review meetings, community-based program meetings) as facilitating regular opportunities to learn how each organization’s roles contributed to the multifaceted CCM model. One program staff interviewee shared, What really helped was understanding what [the intervention] was trying to do as a whole and to . . . provide these services together . . . community partners felt like they were the sole contributor to lowering this person’s A1c and some of them felt a little overwhelmed with that and we had to reassure them that . . . not only are they going to La Cocina Alegre, but this person is receiving medication therapy management, . . . peer-led support groups as well. . . . And you know, seeing that come together, I think that it put everybody at ease and they were able to communicate as well.
Community-based partners perceived that regular meetings and colocated workstations facilitated communication about each organization’s role in the day-to-day delivery of CCM. As one community-based staff interviewee shared, I sit next to the promotoras, so I hear that and then I’m familiar with the transportation services that they offer . . . I just hear the programs that they used . . . We talk about referrals.
Subtheme 1.2: Limited knowledge of roles early on and among staff not engaged in regular in-person meetings
While several interviewees described understanding their organization’s role within the CCM model and becoming familiar with other organizations’ CCM roles, some interviewees perceived their colleagues did not fully understand each organization’s unique and complementary contribution to the model. Additionally, some interviewees noted they themselves were not familiar with each organization’s roles. According to interviewees, limited knowledge of each organization’s role was more common during the early implementation stages and among team members who did not regularly participate in structured in-person communication opportunities. As one program staff interviewee described, “I don’t think that Cocina Alegre understands their relationship to behavioral health. And the behavioral health doesn’t understand the relationship to peer-led support groups.”
Some staff interviewees suggested enhancing each organization’s understanding of how their activities are connected would help participants see the SyV 2.0 services as integrated. For example, a program staff interviewee shared how during early implementation some partners were unclear about roles of others: “The key is the communication. Making sure that we are all clear and know what’s our mission as a whole, not just you as being a community partner.”
In sum, this theme highlights the importance of structuring ongoing opportunities for organizations to discuss their unique and complementary CCM roles.
Theme 2: Building Relationships Across Organizations
The second theme pertained to building relationships across organizations, which was a facilitator and challenge to implementation. There were two subthemes (1) developing relationships that facilitate CCM model implementation and (2) limited relationship building across organizations.
Subtheme 2.1: Developing relationships that facilitate CCM model implementation
Several interviewees perceived case review meetings and community-based meetings supported the development of relationships among organizations. Interviewees characterized relationship building as contributing to increased communication among organizations, enhanced awareness of community-based programs, and identification of improved CCM workflows. For example, one clinic staff interviewee explained how regular meetings fostered relationships that they could call on when implementing the CCM model on a daily basis: I found myself being more interactive with the doctors I guess because I know the importance, and this is what we should be doing . . . I’m looking at the medication, I’m educated on both types of medication but now I need the help of the doctor. So it just opened the door for communication of course with the patient but also to get the doctor involved to return our calls.
Relatedly, community-based staff discussed how meetings strengthened relationships with other community-based programs within the CCM model. As one program staff interviewee described, Within our partnerships, our relationships have grown stronger . . . We learned how to better communicate with each other and then for the partners that were new . . . we were able to kind of establish what those lines of communication would be.
According to some interviewees, community-based meetings and partnership meetings provided an opportunity to learn about the culture and organizational structures that shape clinical and community-based organizations. As one program staff interviewee described, We’re different organizations with different policies, so it’s [a] matter of working together to build those relationships . . . with the clinics it’s been a challenge not only because it’s working with . . . a partner, right? . . . it’s also working with the different systems at the clinic and different challenges with their own staff and the different resources at each clinic.
Subtheme 2.2: Limited relationship building across organizations
Some interviewees described lacking personal connections to staff at other organizations, which presented a barrier to coordination and communication across the CCM model. Often, clinic staff interviewees described limited personal connections with community-based staff, in particular CHWs. In contrast, community-based staff described knowing and communicating regularly with CHWs. As one clinic staff interviewee explained, “So I knew that [the promotores] were there, I didn’t know who they [promotores] were and how to get a hold of them.” Several clinic staff interviewees perceived that program staff played an important role in connecting clinic staff and CHWs. As one clinic staff interviewee shared, At the onset of this program, I do remember asking [program staff] if there is any way that we can meet or have the name or phone numbers [of promotores] so that we can at least make some sort of communication happen so that we are all on the same page, but that never really happened.
