The Potential of a Population Health Strategy to Improve Healthcare Outcomes and Reduce Costs for Medicaid Programs

The rising cost of healthcare in the United States is a concern not only for individuals and families but also for corporations and both state and federal governments. This is particularly true of Medicaid programs across the country. The ever-escalating costs of medical goods and services are driving up the cost of Medicaid programs and making it difficult for state governments to fund their portion of the program.¹ State legislatures are increasingly faced with making difficult choices of fundingthe Medicaid program at the expense of other, equally important priorities.

Partly driven by healthcare funding pressure, as well as the Affordable Care Act, the concept of the “triple aim” was proposed by Berwick and colleagues in 2008.² The triple aim describes 3 linked and concurrent goals for healthcare delivery: improved care for individuals, improved health of populations, and reduced per-capita costs. Medicaid managed care programs have adopted the concept of the triple aim and attempt to achieve these goals by case management, which directs Medicaid recipients to the most appropriate setting for the healthcare that is needed, limiting duplicative or unnecessary procedures and relying heavily on generic drugs. While this approach has seen positive results in containing costs, it is still unclear whether the overall health of the population is improving. Perhaps a slightly different focus could produce even greater results. That focus incorporates the concept of population health.

The term population health was coined by Kindig and Stoddart in 2003.³ Their original definition was concerned with the health outcomes of a group of individuals. While there is no single overarching definition of population health, the focus on health outcomes seems to be the unifying factor in this still-emerging discipline. For me, the whole tenet of population health is to understand the health risks of individuals and design interventions that mitigate the risk in an attempt to halt or stabilize the progression of the health condition or disease moving forward. Keeping people healthy by focusing specifically on their health conditions should lead to a healthier population and subsequently a lower cost of care. How do we do this, and where do we begin?

The first step is to understand the current health status of the individuals enrolled in the program. On the surface, this may sound like a fairly easy task, but it is not (Figure 1). Most Medicaid programs have large quantities of data residing in their Medicaid Management Information System (MMIS) regarding claims payments for their beneficiaries. This data set is very important to the understanding of the health status of the individual. However, no matter how well this data set is mined, it will not be sufficient to get a clear picture of the overall health status of the individual. MMIS data will provide some information, but it is not the whole picture.

For a true health-status picture to emerge, the MMIS data (which contain medical claims data for point-of-care visits as well as pharmacy data) must be merged with available electronic health record (EHR) data. Once this is done, a clearer picture of the health status of an individual will begin to emerge. These individuals can then be stratified into risk categories based on their overall health status. Those with the least risk for adverse health outcomes or disease progression can be stratified on one end of the spectrum and those with the greatest risk for adverse health outcomes on the other.

Once this is done, gaps in care can be identified and addressed. In addition, targeted strategies for interventions can be developed and implemented. This sounds logical and practical, right? But it is easier said than done, particularly in the Medicaid population. Eligibility for Medicaid is based on resources relative to some measure of the federal poverty level or disability status. The social determinates of health, such as food insecurity, the availability of fresh fruits and vegetables, adequate housing, safety concerns, limited educational opportunities, and environmental factors all work to negatively affect the health status of Medicaid beneficiaries. A population health approach in a middle-class or more affluent population is a challenge; add in the complex interaction of poverty and health, and it becomes an even greater challenge for those receiving healthcare benefits through a Medicaid program.

The Mississippi Division of Medicaid began investigating the potential of a population health approach to achieve better health outcomes and reduced cost in 2014. The Medicaid program continued to be a significant cost driver in Mississippi’s state budget, in part because of enrollment growth (Figure 2). (According to the Kaiser Family Foundation, in 2017 Medicaid accounted for 12% of the state’s general fund spending and 46% of all federal funds directed to the state.⁴) Newer, more innovative solutions had to be identified to control costs.

The public investment in the digitization of medicine, including Meaningful Use and Interoperability funding, set the stage to enable the Division of Medicaid to leverage clinical data generated through the delivery of care. By combining clinical EHR data and claims data from the MMIS, the stage was set to monitor health outcomes of Medicaid beneficiaries as well as the progress of the managed care plans to produce improved outcomes.

