Uncertainty in Medicine: the Last Refuge of a Scoundrel

The way we pay hospitals, doctors, and other health-care providers can have a big impact on how aggressively they try to reduce medical errors. That is one of the themes of my “Universal Coverage Kills” oped, where I argue that Medicare locks most of the market into a payment system that actually rewards medical errors. (You read that right, rewards.) Last month, Medicare launched a new policy that attempts to reduce medical errors by forcing providers to bear the costs of what it calls "never events."

The blogger WhiteCoat was among the many who responded (sometimes angrily) to that oped, and has since responded to my response to those responses. (And here I go, responding...)

WhiteCoat still believes I support Medicare's "never events" policy and that I want to lump everyone into a "national HMO." I'm not sure why, given that I advocate "a market free to experiment with different payment systems" and write, "We need experiments with different payment policies to see which produce the best outcomes for patients, and the rigidity that government brings to that process is downright harmful."

WhiteCoat does, however, raise two interesting issues.

1. Should payers attempt to punish medical errors when it's hard to tell whether anyone committed an error?

All agree that providers should bear the costs of obvious medical errors. But what if it's unclear whether an adverse event was the result of error?

Clostridium difficile-associated disease (CDAD) is a growing problem in hospitals. Sometimes, hospital workers pass the bacterium to patients. Other times, patients bring it in with them. (An estimated 3 percent of the adult population already has C. diff living in their intestines.) Who should bear the cost of CDAD in hospital patients?

WhiteCoat argues that forcing providers to bear the cost inevitably punishes them for cases of CDAD that weren't the providers' fault. Since that would be unfair to providers, WhiteCoat (unless I'm misreading him) supports fee-for-service payment in the absence of obvious errors.

In the name of fairness to providers, however, that payment system would be decidedly unfair to payers. Patients, insurance purchasers, and taxpayers, would have to pay for CDAD cases where a provider was at fault.

Then again . . . if providers bear the cost, they will avoid high-risk patients who are likely to suffer such adverse events.

Then again . . . if payers bear the cost, providers will have no financial incentive to reduce medical errors.

Here's the point: there is no obviously superior way of paying doctors, just as there is no obviously superior way of paying managers, CEOs, taxis, lawyers, roofers, etc. Every payment system involves tradeoffs. That's why we need competition between different payment systems: "to see which produce the best outcomes for patients." My guess is that, with all government does to encourage fee-for-service and discourage prepayment, we've probably erred on the side of the former and that the latter would be more prevalent in an open marketplace.

For providers to say, "We must have fee-for-service wherever there is uncertainty about the cause of an adverse event," is self-serving crap. Cries of "uncertainty in medicine!" are often the last refuge of a scoundrel.

2. Who is most likely to discover new ways of avoiding medical errors?

I'll give you a hint: it ain't me.

Sure, I could tell you that providers should pay closer attention to their patients, wash their hands more often, deploy electronic medical records, e-prescribe, use bar-coding for medications, use standardized checklists more often, blah, blah, blah.

But let's assume that health-care providers are currently using everything tool technology offers to prevent medical errors. (Stop laughing. C'mon, this is serious.) How can we generate innovations that allow us to prevent adverse outcomes that were once thought unpreventable?

I'll give you a hint: M-O-N-E-Y.

Sure, most providers already do whatever they can to avoid medical errors. They're good people. But under fee-for-service payment, reducing errors means they take a financial hit, individually and institutionally. (Some thanks.) We're just not going to get as much error reduction under pure, blind fee-for-service as we would if providers could profit by reducing errors.

The beauty of prepayment (combined with an integrated delivery system) is that not only does the provider profit from eliminating adverse events that are known to be preventable, but the provider also profits by finding ways of preventing adverse events that were once thought unpreventable.

Prepayment thus has the potential to generate new knowledge about how to save lives. Pretty cool stuff, I know. That's economics for you.

I'll close by suggesting that everyone who still thinks I support Medicare's "never events" policy or a "national HMO" should go back to the top. And. Re-read. Carefully.