Implementation of the Accountable Health Communities Model in Arizona

ABSTRACT

Objectives: The Accountable Health Communities Model facilitates multisector coordination within the health care delivery system to address health disparities driven by social determinants of health. One such initiative, the To Match and Align Through Community Hubs (2MATCH) project, was examined using the policy ecology of implementation (PEI) framework.

Study Design: Directed content analysis of a case study was conducted using a purposeful sampling of public and organizational documents published between 2017 and 2021.

Methods: Initial coding categories comprised 4 ecological levels, and a deductive approach was used by 2 researchers to systematically categorize and independently code the data.

Results: City-, county-, state-, and nongovernment-level initiatives were identified, and (1) the initiatives intersected across the political context and over time, (2) city and county initiatives were sparse, (3) the initiatives across levels did not directly intersect with the 2MATCH project, and (4) the initiatives across levels indirectly supported the objectives of the 2MATCH project. Furthermore, the government-led and nongovernment-led initiatives aligned with the 2MATCH project on some but not all the targeted health-related social needs. The most common needs were housing, food insecurity, and transportation; however, no initiatives directly or indirectly centered on utility needs or interpersonal violence.

Conclusions: The PEI framework elucidated the contextual factors that facilitated the implementation of the 2MATCH project. Future research examining implementation strategies that may facilitate pathways across contextual levels, especially localized efforts outside the traditional health care delivery system, to improve the adoption of multisector coordination is needed.

The American Journal of Accountable Care. 2023;11(3):14-21. https://doi.org/10.37765/ajac.2023.89434

Implementation science (IS) is the study of methods and approaches to integrate evidence-based research into real-world practices.^1,2 One of the key constructs studied in the IS literature is the context, which influences the adoption and the implementation of an intervention.³ The unique features of the context in which an intervention is implemented, including the social, cultural, economic, political, and organizational factors (ie, contextual factors), facilitate or hinder implementation.^3-5 The importance of identifying and understanding how contextual factors affect implementation is well documented. For example, social relationships and networks play a critical role in the adoption and implementation of interventions.⁶ Factors related to the organizational context, including teamwork, resources, and communication, influenced the implementation of rehabilitation models for individuals with cognitive impairments.⁷ Policy and regulatory changes were critical for the successful implementation of an evidence-based intervention in health care settings.⁸ The literature shows that contextual factors influence the success or failure of an intervention, and understanding these factors is crucial for improving the design of interventions for real-world settings. Although IS has made significant progress in understanding the role that context plays in the implementation of an intervention, gaps remain in the application of IS methods for examining the fit of an intervention in an existing context, particularly whether there is a good fit between the two.^9,10 Therefore, studying the alignment between context and an intervention and then assessing how the context directly or indirectly supports an intervention need to be considered.

This article reflects the call for smaller-scale evaluations of the Accountable Health Communities (AHC) Model in communities using IS methods.¹¹ The Center for Medicare and Medicaid Innovation introduced the AHC Model to assess whether systematically identifying the health-related social needs (HRSNs) of community-dwelling Medicare and Medicaid beneficiaries and addressing their identified needs affect their total health care costs and their inpatient and outpatient health care utilization. In the AHC Model, beneficiaries were screened for the following 5 HRSNs: (1) housing instability and quality, (2) food insecurity, (3) transportation needs, (4) utility needs, and (5) interpersonal violence.¹² These needs are driven by what are known as social determinants of health (SDOH), which are conditions in an individual’s environment (eg, home, work) that affect their health.¹³ Evidence suggests that HRSNs are related to poor health outcomes and higher health care costs.^12,14 To examine the implementation of the AHC Model in Arizona, this study applied the policy ecology of implementation (PEI) framework to examine the fit or alignment between the To Match and Align Through Community Hubs (2MATCH) project and the context and how the context supported its implementation. Implemented by a hospital and medical center (ie, bridge organization), the 2MATCH project screened Medicare and Medicaid beneficiaries with unmet HRSNs in need of health care services. Using an information technology solution, beneficiaries were connected with patient advocates to access services in the community.

