The “Big Five” Medicaid managed care companies—Centene, CVS Health (Aetna), Elevance Health (formerly Anthem), Molina Healthcare, and UnitedHealth Group—have reported their Medicaid enrollment for the quarter ending September 30. The downward trend that began in Q2 continued for all of the companies except Molina, which experienced a small net increase (+16,000). Centene reported the largest quarter-to-quarter decline (-819,000), followed by Elevance (-741,000), UnitedHealthcare (-290,000), and CVSHealth/Aetna (103,000). (Table 1).
These enrollment declines are a net of offsetting increases and decreases. They were driven largely by the unwinding of PHE continuous coverage, but not entirely. For example, Molina reported an increase in Medicaid enrollment in Iowa of 180,000 during the quarter resulting from the state’s decision to increase the number of MCOs from two to three. Molina’s gain came at the expense of the other two MCOs in the Iowa Medicaid market, Centene and Elevance. That increase, combined with the acquisition of an MCO in another state, offset 200,000 in losses from that Molina management attributed to “the net impact of redeterminations,” resulting in a small Medicaid enrollment increase in the quarter. (None of the other companies provided estimates of enrollment losses due solely to the unwinding.)
As shown in Figure 1, these underlying dynamics do not change the overarching narrative. Medicaid enrollment for the Big Five in total grew from 30.1 million in March 2020 to 44.2 million in March 2023, and, six months into the unwinding, has fallen to 41.7 million, a drop of 2.5 million. This reflects the overall decline in national Medicaid enrollment since the unwinding began, whether measured as total disenrollments (10 million) or net disenrollments (5.3 million).
The unwinding is only six months old. The pace and execution varies from state to state, but nationally a third of all redeterminations have been conducted. In many states, rates of procedural disenrollment—terminations of coverage when a state has not actually made an eligibility determination based on the current circumstances of the individual—are unacceptably high, especially for children. CMS has already intervened to require some states to pause disenrollments and reinstate those incorrectly disenrolled, but much remains uncertain about exactly how the unwinding will play out in the coming months.
For their part, some Big Five managements are telling financial analysts on the Q3 earnings calls that, while they expect net losses in Medicaid enrollment by the time the unwinding ends next year, they also expect a significant percentage of those whose coverage is terminated to re-enroll. (Curiously, recent Congressional oversight letters to the Big Five and two nonprofits requesting information on denials of prior authorization did not come up on any of the calls). Those who spoke in some detail on the unwinding reported procedural disenrollment rates at 70 percent or above, consistent with what CCF tracking has found.
The Elevance CEO described the situation as follows:
“In our Medicaid business, rates are actuarially appropriate, but we are absorbing a membership headwind related to the pace of Medicaid redeterminations, especially in states that have adopted accelerated timelines. Nearly three quarters of all Medicaid beneficiaries terminated in our markets to date have lost coverage for administrative reasons, and 37 percent of the attrition from our own health plans has been driven by individuals under 18 years of age, many of whom may still be eligible for Medicaid benefits.”
During the Molina earnings call, the CFO presented the following data:
“Across our states, approximately one-third of our members reviewed had been terminated, of which over 70 percent have been procedural disenrollments rather than due to verification of actual ineligibility. As a result, we are seeing nearly 30 percent of those terminated being reconnected and we expect these numbers to grow. In the quarter, we estimate that we’ve lost approximately 200,000 members due to the net impact of redeterminations, bringing the year-to-date figure to approximately 300,000.”
The Molina CFO went on to posit that re-enrollments were likely to be high in part because of the high rates of procedural disenrollments:
“Given the high number of procedural terminations, and increasing state and CMS interventions, we expect reconnects are likely to continue, decreasing currently reported membership losses.”
The assumption seems to be that individuals who leave an MCO because they are terminated from Medicaid are not leaving due to dissatisfaction with the MCO and are therefore more likely to return to the MCO. That is not implausible. Of course, for this scenario to unfold, individuals terminated from Medicaid for procedural reasons would first have to re-establish eligibility for the program (or be reinstated in the program by the state). Depending on the bandwidth of the state Medicaid agency to process new applications at the same time as it conducts redeterminations, that could be much easier said than done.
The best path forward would be for states to reduce the rate of procedural terminations so that only those children and adults who are actually ineligible are disenrolled. That would lower the number of uninsured, decrease the burden (and cost) of administering this eligibility churn for state Medicaid agencies, and reduce the disruption for MCOs and their network providers. As our colleague Allexa Gardner has explained, there is no shortage of strategies available to states for ensuring continuity of coverage and minimizing procedural disenrollments.
September 30, 2023, is a point in time, and the unwinding data that the Big Five have reported need to be understood as such. We are in the midst of an enrollment event with no precedent, and much remains unknown. Will procedural disenrollments remain high? If so, how many of those disenrolled will re-establish Medicaid eligibility, how long will it take them to do so, and in which MCO will they enroll? Among those disenrolled because they are actually ineligible for Medicaid, how many will obtain coverage in CHIP, in the Marketplace, or through their employers, and how long will their gap in coverage last? The answers to these questions will have important implications for the Medicaid, CHIP, Marketplace, and commercial businesses of the Big Five, but they are several points in time away.