HLM: Is value-based healthcare a particularly challenging sector?
Miller: Every patient is different, but on the other hand, how do health insurance companies operate? The law of large numbers says that on average, patients are fairly similar. You don't have to deliver the exact same treatment to everybody to estimate on average what it is going to be like.
If you get the unusually expensive case—the patient who is an outlier with unique health problems— that is what insurance is for.
On the other hand, saying 'We shouldn't be giving an MRI to everyone who comes in with lower back pain. Most of them should probably go to physical therapy first.' That is something you can do across a broad number of patients. That is going to save money on average and probably be better for the patients.
HLM: Is there common ground for fee-for-service and value-based models that providers can build on?
Miller: A lot of the payment reforms that are being done actually build on fee-for-service. The idea is you don't just leave it in place and try to pile something on top. The problem with fee-for-service now is that it says you get paid the exact same amount to do something whether you do it well or poorly and whether or not [or whether] there are complications or infections that occur. And in fact you may get paid more.
But you don't fix fee-for-service by sticking little penalties or bonuses on top. You have to change the fundamental way it is delivered.
For example, for patients who have health problems, we are looking at payments based on the patient's condition and not based on exactly the procedure you used. A good example is delivering a baby. You get paid more to do a caesarian section than you get paid than a vaginal delivery. Yet the vaginal delivery takes longer, and is better for the mother and the baby.
So why do we now have a 33% C-section rate in the country? Because the fees we pay are not based on the actual value.
John Commins is a content specialist and online news editor for HealthLeaders, a Simplify Compliance brand.