The PEI framework is a conceptual framework used to examine the ecology of the implementation of evidence-based interventions and practices (Figure 1).^15-17 This framework elucidates downstream factors across multiple contexts (eg, public health initiatives, government, organization policies) and the congruence between these factors and the new intervention.¹⁶ These factors make up the broader ecology of facilitators and barriers that may affect the implementation process of interventions or practices in organizational settings.¹⁷ Using this framework facilitates an understanding of the system interactions and contextual factors that may enhance the adoption and sustainability of a new intervention.¹⁵

The PEI framework was chosen over other implementation frameworks because it presents the implementation of a program, strategy, or evidence-based practice beyond the constraints of the immediate environmental context. Unlike the Consolidated Framework for Implementation Research¹⁸ and the Exploration, Preparation, Implementation, Sustainment model,^19,20 the PEI framework considers the outer and inner contexts (eg, environment) within the greater ecology of public health strategies and societal factors (eg, stigma).²¹

CASE STUDY METHODS

Analytical Approach

A directed content analysis of a case study was applied to investigate the implementation of the 2MATCH project within the broader context, defined as the strategies implemented in Arizona that affect the SDOH. The case study research design is appropriate for an in-depth understanding of a phenomenon in its natural setting.^22,23 Search terms used Boolean logic and included AND combinations of “Arizona,” “social determinants of health,” “health-related social needs,” “Maricopa County,” and “Phoenix.” Government websites, news media publications, and organization documents related to SDOH implementation in Arizona between 2017 and 2021 were selected through purposeful sampling.

Following the PEI framework, initial coding categories comprising 4 ecological levels—the social, political, agency, and organizational contexts—were identified and used to inform the analysis of the selected resources.^16,24,25 Applying the conceptualizations of Raghavan et al¹⁶ broadly, the social context encompasses societal factors that may affect the implementation of a program, such as culture or discrimination. The political context includes legislative and advocacy efforts that directly or indirectly support the implementation of the program, strategy, or evidence-based practice. Agency and interagency contexts were defined as strategies and practices that affect the implementation efforts of a practice by an agency. Last, the organizational context involves the environment in which the delivery of a program occurs within an organization. Using a deductive approach, 2 researchers (M.M. and W.W.) systematically categorized and independently coded the data by applying these 4 ecological levels.

Through an iterative process, the number of codes produced was reduced—for example, due to redundancy—to a short list of categories. The researchers reduced bias during the coding process by leveraging peer debriefing, a strategy for trustworthiness and rigor in qualitative research.²⁶ Discrepancies in coding were resolved through discussion and consensus by the 2 researchers.

RESULTS

The findings of this study support the work of Raghavan et al¹⁶ and show that the implementation of the AHC Model was embedded in and indirectly influenced by broader public health strategies designed to address SDOH. That is, the policies and initiatives identified in the search explicitly referenced “social determinants of health” or “health-related social needs” or a specific social need (eg, housing). The PEI framework elucidates the contextual factors, which are inclusive of broader factors often not examined in other IS models. Summarized in the Table^27-40 are the initiatives and strategies identified across multiple levels, including the organization level.

Figure 2 illustrates the diverse set of initiatives across the levels of the PEI framework over time (ie, 2017-2021). The dotted circles show that the political and agency contexts overlap at the state, county, and city levels. This suggests that efforts to address SDOH in Arizona involved entities representing diverse stakeholder participation. At the state level, 5 initiatives were identified—1 addressed SDOH directly (ie, Lyft services for Medicaid patients), 2 initiatives highlighted SDOH as important factors to improve Arizonans’ health (ie, Arizona Health Improvement Plan, Whole Person Care Initiative), and 2 efforts systematically assessed SDOH in Arizona (ie, International Classification of Diseases, Tenth Revision SDOH; CommunityCares).