The Mississippi Division of Medicaid partnered with the Delta Health Alliance (DHA) to perform a pilot program called the Mississippi Delta Medicaid Population Health Demonstration Project. The Mississippi Delta region is one of the most impoverished regions in the country, with a large Medicaid population⁵ and a high disease burden.⁶ DHA is a nonprofit 501(c)(3) whose mission is to research and identify the causes of poor health and lack of adequate educational opportunities as well as inform residents of the Mississippi Delta on how to adapt a healthier lifestyle.

Because the disease burden is high and the health status of the region is poor, this demonstration project had multiple disease conditions to choose from. In addition, because the region is known for its poor health outcomes and lack of infrastructure, any model developed under these difficult conditions should be able to meet the challenges of any other region of the state.

One of the key elements of the study, as well as the power of the population health approach, is to identify individuals at risk for disease progression and implement strategies that will mitigate or perhaps reverse that progression. Two conditions were chosen to explore in this study, the progression of chosen to explore in this study, the progression of diabetes from its precursor form, prediabetes, to full blown type 2 diabetes (T2D) and preterm delivery. The incidence of obesity and the subsequent burden of T2D in the Mississippi Delta region is high. A clinical profile was developed of an individual progressing along the path of prediabetes to T2D. This profile was used as a template to identify those individuals meeting these clinical criteria.

The demonstration project used a proprietary population health platform developed by the Cerner Corporation to create a longitudinal record that contained Medicaid claims data and clinical EHR data for Medicaid beneficiaries in select clinics in the Mississippi Delta region. Using a predictive algorithm, risk scores for progression of prediabetes to T2D were developed for Medicaid beneficiaries receiving care at the select clinics. Targeted interventions were applied through intensive care management techniques.

Mississippi has a high burden of preterm birth. Preterm birth refers to babies born before 37 weeks of gestational age. The National Center for Health Statistics reports that 9.85% of the births in the United States are considered preterm.⁷ The preterm country, at 13.6%.8 Data at the Division of Medicaid show that the preterm birth rate for women receiving benefits is 18%.⁹ Therefore, the Medicaid population accounts for a disproportionate share of the high preterm birth rate in Mississippi.

Although the cause of approximately 50% of the preterm births is unknown, a wide variety of risk factors are known to play a role in the birth outcomes. Some of the known causes of preterm birth are vaginal infections, short intervals between pregnancies, maternal stress, anxiety or depression, smoking, obesity, diabetes, and low socioeconomic status. In addition, African American mothers are almost twice as likely to have a preterm infant as white mothers in the United States.¹⁰

Prenatal care is integral to lowering the risk and rate of preterm births. Often access to prenatal care is tied to insurance coverage. It is worth noting that the socioeconomically disadvantaged women with higher risk factors for preterm birth do not become eligible for coverage until they are pregnant. Access to and timely attainment (in the first trimester) of care for women who were previously uninsured depend on a rapid enrollment for Medicaid benefits. Unsurprisingly, prenatal care in this population is often delayed.

Using the population health platform described above, longitudinal records were created, and predictive algorithms were applied for risk of preterm delivery for women who became pregnant. These women received intensive prenatal care as well as education that was related to risk factors such as infection, smoking, stress, and other lifestyle risks.

While the study is still ongoing, preliminary results indicated a considerable cost savings through the application of population health tools. Preventing the progression of prediabetes to T2D is estimated to save approximately $9700 per year per individual.¹¹ Although the estimates vary on the cost savings of preventing preterm delivery, Mississippi Medicaid data show the average stay in the newborn intensive care unit for a preterm baby costs approximately $57,000.

These preliminary results, which are exciting, need to be verified and repeated not only in other settings but also with other health conditions and disease states such as hypertension, heart failure, asthma, and metabolic syndrome. The care management model used in this approach is significantly high touch, but it is important to weigh the cost of the intervention against the healthcare outcomes achieved.

There is a heightened interest in the healthcare outcomes of managed care programs from the federal regulatory body, the Center of Medicaid and CHIP Services (CMCS), that oversees the Medicaid managed care programs. The recently promulgated Medicaid Managed Care Final Rule¹² puts a great deal more emphasis on state oversight and health outcomes. These new rules indicate that the federal government is now interested in the cost-to-benefit ratio and whether managed care can improve health outcomes and lower costs as a result. As the cost of healthcare and the cost of the Medicaid program at both the state and federal levels increase, there is likely to be further emphasis on cost containment as a product of improved health outcomes. Population health platforms offer a new tool that can be applied to achieve these results.

