Today the Centers for Medicare & Medicaid Services released new data that for the first time, according to Health and Human Services Secretary Kathleen Sebelius, "gives consumers information on what hospitals charge." It laid out what every hospital in the U.S. bills CMS for the 100 most common Medicare procedures and conditions, and how much the hospital is reimbursed.
The big news, Sebelius informed reporters on a call-in news conference, was that the amounts hospitals bill Medicare charge for the same procedure "vary dramatically in ways that can't be easily explained."
The massive spreadsheet offered plenty of examples. For an uncomplicated spinal fusion, Lourdes Medical Center in Willingboro, N.J., billed Medicare an average amount that approached half a million dollars — $471,000 and change. It was reimbursed about $31,000. Inland Hospital in Waterville, Maine, charged Medicare roughly 1/24th as much, slightly less than $20,000. It, too, was paid about $31,000 on average.
Shocking. But the data aren't new, only updated. CMS had made the data available, if you hunted hard enough for it, for several years. I tracked it in my files.
Nor is the message that hospital charges have little connection with reality new. I wrote that in a blogpost four years ago. (The relevant portion appears below.)
And unless you have a touching belief in the power of moral suasion, publishing the data won't make those prices drop. Why would hospitals do that?
Which means that the data won't help consumers other than those who have to pay list price because they are uninsured and yet somehow can afford to shop around — and travel — so they can pay merely $50,000 or $100,000 out of pocket rather than double or triple that amount.
[Compare U.S. News-rated health plans.]
One more frustration is that the charges displayed only reflect those submitted by the hospital, not by physicians or other providers covered under Medicare Part B. Their bills can jack the cost by more than a little. For a routine spinal fusion, the average surgeon's charge is about $12,000. Jonathan Blum, the CMS official who oversees Medicare payments, said in response to a reporter's question that there's no timetable for rolling out data on payments to doctors — which used to be available — for various procedures. Or for providing data on hospital outpatient charges.
My take on this has not changed from 2009, as is evident from my January 21 post in the blog Comarow on Quality, which preceded Second Opinion. Here's an excerpt from what I wrote then:
In hospitals, cost data is generated in a looking-glass world. Statements received by patients show how much the hospital charged the insurer, Medicare, or other payer for care, how much the hospital received, and how much the patient must pay. The total of those charges equals the sum of individual charges for each procedure, test, drug, and gauze pad. These figures, in turn, are plucked from the Chargemaster, a 3M tool used by almost all hospitals.
The total charges — call them the list or sticker price — has almost no basis in reality. For the most part, the only patients billed for the full list price are those who are admitted through the ER and are uninsured. Of course, the hospital doesn't expect such patients to pay more than a fraction of the bill. All or most of the charges are written off.
Hospitals don't expect insurers to cough up the list price, either. The average national hospital charges for a knee replacement under Medicare came to more than $38,000 in 2007. Medicare paid an average of less than $12,000. It is almost impossible to know what commercial insurers pay hospitals; such arrangements are the product of intense negotiation and vary considerably depending on the hospital, the insurer and economic factors such as location and competition.
Do uninsured patients have any leverage to bargain over their bill? "If somebody came in and said, 'I can't pay full freight,' we would work with them and very likely they would be charged an amount not much different than we would receive from third-party payers," a hospital executive in New York told me. The sum, he said, would be determined by their financial and credit status and payment arrangement. "Do they have a credit card? How much can they pay up front? How good a credit risk are they? We're not going to try to charge them the last dollar."
I don't think the hospital is unique, and I'm not sure how well the new transparency movement to open up procedure and other costs to consumers really measures up. When I checked a couple of pricing sites offered by health insurers to members, the figures I saw were list-price charges. They don't necessarily have much to do with what a provider gets, or what a patient pays.
To solve healthcare problems, we need to measure them first. But with a bunch of you-pick-'em yardsticks for measuring "costs," how can we possibly know how large they are, or the true savings from reducing them?