Compared with the state’s efforts to address SDOH, the city- and county-level initiatives were sparse. At the county level, the Health Improvement Partnership of Maricopa County, a collaborative of organizations in partnership with the Maricopa County Department of Public Health, identified 1 strategy related to SDOH: access to healthy food. At the city level, the Housing Phoenix Plan aimed to address the housing crisis in Phoenix by creating or preserving 50,000 homes over the next decade. The second initiative identified at the city level, the Phoenix Food Action Plan, addressed access to healthy foods, with a focus on individuals at risk for food insecurity and hunger. In addition to government-led initiatives, 3 nongovernment-
led efforts were identified: the Arizona Partnership for Healthy Communities, which focused on all SDOH; the Maricopa County Food System Coalition, which focused on expanding access to food sources; and the Arizona Community of Care Network, which focused on affordable housing.

At the organization level, 6 initiatives addressing SDOH were identified. The Community Health Implementation Strategy identified housing and homelessness, access to food and grocery stores, and transportation as primary SDOH that needed to be addressed in its service area. Three initiatives (ie, Homeless Discharge Initiative, Coordinated Community Network Initiative, Care Connection Resources) focused on SDOH through cross-sector collaboratives bringing together community partners and government entities to improve coordinated care in the community and address homelessness. Two programs (ie, ACTIVATE program, Clients Aligned Through Community and Hospital [CATCH]) focused on intensive case management in a similar way to the navigation services that beneficiaries received from the 2MATCH project. Even though the 2MATCH project was emulated to expand the ACTIVATE program, which primarily served beneficiaries with multiple chronic illnesses and repeated hospitalizations, the ACTIVATE program did not directly intersect with the implementation of the 2MATCH project. Despite the bridge organization’s focus on SDOH, these efforts, for the most part, were on the periphery to the 2MATCH project.

DISCUSSION

We leveraged the PEI framework to inform the analysis of this study. The strength of this model compared with other IS frameworks is its inclusivity of ecological factors outside the immediate organizational context. Using directed content analysis, initiatives related to SDOH and how these activities aligned with the 2MATCH project in Arizona were examined. Findings show that several state-, county-, and city-level activities along with nongovernment-led initiatives addressing SDOH aligned with the 2MATCH project on some but not all HRSNs. The most common SDOH activities were housing, food insecurity, and transportation; however, there were no efforts to address utility needs or interpersonal violence. This finding suggests that a potential gap or barrier exists for addressing these 2 HRSNs across levels. For example, when considering the ecology of implementation of the 2MATCH project, the alignment across levels pertaining to interpersonal violence, as an example, is key for understanding what factors (eg, social stigma) may facilitate implementation and what factors could effect change.²¹ This gap also suggests a need for programs on interpersonal violence and initiatives promoting prevention services to address utility needs in Arizona. Barriers to addressing these 2 HRSNs highlight opportunities for future research in these areas.

Findings also indicate that despite the state- and county-
level initiatives addressing SDOH, these activities did not directly affect the implementation of the 2MATCH project. These activities were on the periphery of the 2MATCH project, rather than directly affecting its implementation. Results also show that only 2 city-level activities, the Housing Phoenix Plan and the Phoenix Food Action Plan, aligned with 2 HRSNs, housing and food insecurity, respectively. SDOH initiatives at the city level were the least common compared with the state level and not that much greater in number than county-level efforts. Interestingly, research shows that the local community context is a significant driver of health among individuals, especially those with low income.⁴¹ Although the function of the AHC Model was designed for bridge organizations to directly engage partners to advance efforts to address the HRSNs of beneficiaries enrolled in the 2MATCH project, these study results suggest that existing pathways to do this within the local context may be absent. On the other hand, 6 initiatives across the bridge organization were identified, yet these efforts did not directly intersect with the 2MATCH project. This finding aligns with that of a study by Highfield et al,¹² which noted that organizational integration, in addition to the integration of partner organizations and the collaborative process, was a challenge in the implementation of the AHC Model.

