By MERRILL GOOZNER

The Sunlight Foundation today gave us a fascinating first peak at the hospital safety data from the Centers for Medicare and Medicaid Services, which was finally convinced to release the information after years of stonewalling by the American Hospital Association. For the first time, the public can compare less-than-stellar performance at competing local hospitals on key indicators like catheter or urinary tract infections or bed sores.

As their story points out, the data only cover about 60 percent of hospitals since many states, like Maryland, failed to cooperate with the voluntary CMS program. They also caution that any comparison of the raw numbers must take into account the numerous confounding variables that can make one hospital look more slipshod in its practices than another. Some hospitals take in many more older and poorer patients, who are more likely to have multiple chronic conditions that make them prone to complications during their hospital stays.

Yet as Arthur Levin of the Center for Medical Consumers, a New York-based advocacy group, pointed out, “”I think it’s fair (to release the data) as long as everybody agrees on what the limitations are, and what the caveats are. There are those who say this data isn’t ready for prime time and public review. If we waited for perfection, we wouldn’t have anything out there.”

Speaking of limited databases, Medicare has its own website — “Hospital Compare,” which allows the public to review hospitals’ performance on key indicators of quality performance. These are indicators that clinical practice guidelines from professional medical societies suggest are more likely to lead to positive outcomes. They include such things as giving an aspirin when a heart attack patient presents at the emergency room door, or ensuring that patients receive the right antibiotic prior to surgery. Go to the website and check out the hospitals in and around your zip code. You may be surprised by the results.

Will people use these websites? The idea that “consumer-driven medicine” will drive down health care costs would seem to require this kind of information. How can you make an informed choice on how to spend your money unless you have good data on which to compare quality and cost? Yet the error-rate website set up by the Sunlight Foundation is purely about safety, which is only one variable that goes into determining the overall quality of a hospital. And while Hospital Compare does contain some pricing data, it only contains prices that CMS paid, not what the patients paid out of pocket, or, more significantly, what non-Medicare patients paid.

Here, for instance, you can see what three hospitals within 15 minutes’ drive from my home charged CMS for three different types of heart attack cases. It turns out that the one farthest from my home had the cheapest rate for one of the procedures, while all three charged about the same for the other two. If, God forbid, I find myself clutching my chest in pain tomorrow morning, and tell my wife to rush me to the hospital, am I going to tell her to go to the one that is 15 minutes away because it’s cheaper, or the one that is five minutes away?

These databases are great tools for patient advocacy groups pushing to improve local performance. They can be used by hospital administrators and physicians to spur on constant improvements at their hospitals to win a race to the top with their peers. They can also be used by accreditation agencies and regulators to measure and reward better performance.  What they’re not very good for is comparison shopping by patients, who are ill-equipped to make sense of their undigested data.

Merrill Goozner has been writing about economics and health care for many years. The former chief economics correspondent for the Chicago Tribune, Merrill has written for a long list of publications including the New York Times, The American Prospect, The Washington Post and Financial Times. You can read more pieces by him at GoozNews, where this post first appeared.