ARCH: A Look at Data Systems

Advancing Resilience and Community Health (ARCH) is a project exploring what it takes for networks of community-based organizations (CBOs) to come together and partner with healthcare institutions (payors and hospital systems). NFF has been working with three networks – EngageWell IPA (New York), Metropolitan Alliance of Connected Communities (Minnesota), and Thomas Jefferson Area Coalition for the Homeless (Virginia) – to understand some of the challenges and opportunities involved in working toward partnerships with healthcare, including how to manage data.

Operating in a network requires data sharing across member organizations, the network itself, and healthcare partners, all while considering the confidentiality of people who are accessing services from multiple providers. Data helps organizations understand what’s working for their clients, and among networks of CBOs, data can be used to:

Inform new interventions and approaches for improving community health
Make the case for investments needed
Demonstrate value in partnership opportunities
Help define and track outcomes for agreements with healthcare partners

With the goal of making sure that people receive the care and resources they need, no matter what life circumstances they are facing, data is a foundational part of collaborating as a network and fundamental to successful healthcare partnerships. Following are some of the main themes and considerations that ARCH’s three networks are exploring around data sharing and integration as they build partnerships with healthcare.

Maximize data that’s already being collected for compliance and use it strategically

What this looks like in practice:

EngageWell uses a statewide system called PSYCKES (“Psychiatric Services and Clinical Knowledge Enhancement System”), which was designed by the NYS Office of Mental Health to capture behavioral health claims data for Medicaid beneficiaries receiving mental health or substance use disorder treatment services. EngageWell draws population health data and recipient-level reports from PSYCKES to explore behavioral health needs of the network, identify high-need populations, and uncover trends in service utilization. EngageWell also uses PSYCKES to understand how the network as a whole is performing, identify areas for improvement, develop value propositions, and inform clinical interventions that could be impactful on specific quality measures important to healthcare payers. While a great source of data, PSYCKES is not without its limitations, notably: recipient-level data is only available if a patient has consented to share; various data elements are refreshed at different times, leading to lags in data access; some EngageWell members do not provide Medicaid-billable behavioral health services and are not included in the system; and the platform has limited demographic data (e.g., no race or ethnicity data), which is an impediment to identifying important population health disparities. To uncover data by key demographics, the IPA must request data directly from providers and perform a data match, which is costly and time consuming.

Metropolitan Alliance of Connected Communities (MACC) serves as the data management intermediary for their network of members and uses a robust data system – ClientTrack, supported by software company Eccovia Solutions – to support members in getting set up for compliance-based reporting and then using data for deeper level strategies and storytelling. Because of the deep expertise on MACC’s data team, they are able to work closely with member organizations to create customized database experiences to match individual compliance, operational, and strategic needs, while maintaining the core architecture of the shared platform such that network-level findings and trends can still be shared.

MACC also uses their technical expertise and deep connections and roots in the nonprofit community to help as an interpreter when it comes to data strategy. MACC recognizes that CBOs are best at describing their impact and mission, and in conversations with new health and other partners, MACC is able to help by raising common data management and sharing challenges experienced by all nonprofits – helping to take the burden off of any one organization alone and advocating for more equitable partnerships. Collecting too much information is not just burdensome for CBOs and participants but ultimately takes away precious time that could be spent on programs and elsewhere. The MACC team has seen how resource-starved communities are often the most highly researched and over-surveyed. In response, MACC always focuses on the purpose and reason for different data collection – aiming to reduce it down to only what is necessary.

Thomas Jefferson Area Coalition for the Homeless (TJACH) manages a shared database system called the Homeless Management Information System (HMIS) for homeless service providers. It’s used as a client case management and referral platform that captures the services received by clients who are experiencing homelessness, and the outcomes of those services. TJACH uses this data to track performance measures for individual programs as well as on the community level. Though it has primarily been used to track who TJACH is sending referrals to, the goal is to use its referral capability to create a better closed loop referral system to track services rendered and outcomes achieved for shared clients between different systems.

TJACH and its partners have also used their data to evaluate the impact of racism on housing in Charlottesville. They are using what they are learning to redesign the assessment and prioritization process to more equitably respond to housing and service needs of those in its community.

What to consider:

Which data are most useful for your organization to collect to understand how your services fulfill your mission?
What metrics are needed to see how the network is performing collectively toward improving community health and wellbeing?
Is your organization identifying and using data that has been defined as a priority by those served?
What data can be used to inform new approaches or interventions? To avoid collecting “to much data”, what data or outcomes do you want to prioritize and focus on for your reporting year?
And what metrics can help your network demonstrate its value to healthcare partners and/or payors?
How will data be used in a new partnership? What amount needs to be collected?
Who has access to data in the partnership? What will be reported?