Limitations

In terms of study limitations, the use of theory to guide the content analysis of this study inherently introduces a strong bias. This bias stems from the researchers’ informed approach to data analysis using a framework. However, the main strength of using directed content analysis is that the theory can be supported and expanded by the data.²⁵ Second, the selection of documents reviewed for this study may not be exhaustive of all the existing literature or relevant activities relating to SDOH in Arizona. However, multiple sources were used to triangulate the data to support these findings.²⁶ Moreover, the 2 researchers utilized reflexivity when coding the data. They also leveraged peer debriefing, discussing the codes and any judgment decisions that needed to be made. These processes were in place to ensure the trustworthiness and rigor of the study’s qualitative approach.²⁶

CONCLUSIONS

This study applied the PEI framework to understand the policy ecology for implementation of the 2MATCH project in Phoenix, Arizona. The findings advance our understanding of the ecological factors that directly or indirectly supported the implementation of a project that aimed to connect beneficiaries to social services in a timely manner. Using this framework, initiatives were identified that aligned with and/or affected its implementation. Other studies can build on this model to demonstrate how an IS framework can bolster future community health and well-being work.^3,42 Future research is needed to advance our understanding of implementation strategies that can facilitate pathways across contextual levels, especially localized efforts outside the traditional health care delivery system, to improve the adoption of multisector coordination aimed to address SDOH.

Author Affiliations: School of Social Work and Southwest Interdisciplinary Research Center, Arizona State University (MM, WW), Phoenix, AZ; Dignity Health St Joseph’s Hospital and Medical Center (AA), Phoenix, AZ.

Source of Funding: This research was funded by Dignity Health dba St Joseph’s Hospital and Medical Center (ASU FP00023488, AWD00032621; Dignity Contract Number: 17-0122-10-21/19-500-106-70-28).

Author Disclosures: The authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (MM); acquisition of data (MM); analysis and interpretation of data (MM, WW); drafting of the manuscript (MM, WW, AA); critical revision of the manuscript for important intellectual content (MM, WW, AA); statistical analysis (MM); provision of study materials or patients (AA); obtaining funding (WW, AA); administrative, technical, or logistic support (AA); and supervision (WW, AA).

Send Correspondence to: Micaela Mercado, PhD, LMSW, Arizona State University, 411 N Central Ave, Phoenix, AZ 85004. Email: mfmercad@asu.edu.

REFERENCES

1. Koh S, Lee M, Brotzman LE, Shelton RC. An orientation for new researchers to key domains, processes, and resources in implementation science. Transl Behav Med. 2020;10(1):179-185. doi:10.1093/tbm/iby095

2. Bauer MS, Damschroder L, Hagedorn H, Smith J, Kilbourne AM. An introduction to implementation science for the non-specialist. BMC Psychol. 2015;3(1):32. doi:10.1186/S40359-015-0089-9

3. Nilsen P, Bernhardsson S. Context matters in implementation science: a scoping review of determinant frameworks that describe contextual determinants for implementation outcomes. BMC Health Serv Res. 2019;19(1):189. doi:10.1186/s12913-019-4015-3

4. May CR, Johnson M, Finch T. Implementation, context and complexity. Implement Sci. 2016;11(1):141. doi:10.1186/s13012-016-0506-3

5. Pfadenhauer LM, Gerhardus A, Mozygemba K, et al. Making sense of complexity in context and implementation: the Context and Implementation of Complex Interventions (CICI) framework. Implement Sci. 2017;12(1):21. doi:10.1186/s13012-017-0552-5

6. Bunger AC, Yousefi-Nooraie R, Warren K, Cao Q, Dadgostar P, Bustos TE. Developing a typology of network alteration strategies for implementation: a scoping review and iterative synthesis. Implement Sci. 2023;18(1):10. doi:10.1186/s13012-023-01266-3

7. McGilton KS, Cumal A, Corsi D, Gingrich S, Zheng N, Escrig-Pinol A. Context matters when implementing patient centred rehabilitation models for persons with cognitive impairment: a case study. BMC Health Serv Res. 2021;21(1):204. doi:10.1186/s12913-021-06206-9

8. Chambers DA. Context of health policies and the impact on implementation of health care and health interventions. In: Meier N, Dopson S, eds. Context in Action and How to Study It: Illustrations From Health Care. Oxford University Press; 2019:86-103.

