The recent Medicare report on variation in hospital “prices” is not exactly news. In fact, I wonder why anyone (including the NY Times and NPR) covered it, let alone make it a lead story.
As you probably know, Medicare reported that hospital charges for specific treatments, such as joint replacement surgery, greatly vary from one hospital to another. (This includes charges for all services during the hospitalization, including room charges, drugs, tests, therapy visits, etc.) Everyone in the healthcare business knows that charges do not equal the actual prices paid to hospitals, no more than automobile sticker prices equal the prices that car buyers actually pay. Except that for the past thirty years, the gap for hospitals greatly exceeds (in percentage terms) the gap for cars. This is not just a nonstory, it is an old nonstory.
So reporters tried to give it a new spin. One angle concerns the uninsured, who may have to pay full charges. I will write about this in a future blog. Another angle is that by publishing these charges, Medicare will encourage patients to shop around. That is the subject of this blog.
I suppose it is okay to tell patients that the amount they might have to pay out of their own pockets may vary from one hospital to the next. But the published charge data is useless for computing out of pocket payments; in fact, it may be worse than useless. As even the NY Times noted, insured patients make copayments based on prices that their insurers negotiate with hospitals. These prices are essentially uncorrelated with charges. So a patient who visits a hospital with low charges may well make higher out-of-pocket payments than a patient who visits a high charge hospital. It is a crap shoot.
Even if charges did correlate with prices, a simple comparison of charges for a given treatment is useful only if hospital care is a commodity. You can compare the prices of a Toyota Prius or my latest book from one seller to another because they are selling identical products. But the cost of treating a patient, and therefore the price of treatment, depends a lot on the severity of the patient’s condition. This can make for very misleading comparisons.
Here is a simple example. Suppose there are two types of patients receiving joint replacements – those with simple problems and those with complications. Suppose that Community General Hospital and Doctors Township Hospital both set prices of $20,000 for simple cases and $40,000 for complicated cases – they have identical prices. But suppose further than 25 percent of CGH’s cases are simple, whereas 75 percent of DTH’s cases are simple. We would report that CGH’s “price” for joint replacement is $25,000, while DTH’s “price” is $35,000. The failure to control for complexity makes the pricing comparison all but useless.
To make matters worse, publishing prices without publishing information about quality may encourage patients to pretend that hospital care is a commodity and choose providers that skimp on quality. As I showed in a 20 year old paper with Mark Satterthwaite, this can also encourage a disastrous race-to-the-bottom where hospitals deliberately disinvest in quality in order to bring down their prices. Medicare has published hospital quality report cards, but these receive little media attention, certainly not like the lavish attention given to this new report on charges. Most consumers remain unaware of Medicare’s hospital quality ratings and, those who do take a look may find the data-filled tables too much to handle. Many consumers will be tempted to shop only on the basis of price.
So what is a consumer to do? It is all but impossible for individuals to comparison shop for the best hospital price; with rare exceptions hospitals cannot and will not tell anyone their prices in advance of admission. (They will tell you their charges.) Fortunately, market forces may be coming to the rescue. As more consumers enroll in health plans with high deductibles and large copayments, there is demand for information about actual prices. And where there is demand, supply usually follows. Some private insurers are beginning to post pricing comparisons on their websites, allowing consumers to determine which hospital is likely to offer lower out-of-pocket costs. Insurers are naturally reluctant to disclose the actual amounts they have contracted to pay hospitals, so this information is somewhat sketchy. Several consulting firms have sprung up to work with self-funded employer-sponsored health plans. They use employers’ own data to help employees find the best prices. To my knowledge, these insurers and consultants are not doing much in the way of risk adjustment; the real world equivalents of my fictional CGH will come off poorly in such comparisons when, in fact, they may offer very good deals. But it is still early days and such refinements to the data are sure to follow.
By intelligently shopping around, employees can save hundreds or even thousands of dollars in copayments for costly hospitalizations. But in order to shop around they require valid data. Employees should demand this information from their employers. A nascent market for pricing data has already formed. It only needs an extra nudge.
And, with apologies to the late Paul Harvey, that is the rest of the story about hospital pricing.