Data integration across multiple systems is hard to get right and requires creative problem-solving

What this looks like in practice:

EngageWell has 15 active Member/Owners in the network with 15+ care management systems, electronic medical records, and propriety databases in place. It’s hard to come up with a single solution that unifies all these different systems. EngageWell is working with Data Across Sectors for Health (DASH), supported by RWJF, to develop a proof-of-concept data collection strategy that moves the network toward systematic data collection and analytics. Inside this project, EngageWell is gathering copies of data reported to common funders to analyze more detailed utilization of the network (e.g., Is John Smith receiving services from multiple providers within the network?). The first step was the completion of an environmental scan to look at what databases are in use, the reporting functionalities, and what’s easy for members to access and share with the IPA. Next, with the help of data consultants from CCSI, EngageWell is identifying data points – medical, behavioral health, and SDOH – from each database that will enable client demographic and service information to be matched, aggregated, and analyzed. The goal is to develop a data repository that, when combined with data available in PSYCKES, will paint a broader picture of population need and become a useful quality management tool, benefiting providers, managed care plans, and other payors.

MACC has worked hard to address some of the challenges that come with using the same database across multiple organizations, recognizing that one data system doesn’t always match each member’s internal reporting needs, workflows, or processes – and each organization has its own culture, staffing structure, and level of resources available. In addition to creating custom database experiences for members to solve for disparate data needs, MACC takes advantage of having a common database to replicate learnings and creative solutions across member organizations on the platform.

TJACH is working with a local health system to figure out how to integrate data systems for referrals to ensure that patients are following up on social determinants of health services once they leave the health system. However, two key challenges around this integration include trying not to burden staff with double data entry and figuring out how to amend current release information given HIPAA constraints. TJACH is also using the data it collects to take a deep look at the entire homelessness system to better understand the gaps in service delivery as a first step towards generating shared commitment to addressing those gaps and identifying resources to fill them.

What to consider:

How can you integrate or create systems that unify various proprietary databases, provide necessary reporting information for payors and providers - while avoiding double entry requests and addressing HIPAA compliance issues?
How do you maximize use of your integrated data for the best client/patient service?

Defining agreed upon outcomes upfront is critical when establishing contracts with a healthcare partner

What this looks like in practice:

EngageWell is pursuing two value-based contracts where they need to demonstrate meaningful improvements on a variety of quality measures and medical and behavioral health outcomes while reducing overall healthcare spending. Part of their work involves using data to understand the characteristics of the clients they would be responsible for, risk they would need to take on, and metrics that are ambitious enough to appeal to the health care partner, but realistic enough to be achievable by the network.

MACC is involved in a pilot project with Allina Health and Children’s Hospitals in Minnesota to explore a referral process from healthcare providers to a small group of MACC member organizations. Much of the work upfront involved convening healthcare partners and participating members to build a collective understanding around what exactly should be piloted. Before even getting into the technical implications of how to share data, they needed to agree on what data should be shared and what success means. One of the biggest ah-ha moments for healthcare partners and MACC members was coming to an agreement that the group is sharing service outcomes and not health outcomes, which helped to set expectations going into the technical design and implementation portions of the pilot work.

What to consider:

Does the contract reflect or ensure a shared benefit for all partners? 
Is there alignment in expectations and clarity on what the contract is trying to establish (i.e., what is the goal, and what is the network responsible for)?
How is success measured and what are the systems that need to be in place to evaluate the outcomes? 
Is there shared power and decision-making around defining and acting on outcomes as defined by the community being served? And are there mechanisms in place to hold partners accountable?

Data management requires significant capacity but is often an un- or underfunded expense

What this looks like in practice:

EngageWell is evaluating significant investments in an upgraded data system to be able to deliver on the contracts it is pursuing. Although it did receive some investment funding from the state for this purpose, it is not enough to cover the anticipated full cost of implementation. The IPA has regularly asked itself, “Should we buy a data system or build one?” No matter the approach, costs quickly mount and become eventually prohibitive without additional investment. Moreover, its healthcare partners have made clear that they do not believe it is their responsibility to build the infrastructure of the network. Their interest is in buying service delivery only. So we are left with the question, whose responsibility is it to pay?

MACC has prioritized capacity building to meet the needs of different organizational cultures around data collection, which they do through a “train the trainer model” with members. Some of their recommended best practices are to:

Have a dedicated data administrator at the organization who supports training, monitors data quality, and understands broader agency goals for data collection
Host regular meetings among data administrators across the network to share practices around how to address data quality issues, build buy-in for end users, addressing data security, and more
Ensure that the data administrator is connected to program staff and leadership so that everyone is working in partnership to address needs and concerns so that data can be used for strategic planning, reporting, and more

What to consider:

How is your network being funded (or not) to develop data integration platforms?
What’s needed at your organization and/or network to create a culture of quality data collection? What type of staffing and capacity is required?

As evidenced by the three networks participating in ARCH and countless others working toward partnerships with healthcare, managing data systems is nuanced and complex, but ultimately serves to keep participants and mission delivery at the forefront.

Click here to learn more about DASH’s efforts to support collaborations that work together to improve the health of their communities, promote health equity, and contribute to a Culture of Health. And click here to see what else NFF has learned through the ARCH project about what it takes to advance community health.

Support for Advancing Resilience and Community Health (ARCH) was provided by the Robert Wood Johnson Foundation. The views expressed here do not necessarily reflect the views of the Foundation.

Join over 24,000 social sector leaders who receive our latest insights and noteworthy stories throughout the year.