Mediating churn, Arkansas edition

by Adrianna McIntyre

Arkansas thinks that their “private option” plan for expanding Medicaid—putting newly-eligible individuals into the state health exchange using Medicaid dollars—will provide buffer against individuals churning between coverage types. And a study recently published in the New England Journal of Medicine suggests that they might be right. The unresolved question: at what cost?

Options available to other states are closed off to Arkansas. Throughout the nation, 70% of Medicaid beneficiaries have plans administered through managed care organizations (MCOs), private companies that are contracted to coordinate care . States like CaliforniaTennessee, and Nevada are using their Medicaid managed care infrastructure to protect folks whose income mobility could make them vulnerable to churn; these are sometimes called “bridge” plans. By inviting MCOs to offer private plans on the exchanges—plans that are comparable to their existing Medicaid contracts—individuals can keep core coverage even as their Medicaid eligibility changes. It may not be seamless, but such strategies will offer some continuity of providers and benefits. Arkansas doesn’t have a managed care infrastructure to capitalize upon. That’s why the “private option” is so appealing to the state; it’s a different way to circumvent gaps in coverage.

Benjamin Sommers and Sara Rosenbaum crunched the numbers and found that Arkansas’s proposed premium-assistance model could prevent up to two-thirds of predicted churn in the state, relative to a “traditional” expansion.


The graph consideres three different scenarios: a “Baseline Affordable Care Act”, where coverage changes with eligibility when income fluctuates over the 138% FPL line; their findings suggest that this could affect up to 50% of enrollees during a calendar year. It’s important to note that this is not meant to model churn outcomes in states with the aforementioned “bridge” programs.

Compared to baseline reform, the proposed premium assistance model results in dramatically less churn for Arkansas and Ohio (the state had previously been negotiating with the federal government about a premium assistance approach, but is not on track to expand Medicaid for 2014). This assumes that existing eligibility criteria for parents and disabled adults remains the same, with all other adults below 138% FPL directed to the exchanges. Arkansas benefits to a greater extent because existing eligibility is very narrow; a greater proportion of adults will find themselves enrolled in the “private option” and less prone to churn. Of note: this reduction is is predicated on the assumption that private-option beneficiaries will not experience any disruption in benefits or provider networks when the source of funding for their plan (Medicaid vs. subsidies) changes, which is not guaranteed.

The Arkansas plan requires a waiver from the federal government to move forward. The state’s Department of Human Services released a draft of their 1115 waiver a few weeks ago (you can find a summary and link to the full waiver here). The punchline is that criteria for determining cost comparability (a condition of waiver approval) is still “to be determined.” Arkansas officials have consistently maintained that the private expansion will not cost more than a traditional expansion would. The feds will likely work with state officials to determine the standards for assessing budget neutrality. Even so, Sommers and Rosenbaum caution:

… [C]ost may remain the largest obstacle to a premium-support model … Some analysts argue that private insurers in a competitive marketplace will find ways to save thousands of dollars per beneficiary, but no evidence from either the private insurance or Medicaid managed-care markets supports such claims. The DHHS’s recent guidance does not elaborate on existing federal regulations prohibiting states from purchasing private plans unless costs are “comparable” with those of traditional Medicaid. But unless private insurers reduce provider payments to Medicaid rates (or close to them), it’s unclear how most states could meet that standard.

Adding to the uncertainty, Arkansas will not have competitive bidding during the first year. Under competitive bidding, only certain, less-expensive plans would be fully covered for new Medicaid beneficiaries; without it, enrollees will be insensitive to price signals—they can select any silver plan and Medicaid will cover its costs. Despite these concerns, it seems unlikely that HHS will deny the Arkansas waiver on financial grounds. The waiver is time-limited, terminating in 2017.

It also provides an opportunity: existing literature on outcomes associated with churn is thin, and studies comparing Medicaid coverage to private insurance suffer notoriously from selection bias. If states take initiative in collecting data, Arkansas might prove a natural experiment on both fronts.


Click here to subscribe, for the latest news in and around health policy. 

Adrianna is a graduate student in public policy & public health at the University of Michigan. Follow her on Twitter @onceuponA.


2 thoughts on “Mediating churn, Arkansas edition

  1. Brad F says:

    Suppose a single mom receives Medicaid in AK in 2013 (destitute). Due to a life change, in 2014 she begins to earn >138% through a new job. She loses the job 6 months later.

    Assuming the waiver goes into effect, does she reinstate into traditional Medicaid, or does she refresh into Medicaid 2.0 via the exchange?

    • My understanding is that the exact way that would work out will depend on reporting requirements for income eligibility, which will be determined by each state. According to the waiver application, Arkansas intends to maintain existed income thresholds for traditional Medicaid (<17% FPL for parents). If a mother moves across that line between reporting periods, I believe she would churn between the two coverage types. This is (and income fluctuations above 400% FPL) is why some ~15% of Arkansas beneficiaries are still expected to experience coverage disruption over a two-year period.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: