X

ONECare Population Health Academy – Join For Free

"*" indicates required fields

Already a member of the OPEN MINDS network? Click here to login.
Name*
Address*
This field is hidden when viewing the form
This field is hidden when viewing the form
MM slash DD slash YYYY
This field is hidden when viewing the form
This field is hidden when viewing the form

X

Oops! You need to be logged in to use this form.

Two Health Plans, Same Value-Based Data Challenge

|

By Monica E. Oss, Chief Executive Officer, OPEN MINDS

Over the past year, I’ve met with countless health plan and provider organization executives, and one subject always comes up—the current and growing financial challenges for provider organizations. While the pandemic and inflation have increased the costs of delivering services, fee-for-service reimbursement rates are not following suit. Most executives agree that the solution is that provider organizations should be reimbursed based on value rather than volume, but there is a lack of consensus about what exactly constitutes “value.”

 width=

The numerous challenges of moving toward provider reimbursement based on value in the behavioral health field were explored in the recent 2023 OPEN MINDS Performance Management Institute session, What Health Plans Want: Measuring & Communicating Performance To Payers. The session featured two executives who are in the midst of the transition of behavioral health service reimbursement—Monica Collins, Senior Director of System Transformation for Magellan Behavioral Health of Pennsylvania (Magellan), and Susanna Kramer, Director of Performance Evaluation for Community Behavioral Health (CBH). One common theme quickly emerged during the discussion—the lack of standardized, reliable performance data is a significant and pervasive barrier to implementing value-based purchasing (VBP) reimbursement.

“There are no plug-and-play models,” according to Ms. Collins. “Are there frameworks that may work? Yes, but you must understand the problem that managed care and provider organizations are trying to solve. There isn’t an industry standard tool to track behavioral health outcomes. Everyone measures things a little differently. Infrastructure is limited, and provider organization readiness to measure risk is spotty at best.”

Ms. Kramer echoed this sentiment, speaking to the challenges of collecting standardized, value-based data across their network of provider organizations. “There is an ongoing challenge of creating performance measures that don’t create a perverse incentive to cherry-pick our members. We currently have no standardized data collection methods across our provider network, especially for social determinants of health (SDOH), which makes risk tiering and developing non-claims-based measures more difficult.”

Both Magellan and CBH are managed care organizations (MCOs) participating in HealthChoices, the Pennsylvania Medicaid behavioral health managed care program. In 2018, Pennsylvania Medicaid implemented requirements for a value-based purchasing (VBP) initiative, including fee-for-service (FFS) linked to performance (low risk), supplemental payments attached to shared savings and risk (medium risk), bundled payment arrangements (medium risk), and global payments based on quality measures (high risk). In year one, the financial targets set for the model are 5% of total reimbursement. By 2023, it is supposed to be 30% with additional stipulations: half must be considered moderate-to-large risk and 85% must address SDOH.

Magellan serves five Pennsylvania counties for HealthChoices. In 2018, they started with an incentive-based inpatient model that compared provider organization performance against other provider organizations and increased in risk in years two and three. In 2019, it added an assertive community treatment (ACT) framework that held provider organizations accountable for model fidelity, as well as a family-based model that held provider organizations accountable for working together for the consumers. In 2020, the ACT framework became higher-risk, and Magellan also launched an inpatient shared savings model. In 2021, it then implemented a blended case management approach and a non-attendance at outpatient appointments (DNA) model.

 width=

CBH serves the City of Philadelphia’s Department of Behavioral Health and Intellectual disAbility Services (DBHIDS) for HealthChoices. It launched its original pay-for-performance (P4P) initiative in 2007, which it then used to meet the lowest-risk requirements for the HealthChoices program. During additional years of increasing VBR requirements, CBH transitioned its P4P to higher risk VBP models for acute psychiatric inpatient provider organizations and case management provider organizations. Community-based organizations were selected to participate in the VBC initiative only if they addressed one of two critical member needs: housing instability for members discharged from acute psychiatric inpatient care or transportation to community integrated recovery centers.

 width=

While challenging, provider organizations with performance metrics—for consumer access, experience, and outcomes—are going to have a competitive advantage over the organizations that do not. These performance proof points—along with metrics related to HEDIS measures—are top of mind for health plans. And, this performance data, along with the ability to track unit costs, longitudinal consumer utilization, and population health data, are critical to success in non-fee-for-service reimbursement systems.

Regardless of the specific system design in any state or consumer service sector, the ability for executives to measure performance—and manage it—is fundamental to success in a health care landscape that is increasingly integrated, tech-enabled, hybrid, and value-based. My suggestion is: don’t wait to develop the “perfect” metrics-based management system, which arguably doesn’t exist. Rather, keep in mind the wise words of Charles Babbage, an inventor considered to be the father of the modern computer: “Errors using inadequate data are much less than those using no data at all.”