Analytics and the New Medicaid Rules

5 Pieces of Advice About Analytics and the New Medicaid Rules (Industry Perspective)

Analytics have become a vital underpinning to Medicaid managed care policy and program administration.

BY ZACH AMBROSE, JEREMY RACINE AND DAN GITTERMAN / JUNE 16, 2016

In April, the Centers for Medicare and Medicaid Services (CMS) released the first major update to Medicaid managed care in more than 10 years. A lot has changed in 10 years, including the rise of analytics as a vital underpinning to Medicaid managed care policy, program administration and oversight, clinical decision support and patient outcomes.

The new rules (dubbed CMS-2390-F) further validate the need for more comprehensive, advanced analytics for quality of care improvement, value-based care, program and fiscal integrity, and better management of the Medicaid managed care program.

As expected, the new rules were a hot topic at the recent University of North Carolina (UNC) Analytics for Policy seminar on health care and Medicaid managed care. Stakeholders representing policy, program, payer and academia convened to discuss analytic use cases, best practices and the bright future of analytics in Medicaid managed care. 

Here are five pieces of advice, gleaned from the event, that every Medicaid stakeholder should know. 

Get everyone on the same page regarding business intelligence and analytics.

There are many misconceptions about business intelligence and analytics and how they can be used, or not used, to move toward more prescriptive or optimized capabilities. Tarun Kushwaha, associate professor of marketing at UNC’s Kenan-Flagler Business School, argued that business intelligence is often used as an umbrella term to describe predictive analytics, reporting and ad hoc querying.  

Kushwaha made the case for moving beyond ad hoc querying and static reporting, which is what business intelligence really is, toward true predictive analytics such as forecasting, predictive modeling and optimization techniques. He advised striking a balance of good questions, good data, good analyses and enhanced decision-making. This cycle of processes helps to understand analytic hindsight (what has happened), insight (what is causing this to happen) and foresight (how can we intervene and optimize our outcomes). 

You must know the story your data is telling.

Successful analytic initiatives depend on an organizational commitment to analytic culture, clear definition of what analytics is and isn’t, defined roles and responsibilities, and perhaps most importantly, a focus on the outputs of analytics. Bob Gladden, vice president of the Center for Analytics at Ohio’s CareSource, spoke from a Medicaid health payer perspective. CareSource uses analytics for analyzing gaps in care, predictive readmissions analysis, population segmentation, program evaluation and prioritizing research activities. But this isn’t possible if you don’t know what your data is telling you. 

Gladden argued that analytics output only helps if it tells a useful story. In fact, Gladden cites an emerging trend in the industry where people with journalistic backgrounds are enlisted to tell the story of analytic outputs in a manner relatable to the masses.  

See if CMS can help with your analytics efforts.

The release of CMS-2390-F lays out new guidelines, which will require the use of analytics for states and Medicaid Managed Care Organizations to be best positioned for success. Fortunately states are not alone in this quest. CMS leads several analytics initiatives to support states. 

Kimberly Proctor, technical director of the Data and Systems Group for the Center for Medicaid and CHIP Services, detailed one such program. The Medicaid Innovation Accelerator Program, a four-year commitment launched in 2014, has the “goal of improving health and health care for Medicaid beneficiaries by supporting states’ efforts to accelerate new payment and service delivery reforms.” In addition, a new upcoming initiative due in late 2016 will help states increase analytic capacity through new analytic strategies, data integration and SAS/Statistical programming assistance. You can learn more about these programs on the CMS Innovation Center website.

Consider how analytics-driven policy can improve health care and beyond.

Health-care policymakers can use analytics to explore data, identify existing and emerging trends, and become more prescriptive. Former North Carolina Secretary of Health and Human Services Lanier Cansler cited an North-Carolina-based Medicaid population health improvement initiative that used analytics for care coordination and management efforts, involving patients with multiple chronic conditions.

The initiative resulted in:

• reduction in emergency department visits by 10 percent (aged, blind or disabled [ABD] by 3 percent and non-ABD by 12 percent);
• reduction in inpatient admissions by 11 percent (ABD by 4 percent and non-ABD by 20 percent); and
• reduction in preventable re-admissions by 32 percent (ABD by 34 percent and non-ABD by 29 percent).

Cansler advised using analytics to gain a more holistic view of a patient and going beyond health care inclusive of social determinants. The benefits of which could be used to better inform not just health policy but other social-service-based policy as well.    

Prove your analytics prowess with a specific population before expanding.

Nothing proves the value of analytics better than putting it into action. But you need to be deliberate and strategic. Focus on a specific population where deeper insight would have significant impact. Tara Larson, former acting director of the North Carolina Division of Medical Assistance (Medicaid), and Jeremy Racine, health-care director for SAS State and Local Government, led an exercise focused on the Medicaid diabetic episode of care population. They displayed the power of data using medical episodes of care analytics as the basis for informing policy decisions, controlling costs, enhancing access to care and creating better health outcomes. 

Attendees had the opportunity to evaluate the differences between traditional non-analytic methodologies (such as pen and paper) and advanced analytics. As expected, analytics proved to be more than effective at determining causal factors, identifying potentially avoidable conditions, costs and risk adjusted costs, determining ideal strategic interventions such as treatment or policy changes, and projecting financial impact. This simple exercise helps stakeholders make a compelling case for analytics.

“The new regulations set forward an approach for health care that includes all aspects of a person’s life,” Larson said. “Use of analytics to explore, review and display data from multiple sources is critical for efficient and effective operations.”

The new Medicaid rules are an opportunity to better tie patient outcomes to costs, payments and policy. But that requires huge amounts of data. By following the advice of the UNC seminar participants and continuing to expand the promise and possibilities of health analytics, who knows where we’ll be 10 years from now.

Zach Ambrose is a principal at Ambrose Strategy and works with public- and private-sector clients to bring innovation to the public sector. Jeremy Racine is SAS' state and local government health-care director and evangelizes how analytics empower modernization and innovation across the healthcare continuum. Daniel Gitterman is the Thomas Willis Lambeth Distinguished Professor and chair of public policy at UNC Chapel Hill.

Published in Government Technology on June 16, 2016