Jane MetzgerHCI: That speaks to a number of sub-issues, obviously.Metzger: It does indeed, For example, it turns out that medication reconciliation is also going to be pretty important, because that’s how you find out about the meds that aren’t in the EHR, but that the patient is on. The reality is that, even though a lot of people thought, oh, we’ll get to quality reporting eventually, our major conclusion is that you’d better be thinking about quality measures as you develop your Stage 1 plan. Just to take one example, there are a lot of quality measures that use lab results; and for the measures you do pick, you’d better be sure you have all the data.The other conclusion is that there’s a lot of data that is needed for the quality measures that won’t be captured through the inpatient EHR, because there isn’t really a structured designed into the EHR to make capturing that data easily. Fortunately, there is one element in the ambulatory EHR that is useful in this context, and it’s called the health maintenance profile.Vendors and medical groups have relied on the health maintenance profile for years, and physician practices have been making use of it for years. And the health maintenance profile provides, for any given patient, a view of what health maintenance and disease management guidelines apply, and it translates those guidelines into measures.For example, if we’re talking about a diabetic patient, that patient needs a hemoglobin A1c test every six months, or whatever the case may be. A lot of the wellness guidelines you would implement have to do with immunizations. So the health maintenance profile would include an immunization record as one of its items; and in my case, it would be able to note that I received my flu shot, but outside the particular medical practice (because I had mine at a community immunization drive). So there is a way, that’s built into most EHRs, to handle this data gap, but you have to work at it.HCI: What will be the biggest challenges in this for most medical group IT and clinician leaders, in all this?Metzger: I think they’re sort of on their own to figure out the data capture. The technology they use—their EHR will have to be certified that it can calculate and report. But they’re sort of on their own on how to capture the data.HCI: When I read your report, I had the sense that no more than a small percentage of medical groups would be able to achieve this.Metzger: Well, the amount of effort that Jared and I and our team have put into this was quite enormous. And I’m not aware that anyone else has analyzed this quite so thoroughly. And a lot of people thought, oh, their vendor was going to take care of it. And yes, the vendor has to demonstrate that their technology will calculate and report, but they only have to do three measures. And I think a lot of people thought this would be easier than it turns out, and that they’d probably get more help from their vendor on the data capture piece than it seems that they’re getting.As I had mentioned earlier, I think people focused much more on the line items in Stage 1; but they didn’t focus on the measures. And it’s rather overwhelming—I have a notebook that’s probably 4 inches thick, of the measure specifications. And some of them changed from the interim rule to the final rule.HCI: Do you think vendors also realize now that all these things are much more complex than anyone realized?Metzger: As with all of information technology, the reality has always been that success is not about the tool, it’s really about the implementation. And I think there’s a lot about meaningful use that—assuming that a particular product has the basic capabilities—then, implementation builds on that. And I think meaningful use requires a pretty high degree of implementation. Indeed, quality measures are a pretty good example of that. Doing them requires thinking about data capture in a very granular way.I’m actually surprised that people aren’t talking as much about encounter notes, because when, in our analysis, when we said, hey, here’s all the data you won’t have; well, the only place you could possibly find it would be in notes. And the only way you’d ever get it in a way you could use it for quality reporting, would be if the notes were structured, so that you could achieve automated capture. And it’s very hard to do because of the time constraints.If you think about the physician in practice seeing a new patient, say, every 12 minutes or so, and you think about the current technology for achieving a structured note, and the demands of meaningful use, well, the data capture would either have to come out of structured notes or out of the health maintenance profile. Fortunately, the health maintenance profile is very structured, which is why it makes sense as a vehicle for data capture. In any case, it will remain a challenge for eligible providers to figure out how to achieve data capture in this context, which only adds to the complexity of the overall scenario.
