The federal government has just announced that it will let more light shine on a great trove of Medicare data — medical claims records from its 47 million patients. And in particular, it will let employers, consumer groups and insurance companies mine those mountains of claims to help them rate the performance of individual doctors.
So does this mean that something on the scale of TripAdvisor.com — or Rotten Tomatoes! — is coming to American medicine? Not any time soon.
What does it mean? I asked Dr. Tara Lagu, a research scientist at Baystate Medical Center and an assistant professor at Tufts University School of Medicine, who studies the effects of making health quality data public.
Her bottom line: Probably within a couple of years, patients will be able to determine a given doctor’s volume — how many cases like theirs the doctor treats every year — and possibly quite a bit about how well those cases have gone. But quality ratings themselves are likely to remain contentious, and the big question is whether patients will be able to take advantage of the newly available data.
Our conversation, lightly edited:
My main take on this new Medicare announcement has to do with how patients will digest the data. The great potential of this could be lost if it’s not presented to patients in the right way. We really could be missing an opportunity to help patients participate in the quality improvement process.
What exactly is the data that will be released?
Every time a Medicare patient receives health care, the place where they got that service files a claim with Medicare, and Medicare puts it into a big database. So it’s a set of billing records. It doesn’t have things like lab results; it doesn’t have a lot of clinical data. So the first thing patients should know is that their clinical data is not going to be available. All this is, is documentation of interaction with the health care system. Medicare aggregates it on a very big scale: How many patients in Boston saw Dr. X or went to hospital Y?
Medicare also collects quality data, mostly through chart review. For example, how many patients who had a heart attack got a beta blocker? Most of that information is available at the hospital level and presented on the Hospital Compare Website. So part of the data is actually not a new thing — that’s been available for some time. But only 6% of patients are aware that it exists, and probably even fewer are able to see the information and comprehend it.
So now, Medicare is going to release more data — though not just to anybody. You have to be a “qualifying organization” — probably employers, insurance companies, and in Massachusetts, a group like Massachusetts Health Quality Partners, groups that have a track record of using this kind of data to compile quality records.
So it won’t be available on the Web to you and me?
Right. Not directly. But a group like Massachusetts Health Quality Partners, which already has a track record of compiling data on quality, will have a bigger group of patients to analyze. They can compile a report that can be more representative of all the patients that a doctor or a hospital treats.
There’s a really a great potential here to get patients more engaged on what health quality is. One barrier is that patients may not even be aware that there are differences in quality of care between doctors and hospitals.
Another challenge is that these data are really complicated. Even with some of the agencies putting together reports, I think it’s going to be really hard for patients to understand what they’re looking at. That’s a big challenge with Hospital Compare: If you’re told your looking at the risk-adjusted mortality rate of a hospital, as a patient, what does that mean to you? In addition, you’re looking at so few conditions that they’re unlikely to be relevant to you.
What I hope comes of this is that we can start to look at more conditions and more types of doctors. If you can see all the surgeons in an area, you can see which surgeon or hospital does the most of a given surgery. If, say, Baystate does the most of a given surgery, it’s possible we’re best at it because there’s an association between volume and outcome. That would be a really good starting point — to report on volumes.
So are you saying that probably, the central benefit of what Medicare has just announced is that in the relatively near future, we’ll be able to look at volume and possibly quality for just about any doctor?
Yes, volume and potentially quality — though quality metrics remain a challenge. They may include: risk-adjusted mortality rates, surgical outcomes or complications, risk-adjusted re-admission rates. All of those are potential measures of quality, but we don’t have established and agreed-upon measures of risk-adjustment for every condition out there.
We might know more about volume. One of the best answers we’re going to get is that we’re going to know more about who the high-volume providers are.
How hard is it to find out about volume now?
You can buy Medicare data, but you can’t know it at the individual doctor level. The identification at the doctor level will be the breakthrough with this rule. Prior to the federal health reform act, we were not allowed to individually identify doctors. There’s a clause in the health reform act that alters the law so individual data can be released. This is the response to that.
So then, to speak concretely, say I need a hip replacement and my doctor refers me to a surgeon. Soon I’ll be able to see how many hip replacements he does?
It’s pretty hard right now for patients to get those data. For the most part, you can get information at the hospital level on procedures. You can ask the surgeon how many they do in a year, but it’s not publicly reported anywhere.
These changes can also be expected to provide you some information on the outcomes of the patients the surgeon sees. Reports on quality will tell you how many patients got antibiotics at the time of surgery, and how many had complications from surgery, and how many died. Those numbers are supposedly going to be available.
If individual physicians are unhappy with the score they get, they can push back. But theoretically, you’ll be able to look at all the surgeons in volume, outcome, potentially costs — and choose from those things. It will be a change, if it all happens the way it’s supposed to happen.
I do wonder, though, if it’s really going to add much in terms of the way patients choose a doctor. There’s been a lot of research lately on what does lead patients to choose a doctor, particularly because many don’t know about quality difference, and they aren’t aware of online data, and they may have limited choice because of insurance. For the most part, they go to whoever their primary care doctor recommends. It’ll be interesting to see if this changes the way patients look for doctors. We don’t know. A lot will depend on how the data are presented.
So when might we start seeing the fruits of this new data release?
The data will be released in big chunks this year or early next year. Organizations considered “qualified entities” can purchase the data, and then it takes a while — this is a huge amount of data — to clean it up and compile the reports. I assume it’ll be at least a year from when these organizations get the data before any go up online. That’s a guess based on my own experience working on large data.
I think potentially in a couple of years there will be data, certainly on volume and there will be quality data, certainly on outcomes. It will be interesting to see what push-back there is on risk adjustment on how sick the patients are. And the big question is: Is this going to be presented in a way that patients can understand? It’s got great potential…