COVID-19, Medicaid block grants and other notes

COVID-19 is an excellent example of potential shock for Medicaid financing in a block grant context.

If there is a widespread community infections and plenty of people unexpectedly being admitted to hospitals for significant care, things could get expensive fast in an unanticipated manner.  Commercial insurers will pass most of the cost to reinsurers and the federal government will be able to readily absorb the Medicare and VA bumps.  However, Medicaid will likely be a significant payer in a plausibly bad scenario.  Legacy Medicaid eligibility categories will have states pay between a quarter and half the incremental costs while the feds pick up the rest.  Adults who are covered by Medicaid expansion will see their state pay a tenth and the feds the rest.

A few weeks ago, CMS released a letter to state Medicaid directors inviting them to ask for block grants for the adult, non-blind/aged/disabled population. Two flavors of block grants were available; the first was a total program block grant which would grow at CPI-M+.5 and the other flavor was a per-capita block grant which would grow at CPI-M.  The initial allocations for both flavors would be based on trended forward historical data.  Historical data, by its definition, does not contain new shock information.  A block grant for Medicaid transfers shock risk from being split between the feds and the states to entirely the states.  Expensive, external shocks would quickly deplete the federal funds and force the state to either find additional state funds or cut back services dramatically.  CMS has said that in an emergency, the block grant could be altered for special circumstances:

Special Circumstances Adjustment

Recognizing the dynamic health care landscape in which state Medicaid programs are operating, CMS will provide states with the opportunity to propose updates to an approved HAO demonstration to account for any changes to projected expenditures or enrollment in the current demonstration year due to unforeseen circumstances out of the state’s control, such as
a public health crisis or major economic event. Under such circumstances, states will have the opportunity to submit new information and relevant data, describe the circumstances and
proposed amendment, and renegotiate relevant STCs. The data provided by the state will be validated by the CMS Office of the Actuary in consultation with other appropriate federal
entities.

There is a lag and an uncertainty on the part of a state as to whether or not a public health crisis will have additional federal funding under a proposed Medicaid block grant. Under current rules, the new, unexpected claims are submitted to CMS and the funds show up in the normal course of business. The uncertainty would cast a pall on decisionmakers if block grants had already been elected. That uncertainty would be much higher in the world where Cassidy-Graham passed and big chunks of Medicaid were block granted as those changes would require Congressional action which, as we have seen in Puerto Rico which does operate a block granted Medicaid program, is not guaranteed to be quick nor sufficient.

Moving a little wider in the field, I think that states which have Medicaid expansion for the working poor will be in better shape than states that have not expanded Medicaid, all else being equal. The coffee shop worker or the home health aide still can’t afford to take pre-emptive time off but in Medicaid expansion states, they can afford to get checked out which helps with surveillance, isolation and treatment efforts.

Finally, John Graves of Vanderbilt raised one hell of an interesting and important point yesterday; timing of a potential crisis is not good.

The U.S. experience with a pandemic early in the calendar year will be uniquely challenging given that the vast majority of Americans have privately administered plans that (increasingly) rely on deductibles/cost-sharing as a fairly blunt utilization-management tool.

— John Graves (@johngraves9) February 26, 2020

A 1st Quarter crisis looks very different than a 4th Quarter crisis as the number and characteristics of folks who have no remaining cost sharing as a barrier to care looks very different in the Q1 instead of Q4.