Building State Cooperatives for Healthcare Improvement: Meeting Summary

Introductions, Agenda Review, and Opening Remarks

Jesse (Jay) Crosson, Senior Social Scientist, Center for Financing, Access, and Cost Trends, AHRQ EvidenceNOW: Building State Capacity Team 
Tess Miller, Acting Director, Center for Evidence and Practice Improvement, AHRQ

Jay Crosson welcomed the group, thanking everyone for their attendance. Each person in the room introduced themselves, sharing their name, location, and organization. Jay then went over the agenda for the day. 

Tess Miller opened the meeting with a few acknowledgements about the history of state cooperatives at AHRQ. 

Key Takeaways

• AHRQ is celebrating its 35th anniversary and over 30 years of funding primary care research initiatives to address topics such as heart health, unhealthy alcohol use, urinary incontinence, and blood pressure control and tobacco cessation (including ENOW BSC).
• Across each initiative, the Agency continued to learn and refine their understanding around building capacity in practices, delivering more evidence-based care, improving equity, and strengthening connections between primary, specialty, and community-based care. 
• The findings from ENOW BSC will build upon this foundation of knowledge and inform future initiatives, looking towards the next iteration of the work: AHRQ's Healthcare Extension Service: State-Based Solutions to Healthcare Improvement.

State Grantee Presentations

Shari Bolen, MD, MPH – Heart Healthy Ohio Initiative, Principal Investigator
Larry Hearld, PhD – Alabama Cardiovascular Risk Reduction Network, Lead Evaluator
Jim Bailey, MD, MPH – Tennessee Heart Health Network, Principal Investigator
Theresa Walunas, PhD and Anya Day, MPH – Healthy Hearts for Michigan, Co-Principal Investigators

A representative from each of the four ENOW BSC state grantees shared a few of their most important lessons learned from their experience building improvement cooperatives in their states. 

Key Takeaways

Heart Healthy Ohio (Ohio):

• Having strong partnerships involved in the cooperative is key, and Ohio built upon existing relationships with Medicaid.
• Ohio leveraged academic relationships, linking three schools of medicine together to extend their reach and expertise across the state, and these schools were able to provide clinical coaches and other subject matter expertise to the cooperative.
• Ohio tapped into the relationships across three regional health improvement collaboratives to expand their QI infrastructure and brought in the Ohio Academy of Family Practice to engage medical providers and reach primary care Federally Qualified Health Centers (FQHCs).
• Additionally, they brought in various professional organizations across Ohio, with which they did not have existing partnerships, including nursing professional organizations, CHWs, physician assistants, dieticians, and pharmacists.
• One challenge was engaging the business community. While Ohio had a multistate coalition for businesses in their partnership, and they were interested in employee engagement, there were constraints with sharing data and collecting process measures. One way to overcome this challenge in the future would be to explore data-sharing options outside of an EHR. 
• Ohio has a long history of improvement efforts for health, so when approaching legislators, it helped to build upon past successes to demonstrate their expertise. 

Alabama Cardiovascular Cooperative (Alabama):

• Unlike other states, the Alabama team had to construct and operationalize their cooperative without any existing structures. 
• Their backbone organization was the University of Alabama Birmingham, with Auburn University and Alabama Primary Health Care Association (APHCA) as key partners. Their steering committee included major payers as well.
• One key takeaway from the team was the importance of clearly defining a transparent governance structure for the cooperative early on in the process. One challenge they faced was a lack of understanding of roles and responsibilities across the partner organizations. This was mitigated through frequent meetings and clear communication. 
• Alabama also learned the importance of understanding the diverse technology and data infrastructures in primary care to inform interventions and to set the pace for implementation. The range of different EHRs and their capabilities had an outsized effect on efforts to generate reports and do QI. 

Tennessee Heart Health Network (Tennessee):

• Tennessee’s most important takeaway was the importance of engaging stakeholders in what they deem to be meaningful work. It takes significant administrative and governance support to do this well, and the plan must have an intentional, clear, and compelling strategy that is well aligned with partners’ goals. 
• Their governance structure included payers, providers, patients, and CBOs, among others.
• Payer engagement started with a strong PCOR evidence-based goal up front that the payers bought into. They established a reimbursement subcommittee, which first discussed alignment with Medicaid payment codes. The subcommittee spent 2 years working to get all of the payers to use the same codes.
• For patient engagement, Tennessee had a strong patient advisory council from the onset. Tennessee engaged with patients by sharing patient stories and vetting all initiatives through them, which helped center health equity and align the work with patients’ high-priority needs.
• Engaging legislators is an ongoing process, but the President of the University of Tennessee has strong relationships with state legislators. In general, having information presented to legislators by a trusted source is critical to making progress on key issues (e.g., reimbursement codes).
• However, the team avoided certain legislation, such as licensure of health coaches, in order to ensure they could advance the work and avoid delays in decision making.

Healthy Hearts for Michigan (Michigan):

• Michigan did not have a formal cooperative structure upon which to build, but they had existing partnerships to leverage. For their cooperative, it was key to think about cooperative membership from multiple angles, including core infrastructure, expertise, and experience to inform the work. They specifically engaged the Office of Rural Health and State Health Department. 
• Michigan particularly focused on developing partnerships with groups that had relationships with the types of practices they wanted to work with—rural practices. They also sought experts in QI initiatives, practice facilitation, study design, etc. They intentionally sought out a QI organization located in the Upper Peninsula to achieve a greater geographic reach across the state. 
• Michigan engaged physician experts from various organizations, including a rural physician to advise on intervention adaptations for rural health centers. 
• They partnered early with the state Office for Rural Health and the state Health Department was on their steering committee. 
• Additional partners that would have been helpful to include were payers, MCOs, patients and families, and different subject matter experts. 
   

Abt Evaluation Team Presentation 

Robin Bloodworth, PhD, Abt Global LLC
Sarah Shoemaker-Hunt PhD, PharmD, Abt Global LLC

Key Takeaways

Evaluation goals:

• Understand the characteristics of the Cooperative structure grantees built in their states, and the barriers and facilitators they experienced when building their Cooperatives.
• Understand what QI support strategies were provided to practices and to what extent?
• Measure whether there is an association between the QI support provided by the cooperatives and practice capacity and clinical outcomes. 

Evaluation methods:

• Mixed-methods evaluation using baseline and post-intervention practice-level data on:
  • Practice capacity for change.
  • Percent of patients with hypertension in good control.
  • Percent of tobacco users provided cessation counseling.
• Used qualitative data to understand Cooperative models and to explore the barriers and facilitators to building state cooperatives for improvement.

Practice characteristics:

• Most practices had from two to five clinicians, were located in an urban core area, and had a medically underserved area designation.
• To measure the intervention “dose,” the evaluation team used active hours of QI support. Overall, more than half of practices spent 5–8 hours participating in active QI support.
• The regression model found that all main outcomes were significantly higher post-intervention.
• Other interesting findings: 
  • Practice ownership is positively associated with practice-level improvement capacity—compared to clinician-owned practices, practices with hospital/health system and safety net ownership had higher improvement capacity scores. 
  • Value-based payment (VBP) has mixed effects on outcomes. Practices with VBP models with Medicaid as a payer had lower mean tobacco cessation, and practices with VBP models with private/commercial payers had higher mean tobacco cessation.
    • These significant VBP trends were the same in the practice-level improvement capacity model, with the addition of VBP models with an “Other” payer type had lower mean improvement capacity scores.

Conclusions:

• At the end of the intervention, ENOW BSC practices had significantly higher levels of:
  • Tobacco cessation counseling.
  • Blood pressure control.
  • Practice capacity (Change Process Capacity Questionnaire) post-intervention as compared to baseline.
• Practice capacity was improved, especially for practices with a hospital/health system and/or safety net.
• Major disruptive events in primary care were associated with lower practice capacity in our study, but not clinical outcomes.
• Future QI initiatives may benefit from early harmonization and agreement on a minimum dataset and definitions, consistently captured in grantees’ QI support interventions (e.g., dose, mode, content).

Lessons learned:

• Given the well-documented challenges, time, and resources needed to obtain QI measures, consider alternative approaches to obtaining QI data if the data are being used in feedback to practices.
• Multistakeholder engagement is needed to co-develop, coordinate, and align primary care improvement.
• Building on and strengthening existing relationships with state partners is essential for forming and sustaining a primary care improvement cooperative. 
• Payer partners are critical for future sustainability and sustainability should be planned for from the beginning. 

