When it comes to health care, the biggest of the big data are all about Medicare.
So, it's kind of a BIG deal when the government releases what individual hospitals charge Medicare — and what they actually get paid — for the most common diagnoses and treatments.
In a first, the Centers for Medicare and Medicaid Services made those figures from more than 3,000 hospitals public Wednesday.
A quick spin through the data shows that what hospitals in the same town or state charge for the same procedure is often very different. And, as previous studies of anonymized data have shown, there are big variations from one part of the country to another.
Now, it must be said, that what a hospital charges Medicare isn't what the government pays. Medicare and private health insurers get discounts and don't pay anything near the hospital rack rate. The uninsured may pay something closer to these master prices, though, because they don't have bargaining power.
But the Medicare data that were just released includes the amount of money the government actually forked over.
Take, Hartford, Conn., for instance, health insurance capital of the U.S. Hartford Hospital, the biggest one in town, charges a tad over $15,000 to treat a Medicare patient diagnosed with chest pain and receives $4,975 for the service. Little Charlotte Hungerford Hospital, in nearby Torrington, charges Medicare about $7,000 and receives $3,713.
Medicare pays teaching hospitals a little more to compensate them for that work. And hospitals in places with higher costs get more, too.
So, for the first time, anybody can get an idea of what their hospitals are billing Medicare for something and how much the government is paying. The database, such as it is, isn't very friendly. But with the data now public, look for outsiders to make it easier to sort and analyze.
The Washington Post has already taken a crack at it. A handy graphic tool lets you look at charges by state for some of the most common medical conditions, including heart failure, joint replacement and pneumonia.
So how useful is the information? If you're a nerd, maybe quite useful. But the data are a slog.
Carnegie-Mellon's Martin Gaynor, a professor of economics and health policy applauds the move, while saying the numbers only tell us part of the story. "Charges are list prices," he says. "They're sticker prices."
The data on what Medicare actually pays show less variation, and, he says, that's expected, given that administrative formulas dictate the amounts. Data on what private insurers pay would help even more, and that's something he's working on through the Health Care Cost Institute.
Still, he says, there's no harm and maybe some good to be had from the Medicare data dump. "Just sort of getting information out there, even if it's not exactly what you want, it may have some use," he says. "It may make hospitals think a little bit harder about what they're doing."
In some places, the details on charges could spark conversations or pushback about how well nonprofit community hospitals are doing at their public missions, he says.