Skip to main content

How to Use Analytics for MCO Oversight

Today, about 90 percent of Medicaid beneficiaries are enrolled in some type of managed care. In theory, contracting with managed care organizations (MCOs) reduces healthcare costs for the states, while also improving the quality of care for beneficiaries. But how do states know for certain that they are getting the return on investment that they expect? MCO analytics can provide an excellent tool for gaining insight into state Medicaid programs.

Of course, as states look for MCO analytics, they should not be searching for just a reporting dashboard — although that’s certainly part of it. An effective MCO analytics portfolio should be configurable, with filters that provide the ability to drill down into a multitude of parameters: time frames, populations, geographic areas, etc. Then, the solution should look beyond the numbers to help states understand what is going on and what they should do about it. In other words, states should be able to gain actionable insights that will drive performance improvements across all their populations.

Medical Cost Analytics, Trends and Benchmarks

One area in which states see tremendous variation in MCO performance is in how well the MCOs manage medical costs. With a limited amount of money to allocate for their Medicaid programs, states need to have a window into how their money is being spent. To that end, being able to compare MCO performance is vital. Are the MCOs doing a good job of controlling the different areas of spend (i.e. inpatient hospital stays, outpatient surgery, emergency room visits, pharmacy costs, etc.)?

Each MCO will have different memberships, so raw numbers aren’t particularly helpful to the states in making comparisons. Instead, an effective analytics portfolio can help normalize costs so states can make comparisons across the MCOs. Once states have a solid understanding of medical costs and trends, they will know, for example, if an MCO request for a rate increase is appropriate.

Claims vs. Encounter Comparisons

Encounter data from MCOs can sometimes be hard for states to understand or analyze. So, when claims data is available, states can use MCO analytics to compare claims to encounters and determine if there are discrepancies in the information they’re seeing. In essence, MCO analytics can help states perform a quality check on encounter data from the MCOs.

FFS vs MCO Program Comparisons  

Comparing fee-for-service (FFS) program performance against MCO performance has traditionally been problematic, as FFS programs are usually carveouts that contain sicker patients or specialty populations. By providing risk-score-adjusted comparisons and normalizing by membership, MCO analytics can help states then compare apples to apples.

Provider Network Evaluation (Access and Value Optimization)

Having true access to care has always been a problem for Medicaid beneficiaries. Do members in a particular plan have easy access to primary care providers? Are members receiving the appropriate level of care for chronic conditions? MCO analytics can help states see how providers are doing in these critical areas.

Social Determinants of Health Overlay

Many states are now looking to their MCOs to address social determinants of health (SDOH) and reduce any inequities their populations might face. With an overlay of SDOH data — when available — MCO analytics can unearth information to bring about more positive outcomes. For instance, if the diabetic population in a certain health plan is not doing well, what contributing factors might be involved? Do these members have access to good nutrition, or are they living in a food desert? As states launch these SDOH initiatives, MCO analytics could be a helpful tool in measuring effectiveness.

Quality of Care Outcomes

While the Healthcare Effectiveness Data and Information Set (HEDIS) is a useful performance improvement tool, the data used in a HEDIS report can sometimes be a couple of years old by the time states see it. By implementing their own oversight analytics, states can use claims-based measures to gain an almost real-time window into at least some of the quality measures that affect health outcomes. With a better feel for how quality of care is going across health plans, states can determine where meaningful improvements can be made.

Risk Analytics (Predictive and Prescriptive)

States and their MCOs have a limited amount of resources to use for outreach, so who should they focus their efforts on? Traditionally, payers have looked at the people who were the sickest in the past and made assumptions that those would be the ones who would need interventions in the future. But a significant portion of the people who were high utilizers of health resources last year are not necessarily the same group that will have problems in the upcoming year. Predictive analytics that use artificial intelligence modeling and machine learning can help predict who the high-risk patients will be in the future. Then, prescriptive analytics can help states determine why these members are high risk and what should be done to help them.  

It has often been said that if you can’t measure something, you can’t change it. Measuring MCO performance gives states a way to understand what’s going right with their Medicaid programs and what could be improved. And, armed with solid information, states will then have a powerful opportunity to make policy changes that allow best practices to be carried out across their entire Medicaid portfolio.