Thus, strengthening relationships was critical to coordinating care across participating organizations.
Theme 3: Coordinating Care Among Organizations
The final theme, coordinating care among organizations, included two subthemes that reflected facilitators to intervention implementation: (1) CHWs serve as a critical nexus of CCM communication and (2) case review meetings facilitate coordination of the CCM model.
Subtheme 3.1: CHWs serve as critical nexus of CCM communication
According to several interviewees, CHWs served as a critical nexus of CCM communication across organizations. One program staff interviewee described the centrality of CHWs to the delivery of CCM: I think they’re [promotores] the main source of actually kind of selling the services to the participant, motivating them to go and even following up to see if they did go and also to identify any barriers that are happening that are the reason why they are unable to go to the service.
Several interviewees described CHWs as bridges between organizations and participants. As one program staff characterized: “They’ll [promotores] kind of be that point of contact to help make that happen as well on the clinic end.”
While several interviewees described directly communicating with CHWs to facilitate communication, others communicated through the CHW coordinator. For example, one clinic staff interviewee described, If patients are continuing as a no show, we inform the program coordinator . . . to inform them that this patient is simply not coming in, and . . . to relay that information to the promotoras.
As this quote illustrates, the CHW coordinator served as a liaison across the complex and dynamic CCM team, reflecting a hierarchy in the CCM model designed to facilitate communication. While some interviewees described successfully communicating with CHWs, others described challenges, suggesting the need to better streamline communication between CHWs and CCM partners. Thus, the hierarchy of communication in the CCM model may have also impeded direct communication between CCM partners and CHWs, resulting in delayed care coordination.
Subtheme 3.2: Case review meetings facilitate coordination of CCM model
Several program and clinic staff interviewees perceived biweekly case review meetings, which included program staff (e.g., CHW supervisors, program coordinators) and clinic staff, as facilitating coordination of CCM model components. Interviewees explained that during CCM meetings, program and clinic staff discussed updates about participants whom CHWs identified as having increased HbA1c levels, based on CHW case notes developed in collaboration with and presented by CHW supervisors. According to one clinic staff interviewee: We discuss either different strategies or more recommendations to help this participant improve . . . what we can do to help this participant, and that could be referring them to a service that they weren’t initially approved [for] from the beginning of enrollment.
Several interviewees described how case review meetings provided space to discuss stressors affecting participants’ health, which may in turn affect their CCM participation. According to interviewees, these discussions often helped mitigate participant barriers by identifying opportunities to leverage the CCM model. One program staff interviewee described, All of the partners . . . come together to review a case [where] instead of going the way we all hoped, it’s going the opposite way . . . we dedicate that time to review all of the social determinants that are not helping this participant control their diabetes. It’s a good time where the clinic side and community talk together and review each case.
According to clinic and community-based staff, case review meetings facilitated handoffs from one CCM partner to another (e.g., behavioral health to social services, CHWs to community-based programs). Several interviewees attributed this ease of connecting participants with CCM components to the establishment of a communication network among CCM staff and keeping staff informed about each component of the CCM with which they could connect participants.
Discussion
A deeper understanding of factors that facilitate or constrain implementation of expanded CCM models incorporating multiple clinics, community-based organizations, CHWs, and community-based initiatives is imperative for translating sustainable CCM models into health care settings. This study examined perceived facilitators and barriers to implementation of an expanded CCM model serving a predominantly Mexican-origin community with diabetes. Prior to this intervention, CCM organizations did not have a history of working together to deliver integrated care. A central finding from this qualitative inquiry is the importance of understanding roles and building relationships across organizations to ensure effective communication and coordination of care; this is consistent with previous literature regarding multidisciplinary community-based interventions and partnerships focused on understanding and addressing chronic conditions (Garney et al., 2018; Vaughn et al., 2018).
