• Flavors of cost shifting from the uninsured

    The Supreme Court’s ruling on the Medicaid expansion has made cost shifting from the uninsured a polite dinner topic again. From there it has sloshed over to blogs and Twitter. I’ll just make a few quick points before I head off to work:

    • A round up of empirical evidence on cost shifting from the uninsured is here and more on other types of costs shifting can be found in the FAQ.
    • Cost shifting from the uninsured comprises three different phenomena:
      1. Increased taxes to cover uncompensated care. To the extent that policy can get those being cared for to cough up some financial contributions  for that care (whether by purchase of insurance or direct payment, penalty, tax, whatever), this can be reduced.
      2. Premium-increasing adverse selection. If the uninsured are healthier on average than the insured, this causes premiums to be higher than they otherwise would be if everyone were insured. Some consider this a cost shift, a negative externality. Again, the way to remove it is obvious, but not everyone is agreed it’s worth the cost. Moreover, one could certainly call forced contributions toward health care a cost shift onto the healthy. This type of risk spreading really bothers some people, apparently.
      3. Forced charity. (I don’t know what else to call this.) This type of cost shifting is when hospitals and other providers offset shortfalls in payments from uncompensated “charity care.”* The story is that they increase prices for private payers (the insured and others who can pay full cost) so they can afford charity care. I’m not sure in what sense this makes it charity care since providers are, well, shifting the costs and it sounds like private payers are not voluntarily offering the support. We can discuss that in the comments. In any case, the way to reduce this type of cost shifting is to get those who are in need of the charity to pay something for their care, if possible.
    • I don’t believe (but could be wrong) that the empirical literature on cost shifting carefully distinguishes all these pathways.
    • Notice that all three could be reduced through an expansion of the welfare state. Since that’s taxpayer financed it effectively converts types 2 and 3 to a formal version of 1. We don’t tend to call Medicaid a cost shift, but it really is quite similar to 1 only, in my view, more humane and with increased incentive or at least possibility for preventative care, better care management, etc.
    • Finally, type 3 is not likely to be as large as people like to think it is. The evidence suggests it probably is fairly small. And that evidence may include the effects of 1 and 2. There must be limits to the extent to which a hospital can increase prices. Moreover, hospitals can and do adjust costs more than they cost shift. They also can and do cut back on charity care.

    * To be fair, some of this “charity care” could be required by law under EMTALA. I’m not an expert in the extent to which hospitals are compensated by government (e.g., via special Medicare or Medicaid payments and/or other programs) for their EMTALA care, if at all.


    • It is my understanding the the EMTALA is required to “stabilize,” that is, prevent you from dying in their ER. That’s it. To your point. A bill for whatever services are rendered is still given. As you know, qualifying for Medicaid as a childless adult is very difficult and sometimes near impossible in some states. So we can expect lots of “uncompensated care” happening in hospitals if states don’t participate in Medicaid expansion.

      • They do send a bill. I know that from the sad experiences of a friend, who did not survive the ER visit. I can also tell you from that experience that those bills rarely get paid.

    • Another possibility that hasn’t received much attention might be labeled “political cost-shifting:” Hospitals agreed to Medicare reimbursement cuts under the ACA because they expected a reduction in uninsured patients with the exchanges/subsidies and Medicaid expansion. Now, if a significant number of states decline the Medicaid expansion, it would not surprise me if the hospital lobby argues to repeal or reduce the Medicare reimbursement cuts. In addition, I think the cuts in disproportionate share payments may be conditional on reducing uninsurance rates below a certain threshold. If the Medicaid expansion falters, some of the payment cuts that finance the ACA may be in jeopardy.

    • Is this the same Austin Frakt who wrote “The Health Care Cost Shifting Myth” in 2009? Uwe Reinhardt has also done work on this issue, debunking the cost shifting myth- cited in your earlier blog post.

      Cost shifting is a convenient excuse for hospitals to use when negotiating with private payers. It is nothing more than that. See my Sickonomics blog post at http://www.healtheconomy.net/blog/?p=13 .

      As for how much EMTALA care gets written off — it’s not dollar for dollar. Medicare and Medicaid pay hospitals based upon complex formulas which take into account how much charity care, and uncompensated care they give. The hospital does not “lose” this money; it is factored in to other payments.

    • For not-for-profit hospitals, the uncompensated care they provide is also important in demonstrating their community benefit and allowing them to maintain their tax-exempt status. (I believe the IRS is growing less sympathetic to arguments that research, maintaining a medical school, or meeting a payroll constitute community benefit). Where cash-hungry counties see not-for-profit hospitals sitting on expensive real estate, they have in some cases (for example, sought to have hospitals’ not-for-profit status revoked, exposing the hospitals to substantial tax liabilities. So writing off some charity care is probably a net financial winner, at least for the not-for-profits.

    • Are the estimates of how much hospitals spend on caring for people who cannot pay based on how much those hospitals typically charge for similar services or an the marginal costs of providing additional people with those services? If they are based on the former, is the gap between the two great enough to be of concern if you are trying to estimate the financial impact of caring for people who cannot pay.

    • George — uncompensated care (charity + bad debt) is reported at the rack rate — the “full-established charge” from the hospital’s charge master for the care delivered — multiplied by the hospital’s overall cost to charge ratio. Thus any relationship between this reported “cost” and the true marginal service line cost is entirely accidental.

      Gruber and Rodriguez (2007) [http://economics.mit.edu/files/6423 ] looked at physician uncompensated care and found that it might be considerably overstated, but that this might not apply to hospital uncompensated care (“further study is needed”).

      Hadley et al (2008) provides a summary of how uncompensated care is paid for (http://www.kff.org/uninsured/upload/7809.pdf ) that gets at some of the sausage making.

      If you want actual sausage, here’s the guidance for filing your cost report: (Worksheet S-10): https://www.cms.gov/Regulations-and-Guidance/Guidance/Transmittals/Downloads/R1P240.pdf