Breakout Discussion I: Working with Public and Private Entities to Co-Develop, Coordinate, and Align Primary Care Improvement Efforts, Including Building a Quality Improvement Workforce

This breakout group sought to understand the best practices, lessons learned, and key challenges to engaging public and private partners when establishing state cooperatives. 

Key Takeaways

Critical actions for establishing a state cooperative:

• Develop a sustainability plan at the outset of the cooperative and engage purchasers and payers early to maintain momentum when funding ends. 
• Build relationships with both organizations and the individuals that represent those organizations, to build and nurture trust. Have multiple relationships at each organization to sustain partnerships and account for transitions.
• Build time to:
  • Understand the common incentives and challenges among partners, close gaps to arrive to common goals, and build durable relationships.
  • Educate and advocate at the state and federal levels to effectively convey the value of state cooperatives and state-level primary care spending.
• Conduct needs assessments and asset mapping and encourage asset sharing among partners.
• Develop a clear vision of success, including intentional processes on how to get there, and a clear understanding of how success is defined across partners.
  • Co-create with partners from the outset. Otherwise, “storming and norming” efforts could take the group in a different direction than communities’ primary care practice needs. Establish and be clear on what decisions need to be made collectively.
  • Take a human-centered design approach and engage the community, with skilled facilitation, which ensures a shared understanding of co-design and decision-making processes.
  • Invest in “small table work” (i.e., partners with close affinity) that will benefit more productive “big table” discussions (e.g., engage primary care providers and those that can speak for and advocate for their QI priorities).
• Maintain efforts to improve access to care and better address social and structural determinants and drivers of health that will ultimately align disparate partner incentives to cooperate. Refer to Community Health Detailing model, as one example. 
• Engage neutral conveners, including practice facilitators, who are familiar with practice and patient populations.
• Build a QI workforce at the cooperative level with common training and certification processes and recruit non-traditional workers for the QI workforce (e.g., CHWs, youth, patients, peers, CBO staff who provide social services).
• Engage practices that benefit most from the cooperative program, even if it takes more work to engage and recruit them (e.g., community health centers, rural practices, small and medium-sized independent practices).
  • Understand that some hospitals have robust QI operations and a leadership structure that can accept the invitation and connect the cooperative to many sites.
  • Understand that some hospital leaders have little power. Even if they personally want their hospital to participate in a QI initiative, it may not be possible due to other priorities.
  • Be aware of competing demands to build patient care where QI is a more distal concern.
• Engage clinicians to ensure that the QI efforts address what they see as important for care.
• Hone storytelling to bring partners not historically engaged (e.g., payers, hospitals, employers, subspecialists) to effectively describe cooperative efforts and objectives to build understanding, alignment, and buy-in. 
  • Focus on stories that depict different types of return on investment and can show partners and potential partners how the cooperative efforts will meet their distinct missions over time.

Key developments needed for the next generation of state cooperatives:

• Co-create a “community charge” to cooperatives that includes a clear value proposition and economic argument to patient, payers, clinicians, and other key partners.
• Design your implementation strategy with sustainability in mind from the outset by engaging policymakers, payers, hospitals, and other multisector and community partners early and often.
  • Estimate return on investment and value of investment of QI efforts to help make a case for upfront funding.
  • Pool resources and investments from local/regional “outcomes buyers” to fund these supports and reinvestment (i.e., outcome-based contracting that shifts from fee-for-service payment and delivery models to more value-based healthcare). 
  • Leverage hospital community benefits programs.
  • Identify and develop joint investment and action plans with partners to leverage assets and address infrastructure gaps that limit primary care.
  • Include asset mapping, data sharing, funding pool, and local/regional primary care improvement resources in improvement efforts.
  • Align with VBP for primary care and to sustain overall payment reform momentum to ensure sustainability.
• Center and embed equity in primary care and QI efforts. 
• Acknowledge and attend to misalignment of incentives, rewards, resources, motivation, and timing among partners to nurture buy-in and effectively collaborate. 

Breakout Discussion II: Aligning Incentives, Payment, and State-Level Resources with Improvement Efforts

This breakout group sought to understand the best practices, lessons learned, and key challenges related to aligning incentives, payment, and state-level resources when building a state cooperative. Additionally, groups discussed the opportunities in existing work and where the field needs to advance for the future cooperatives. 

Key Takeaways 

Governance:

• Casting a wide net to include a broad range of stakeholders is key for the governance structure of a cooperative. Include partners such as payers, providers, business representatives, Medicaid, MCOs, patients and families, CBOs, and legislators.
• A strong governance structure should be established to align metrics and incentives, which will help to build the trust and relationships needed to execute healthcare extension work. 
• Self-interest is valued in the governance structure, and it is important to understand each entity’s interest, while also understanding when self-interest needs to be set aside to focus on the shared vision and agenda. 
• The state cooperative can serve as the neutral convenor to collect data and share themes around payments, incentives, and other key metrics for the group. 
• As part of the governance structure, key decision-makers are needed to sign off on certain initiatives and funding opportunities. 

Equity:

• Incentives are necessary for practices and communities to execute the work and should be developed equitably. Disaggregated data across populations will assist with understanding what amounts and which types of incentives are needed, based on a given region or population.
• Some communities and providers have been under-resourced across generations, and they might need more upfront funding to participate effectively in an intervention as well as additional technical assistance and coaching.
• “De-medicalize” structures by including resources such as CHWs. They should be brought to the table early on as true partners with transparent payment. These are people who are trusted by the community and affect return on investment – both in the expertise they provide and the relationships they have in the community. Investing in a full-time employee in the community has positive effects that go beyond healthcare. 
• Some states have a director or office of equity in their Medicaid MCOs, and this is sometimes a way to align priorities and secure funding.
• Cities that lie on the border of states might need to consider how Medicaid and MCOs operate in the neighboring states, since the tertiary health center might lie across the border. 

Legislators:

• Strong relationships with legislators can bring in additional funding. 
• Appealing to legislators involves a specific skillset focused on narrative-based organizing techniques and storytelling to encourage more funding and/or changes in legislation. 
• It is important to keep in mind that legislators represent their communities, not the healthcare system. When appealing to legislators, focus on how the cooperative will help the communities they serve. Research the legislators’ priorities and align funding efforts along those priorities. 
• Some states are not able to use the word “equity” for political reasons. Be thoughtful about finding language that resonates with legislators and other stakeholders to align priorities and secure funding.

Payers:

• Medicaid may be a crucial partner for aligning incentives across a state, but there are some key considerations for partnering.
  • In states where Medicaid has deferred its plan responsibility to a third party, these parties have considerable autonomy. It is important to understand the opportunities and barriers each one brings.
  • A Medicaid managed care payer in one county might operate differently than one in a neighboring county, causing local differences in payments.
• It is crucial to have multipayer engagement, but it is also very difficult. Bring payers together around antitrust and encourage them to be at the table and feel comfortable. Together, analyze what is possible in a state and establish norms and rules to foster conversation.
• There needs to be a space just for payers as well as a space for payers and practices to come together. 

Expanding beyond the practices:

• A lot can be learned from the Vermont shared resource model, in which community resources were shared across and used by all the practices. Similarly, providers that are part of a state cooperative can share resources and expertise around key topics (e.g., Health IT). https://blueprintforhealth.vermont.gov/
• QI cannot always be achieved with the practice alone, so explore what is in the community. If there are health councils, do they connect with each other? 

Breakout Discussion III: Using Data for Ongoing System Improvement and to Inform State and Local Decision Making

Breakout groups throughout the day were asked to identify both the critical actions that someone establishing an improvement cooperative should consider, and the key gaps in data that need to be filled to drive the field forward for the next generation of state cooperatives.

Key Takeaways

Data governance:

• Be thorough in establishing data governance when planning a state cooperative. Include people on the team who are knowledgeable about the state’s data structures (e.g., EHRs, health information exchanges (HIEs), community information exchanges (CIEs), Medicaid), limitations, and capabilities and plan for sustainability of data collection. 
• Agree on the key outcomes of the QI initiative at the start and consider whether the outcomes are measurable, and then determine the best methods for data collection, sharing, and translation. 
• Allow ample resources and time to thoroughly train those capturing and analyzing the data to standardize best practices and to create alignment and efficiencies. Understanding and cleaning the data takes considerable time and can reduce resources for the actual intervention. Consider creating a learning community where analysts can share ideas and problem solve challenges together.
• Understand what data are already being captured (e.g., by HIEs, Clinically Integrated Networks (CINs), other networks) so you are not duplicating efforts and/or burdening providers. Consider if there are similar QI programs in your state (e.g., Making Care Primary (MCP), All-Payer Health Equity Approaches and Development Model (AHEAD)).