9. Ellis J, Band R, Kinsella K, et al. Optimising and profiling pre-implementation contexts to create and implement a public health network intervention for tackling loneliness. Implement Sci. 2020;15(1):35.
doi:10.1186/s13012-020-00997-x

10. Escoffery C, Sekar S, Allen CG, Madrigal L, Haardoerfer R, Mertens A. A scoping review of outer context constructs in dissemination and implementation science theories, models, and frameworks. Transl Behav Med. 2023;13(5):327-337. doi:10.1093/tbm/ibac115

11. Gottlieb L, Colvin JD, Fleegler E, Hessler D, Garg A, Adler N. Evaluating the Accountable Health Communities demonstration project. J Gen Intern Med. 2017;32(3):345-349. doi:10.1007/s11606-016-3920-y

12. Highfield L, Ferguson G, Walsh R, et al. A conceptual framework for addressing social needs through the Accountable Health Communities Model. Am J Accountable Care. 2020;8(4):15-23. doi:10.37765/ajac.2020.88681

13. Olayiwola JN, Rastetter M. Aiming for health equity: the bullseye of the quadruple aim. J Hosp Manag Health Policy. 2021;5:11. doi:10.21037/jhmhp-20-101

14. Daniel-Robinson L, Moore JE. Innovation and opportunities to address social determinants of health in Medicaid managed care. Institute for Medicaid Innovation. January 2019. Accessed August 30, 2022. https://healthystudentspromisingfutures.org/wp-content/uploads/2019/07/2019-IMI-Social_Determinants_of_Health_in_Medicaid-Report.pdf

15. Aby MJ. Race and equity in statewide implementation programs: an application of the policy ecology of implementation framework. Adm Policy Ment Health. 2020;47(6):946-960. doi:10.1007/s10488-020-01033-2

16. Raghavan R, Bright CL, Shadoin AL. Toward a policy ecology of implementation of evidence-based practices in public mental health settings. Implement Sci. 2008;3:26. doi:10.1186/1748-5908-3-26

17. Stone EM, Daumit GL, Kennedy-Hendricks A, McGinty EE. The policy ecology of behavioral health homes: case study of Maryland’s Medicaid Health Home Program. Adm Policy Ment Health. 2020;47(1):60-72.
doi:10.1007/s10488-019-00973-8

18. Damschroder LJ, Aron DC, Keith RE, Kirsh SR, Alexander JA, Lowery JC. Fostering implementation of health services research findings into practice: a consolidated framework for advancing implementation science. Implement Sci. 2009;4:50. doi:10.1186/1748-5908-4-50

19. Aarons GA, Hurlburt M, Horwitz SM. Advancing a conceptual model of evidence-based practice implementation in public service sectors. Adm Policy Ment Health. 2011;38(1):4-23. doi:10.1007/s10488-010-0327-7

20. Moullin JC, Dickson KS, Stadnick NA, Rabin B, Aarons GA. Systematic review of the Exploration, Preparation, Implementation, Sustainment (EPIS) framework. Implement Sci. 2019;14(1):1. doi:10.1186/s13012-018-0842-6

21. Powell BJ, Beidas RS, Rubin RM, et al. Applying the policy ecology framework to Philadelphia’s behavioral health transformation efforts. Adm Policy Ment Health. 2016;43(6):909-926. doi:10.1007/s10488-016-0733-6

22. Njie B, Asimiran S. Case study as a choice in qualitative methodology. IOSR J Res Methods Educ. 2014;4(3):35-40. doi:10.9790/7388-04313540

23. Walker SC, Sedlar G, Berliner L, et al. Advancing the state-level tracking of evidence-based practices: a case study. Int J Ment Health Syst. 2019;13(1):25. doi:10.1186/s13033-019-0280-0