Implications for Research, Policy, and Practice
Findings reported here illuminate the critical importance of understanding each organization’s role and building relationships across organizations to enable effective implementation of CCM models and coordination of care across multisectoral teams of clinic, program, and community-based staff (Wagner, 2000). Moreover, regular and structured opportunities to facilitate relationship building fostered communication and coordination across organizations. These findings point to the ongoing nature of relationship building when developing and sustaining interventions.
Notably, CHWs streamlined CCM delivery and served as a bridge between intervention participants, CCM team members, and organizations. However, an important implementation barrier was lack of direct communication between CCM partners and CHWs, given that some staff reported limited connections with CHWs. CCM meetings included the CHW coordinator, who was in close contact with CHWs and represented CHWs in meetings given time constraints posed by CHW’s high caseload across an expansive geography. Additionally, post hoc discussions with CCM partners illuminated how hierarchies in health care settings pose challenges to convening physicians, administrators, and CHWs as equal partners. This finding indicates persistent challenges of integrating clinic and community-based organizations within complex interventions. According to interviewees, CHWs were embedded in the community while clinic staff were based at the health center; this highlights the need to integrate different spheres of influence and expertise in CCM-CHW models. Future CCM interventions would benefit from ensuring CHWs are equitably integrated into the design of the communication strategy.
These results suggest complex interventions require substantial commitment to communicating and training the team in their roles. Moreover, integrating teams to address medical needs and social determinants of health is a relatively new model in medical care (National Academies of Sciences, Engineering, and Medicine, 2019), indicating the importance of clarifying roles. Integrating nontraditional partners intensifies the need for ongoing training on roles and responsibilities and structured opportunities to build relationships across organizations.
Limitations and Strengths
Findings from this study should be understood within the context of some limitations. First, while these findings may extend to other CCM models that seek to mitigate health inequities, the goal of this study was to examine perceived facilitators and barriers to implementing this expanded CCM model. Second, this study involves an analysis of interviews conducted at the end of the 12-month intervention. Studies that longitudinally examine facilitators and barriers of intervention implementation (e.g., baseline, midpoint, summary) are warranted. Third, future studies should examine the perspectives of primary care providers regarding CCM implementation, which were excluded from this study. Fourth, interviews were completed by the outside evaluation team who were not program administrators. It is possible that interview participants shared responses perceived to be socially desirable. However, evaluators perceived that interview participants openly discussed implementation facilitators and challenges. Despite these limitations, the focus on staff perspectives of program implementation offers useful insights into the implementation of interventions integrating multiple sectors to improve chronic conditions such as diabetes.
Conclusions
Community-based CCM interventions involving a multifaceted team of health care providers, CHWs, community-based partners, and program staff hold promise for mitigating inequities in complex conditions such as diabetes. The SyV 2.0 implementation evaluation contributes to the body of evidence regarding integrated health services in a community-based setting within a low-income population of Mexican-origin adults with persistent uncontrolled diabetes. These findings underscore the importance of ensuring all organizations understand their roles in the intervention, as well as roles of other organizations.
Footnotes
Acknowledgements
The authors thank the participants for their involvement in the intervention, the UTHealth SPH personnel and all affiliated clinics and partners for their collaborative work in implementing SyV 2.0 and Health Resources in Action, Inc., as the external evaluator. The authors also acknowledge members of the research team involved in qualitative data collection and analysis, including Erika Gaitan, MSW, and Edlín Maldonado-Fuller, MBA, MA, for conducting interviews, and Allyson Auerbach, MPH, for providing research assistance with coding of interview transcripts. The authors also would like to recognize the funding for this trial from Methodist Healthcare Ministries.
Authors’ Note
Opinions or points of view expressed in this document are those of the authors and do not necessarily reflect the official position of, or a position that is endorsed by the Corporation for National and Community Service.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received financial support for the research, authorship, and/or publication of this article: The authors acknowledge grants from Methodist Healthcare Ministries of South Texas, Inc., the Corporation for National and Community Service (Social Innovation Fund Grant No. 14SIHTX001) and Valley Baptist Legacy Foundation. UTHealth SPH is a proud subgrantee of Methodist Healthcare Ministries’ Sí Texas Project, a Social Innovation Fund Program.