Data collection to improve health equity:

• Thought leaders noted that data should be collected to further the field’s understanding of the effect of social complexity on clinical outcomes.
• QI efforts should prioritize health equity and QI data should be used to identify potential health disparities.
• Some suggested methods of SDOH data collection:
  • CIEs can be leveraged to capture both health and social risk data.
  • Place-based data that capture community-level data on SDOH (instead of SDOH data captured by providers).
  • Data linkages groups can collect some geocoded data on SDOH. 

Data alignment, harmonization, and sharing:

• Despite its complexities, QI data collection should be harmonized between states to evaluate QI efforts. 
• State cooperatives need to share QI data within their own state and with other states to effect changes at a national level. They need to understand the barriers to QI efforts, to analyze and translate the QI data, and to demonstrate the return on investment of specific QI interventions (e.g., use of CHWs). 
• There should be alignment at the state level on qualities of high performing primary care providers and the data that can adequately capture QI efforts, (e.g., quality measures that capture practices’ continuity, comprehensiveness, access, and person-centeredness).
• QI data can be a key component of sharing the “story” of the QI initiative—whether that is to recruit practices or to advocate for funding. 
• Consider other data sources than EHRs to understand the impact of QI initiatives. Capture and share qualitative data from both patients and providers, patient-reported outcomes and/or patients’ health goals, experience of care data from the providers, and financial data, since quality measures do not always tell the full story. More real-time data shared with practices can enable quicker QI interventions. 

Reduce data collection burden on providers:

• Most primary care practices are enrolled in multiple QI programs, which each require related data collection. To the extent possible, researchers should work to reduce the burden on providers. Thought leaders noted either more funding is needed to establish an infrastructure at the provider level to collect and analyze QI data or state cooperatives should fill this role. 
• While collection of SDOH data is vital, resources are needed at the community level to connect patients with resources. Too often, the data are collected but the providers cannot take meaningful action to improve a patient’s situation, which can reduce positive healthcare outcomes.
• Help meet providers where they are, in terms of capturing QI data. Emphasizing the importance of the QI initiative and how it can lead to practice-level improvement will create better buy-in by providers than focusing on system improvements.
• Providers and staff need assistance with both understanding their data and with the methods for using QI resources in their practices to help their patients.

Large Group Discussion

All meeting attendees reconvened for a discussion about their key learnings from others and validations of their experiences that arose during the breakout sessions. Additionally, attendees were asked to inform AHRQ of the main questions they would ask of someone who plans to develop a state cooperative.

Key Takeaways

Learnings and validations:

• QI exists at the intersection of public health, primary care, and community. Multisector collaboration and alignment is the only way to start making meaningful progress.
• Primary care providers understand healthcare complexities. There should be alignment of payers, purchasers, and states in QI efforts with primary care serving as the common denominator across all. 
• Know your story and the levers you can use to appeal to different audiences—including providers, payers, and policy makers.
• Ensure that payers are involved from the beginning of the state cooperative to promote sustainability.
• State cooperatives can serve as the neutral convenor for multipayer alignment.
• Ensure that equity is at the center of investments. This may mean disproportionately investing in certain communities or practices to achieve equitable outcomes.
• Be patient and realistic in terms of the pace at which work gets done; results don’t happen right away.
• Land grant universities can be strong partners because they have expertise in extension services. 
• Ask the community what resources and groups are already available as well as when and where they convene instead of asking them to join new QI efforts. Align extension work with existing efforts and activities in the community so that the work thoughtfully fits into current interventions. Healthcare improvement efforts should be seen as part of community development and engage other organizations working in this same space.

Questions someone developing a state cooperative should consider: 

• How will they fund the state cooperative?
• What is your story?
• What is your overall plan, from evaluation to sustainability?
• What are your goals and are they measurable from the available data? What type of data—quantitative and qualitative—can you collect outside of an EHR?
• What other programs (e.g., Making Care Primary, All-Payer Health Equity Approaches and Development Model) already exist in your state and how can you align?