24. Kohlbacher F. The use of qualitative content analysis in case study research. Forum Qual Soc Res. 2006;7(1). doi:10.17169/fqs-7.1.75

25. Hsieh HF, Shannon SE. Three approaches to qualitative content analysis. Qual Health Res. 2005;15(9):1277-1288. doi:10.1177/1049732305276687

26. Lietz CA, Zayas LE. Evaluating qualitative research for social work practitioners. Adv Soc Work. 2010;11(2):188-202. doi:10.18060/589

27. Arizona Health Improvement Plan. Arizona Department of Health Services; 2016.

28. Arizona Health Improvement Plan. Arizona Department of Health Services; 2021.

29. Whittal K, Rustagi K, Bivins D, Marks W, Filler A. Social determinants of health (SDOH) and risk adjustment: Arizona Medicaid innovations. Wakely. March 2021. Accessed August 30, 2022. https://www.wakely.com/sites/default/files/files/content/social-determinants-health-sdoh-and-risk-adjustment.pdf

30. AHCCCS Whole Person Care Initiative (WPCI). Arizona Health Care Cost Containment System. Accessed September 22, 2022. https://www.azahcccs.gov/AHCCCS/Initiatives/AHCCCSWPCI/

31. Lyft is providing rides for eligible Medicaid patients in six states, increasing access to care for potentially nine million. Lyft. October 24, 2019. Accessed August 30, 2022. https://www.lyft.com/blog/posts/lyft-is-providing-rides-for-eligible-medicaid-patients-in-six-states

32. Arizona’s social determinants of health (SDOH) closed loop referral system (CLRS). Healthcurrent. Accessed August 30, 2022. https://web.archive.org/web/20230401203033/https://healthcurrent.org/sdoh-old/

33. Minnick P. The Arizona Healthy Communities Opportunities Index. 2016. Accessed August 30, 2022. https://bit.ly/47HacEy

34. A comprehensive food assessment for Maricopa County. Maricopa County Food System Coalition. September 2019.

Accessed September 22, 2022. https://marcofoodcoalition.org/wp-content/uploads/2020/05/MARCO_Full-Summary-Report_Final_compressed.pdf

35. Gracia A. Hundreds of Phoenicians don’t have easy access to healthy food. Arizona Republic. September 16, 2019. Accessed September 22, 2022. https://www.azcentral.com/story/entertainment/dining/2019/09/16/phoenix-putting-together-its-first-citywide-food-action-plan/2022725001/

36. Home Matters to Arizona. Accessed August 30, 2022. https://homemattersarizona.com/

37. Coordinated community health needs assessment for Maricopa County. Maricopa County Department of Health. 2017. Accessed August 30, 2022. https://www.maricopa.gov/DocumentCenter/View/40784/MC_2017_CCHNA_Report

38. Housing Phoenix. Phoenix Community Alliance; 2020.

39. Albright R. 2025 Food Action Plan: healthy food for all. City of Phoenix. January 2020. Accessed September 22, 2022. https://www.phoenix.gov/sustainabilitysite/Documents/FINAL%202025%20Phoenix%20Food%20Action%20Plan%20Jan%202020.pdf

40. St. Joseph’s Hospital and Medical Center community health implementation strategy 2019-2021. Dignity Health. 2019. Accessed August 30, 2022. https://www.dignityhealth.org/content/dam/dignity-health/pdfs/implementation-strategies/2019/final-sjhmc-2019-chis.pdf

41. Billioux A, Conway PH, Alley DE. Addressing population health: integrators in the Accountable Health Communities model. JAMA. 2017;318(19):1865-1866. doi:10.1001/jama.2017.15063

42. Rogers L, De Brún A, McAuliffe E. Defining and assessing context in healthcare implementation studies: a systematic review. BMC Health Serv Res. 2020;20(1):591. doi:10.1186/s12913-020-05212-7