Of anecdotal interest is the fact that CMCS was very receptive to the concept of a population health approach to improving quality and outcomes in the Medicaid program. When the Mississippi Division of Medicaid submitted its Implementation Advanced Planning Document for enhanced funding for its population health initiative, it was received with great interest and approved within a 4-week time period, a short turnaround for proposals of this type.

In 2017, the Mississippi Division of Medicaid completed a reprocurement of its managed care program, adding contract language that requires managed care plans to implement a population health-based strategy as part of its ongoing operations. While debate continues over who should deliver managed care to the state’s Medicaid recipients, the future seems to point toward a population health approach.

While the program is still in its infancy, there is significant potential for population health platforms to improve healthcare outcomes, which can not only lead to healthier, more productive individuals but also substantially lower the cost of healthcare in Medicaid programs. Financial Disclosure

There are no conflicts to disclose.

About the Author

David J. Dzielak, PhD, served as executive director of the Mississippi Division of Medicaid from January 2012 to December 2017. He is an integrative cardiovascular physiologist whose research has focused on the neural control of the circulation and the role of immune mechanisms in cardiovascular disease. He received his PhD from the University

of Mississippi Medical Center (UMMC) in 1982 and returned to the medical school in 1987 after working in the private sector. At the time of his appointment as Medicaid director, he was UMMC’s associate vice chancellor for Strategic Research Alliances; he helped the medical center secure more than $74 million in congressional funding and played a major role in promoting technology transfer and biotechnology development with academic and business partners.References

1. Rudowitz R, Gifford K; Henry J. Kaiser Family Foundation. Medicaid moving ahead in uncertain times: findings from the annual Kaiser 50-state Medicaid budget survey. kaiserfamilyfoundation.files.wordpress.com/2017/10/medicaid-budget-event-slides-10-19-17.pdf. Published October 19, 2017. Accessed February 2, 2018.

2. Berwick DM, Nolan TW, Whittington J. The triple aim: care, health, and cost. Health Aff (Millwood). 2008;27(3):759-769. doi: 10.1377/hlthaff.27.3.759.

3. Kindig D, Stoddart G. What is population health? Am J Public Health. 2003;93(3):380-383.

4. Medicaid in Mississippi. Henry J. Kaiser Family Foundation website. files.kff.org/attachment/fact-sheet-medicaid-state-MS?utm_campaign=Revue%20newsletter&utm_medium=Newsletter&utm_source=MHA%20Revue. Published June 2017. Accessed February 2, 2018.

5. Percent of children covered by Medicaid/CHIP by congressional district,2015. Georgetown Health Policy Institute website. ccf.georgetown.edu/map/2015-children-congressional/. Accessed February 5, 2018.

6. County data: diabetes, obesity, and leisure-time physical activity. Centers for Disease Control and Prevention website. cdc.gov/diabetes/data/countydata/statecountyindicators.html. Updated May 16, 2016. Accessed February 5, 2018.

7. Martin JA, Hamilton BE, Osterman MJK; National Center for Health Statistics.Births in the United States, 2016. www.cdc.gov/nchs/products/databriefs/db287.htm. Updated September 26, 2017. February 5, 2018.

8. 2017 premature birth report card. March of Dimes website. marchofdimes.org/mission/prematurity-reportcard.aspx. Published October 2017. Accessed February 5, 2018.

9. Data on file.

10. Preterm birth. Centers for Disease Control and Prevention website. cdc.gov/reproductivehealth/maternalinfanthealth/pretermbirth.htm. Updated November 27, 2017. Accessed February 5, 2018.

11. Delta Health Alliance website. deltahealthalliance.org/project/medicaid-health-demonstration-project/. Accessed February 5, 2018.

12. Centers for Medicare & Medicaid Services. Medicaid and CHIP managed care final rule. medicaid.gov/medicaid/managed-care/guidance/final-rule/index.html. Updated December 2016. Accessed February 2, 2016.

Download PDF: The Potential of a Population Health Strategy to Improve Healthcare Outcomes and Reduce Costs for Medicaid Programs