Have you ever tried to use Medicare data to research a hospital? If so, I bet you've been disappointed.
Let's say you live in Las Vegas and you have heart problems. There are three hospitals nearby. You'd like to know the answer to a simple question. "If I ever have a heart attack where should I tell the ambulance to take me?"
Medicare's Web site that holds this information is called Hospital Compare. If you visit and search for Las Vegas heart attack care, this is what you'll find:
Was that helpful?
As you can see the three hospitals within your search are all rated "No Different than U.S. National Rate." In other words, they are all average. Do the same search in every other city that has more than one hospital and you're likely to find the same thing. Or you might find that there was not even enough data to calculate an average result.
If you want to find a hospital that's exceptional, there's no way to search for it. You can also can't search for a dud.
If you invest more time in the research by clicking down into some of the hospital-specific data, much of it is missing. Here's specifically what you will find in Las Vegas, and that's assuming you can navigate the clunky user interface:
So two out of three measures are missing. And these are just the process measures, which are not the most useful. The best way to rate a hospital is to track results instead of process. Specifically, that means calculating whether patients survive heart attacks at higher or lower rates than would be expected. Medicare does this analysis but will not release the raw data. Instead it grades facilities on such a generous curve that it's hard not to be considered average. Everyone gets a B+.
I tried a similar search for other cities including Sacramento, Philadelphia, Memphis, Fort Lauderdale and parts of Iowa where I live. The same average ranking was all I could find. That's like saying the within a big city it doesn't matter if you stay at a Super 8, a Sheraton or the Ritz.
It was only when I deliberately looked for hospitals I know are good (Mayo's Methodist hospital in Rochester, Minn. or Massachusetts General in Boston) that I found a "Better Than" ranking. Of course Medicare does not disclose how many of the 5,000 or so hospitals nationwide qualify for better-than-average rankings.
I'm not sure what the excuse is here. This is not a brand new site like the Physician Compare, which I panned last year. Medicare bragged more than a year ago about how it had uploaded reems of quality data to its hospital site. And the site itself launched in 2005.
A couple other observations:
*You can't even search for a specific hospital by name.
*You can't rank all hospitals by quality measures as you would be able to on a site like TripAdvisor. (On that site you can specifically look at the best restaurants in Memphis.)
*There's no description of what these measures mean. When Medicare says that a hospital is at the U.S. national average, do they mean between the 55th and the 45th percentile? Or do they mean within 3 standard deviations of the average (99.7% of the total)? I'd guess the latter given the results on the site but have no idea.
*Once you're in the granular parts of the site you can no longer compare the data you're looking at to other specific hospitals.
Since I'm trying to be constructive and not just a grouch, here are a few easily implementable suggestions:
*Invest in the search function. Not sure what that means? Copy the functions of "advanced search" on any other Web site. (Gmail for example.)
*Give numerical ratings to hospitals. Emulate GreatSchools.org, one of my favorite Web sites. Great Schools provides an excellent service to parents by ranking each public school on a scale of 1 to 10 based on test scores and other factors. I'd love to know which hospital is a 10 and which is a 1. So would the hospitals' management teams, directors, and medical staff. Nobody wants to be a 1.
*Stop trying to be the only or best consumer site. Instead provide lots of free raw data or APIs or whatever it takes to proliferate the information. Let other people build Web applications to present and analyze this stuff. HealthGrades and Consumer Reports do a great job already. There's room for more if you lower the barrier for entry.
In a recent interview outgoing Medicare chief Donald Berwick said: "I have come to think that transparency is one of the most important aspects of policy in the nation. I don’t think we’re where we need to be and I think if we can make more information more transparent, almost everything will get better, cost and price."
I'm glad he's come around this view. Ambitious rhetoric is a good starting point. Now build a decent Web site.
