Occasional contributor Matt Quinn has been on a tear this week. He notes the following gem from a recent iHealthbeat story about the use of IT in evidence based medicine.
According to Dr. Bob Williams, principal of Cap Gemini Ernst & Young’s health consulting division, "most physicians want to include data collection in their clinical systems, but they don’t want the guidelines that come from the data to interfere with the care they provide."
Some time back I had a blogging conversation with Robert Centor over at DB’s Medical Rants about evidence-based medicine. He said:
Having physicians enter data on their patients is not an intelligent use of their skills and time. In order to understand quality we would need such a large and broad database that data entry would take longer than patient encounters.
I still have a longer reply to Robert stored up, but the gist of my argument is, just because systems for recording data are not perfect doesn’t mean that we should give up the prospect of using (and improving) them. The lack of the "perfect" in recording clinical data has been allowed too many physicians to drive out the "good" of recording what data they can and using it to guide their care. Regarding Bob Williams’ comments Matt asks:
Does this mean that:
– physicians don’t trust that the data that they’re recording (in clinical as opposed to billing/financial information systems) accurately reflects the care that they’re providing?
– they don’t feel that they should learn (i.e. change behavior) based on what works and what doesn’t work for them and their colleagues (i.e. evidence-based guidelines are so valid/established that one should not deviate from them even based on personal/group evidence)? or that
– it’s OK to collect data but it’s not OK to measure their performance against it (and other docs)?
Like Matt I’m a little puzzled by what Bob means but I suspect it’s the inverse of Matt’s second point in that many physicians don’t want to know about the guidelines because, for want of a better term, "they know better". That’s OK in one of the three methods that Michael Millenson told me are how physicians practise medicine . Millenson’s three ways are:
a) Follow the best evidence based guidelines
b) Innovate by doing something different and record that innovation so that it can be compared to the guideline. Then you can see whether it was worse than the guideline, or better than the guideline and should thus be used to change the guideline, or
c) Ignore the guideline because you just "know" that your way is better.
Millenson tends to think that way too many physicians are in category c), and I suspect that Bob Williams agrees with him. As I’ve posted about before, following the EBM guidelines is not easy, but it should be second nature amongst physicians to figure out what they are, ensure that they are widely understood by the clinicians and patients they are working with, and to check their own practice patterns against them. With no information systems to record what was done, the last part is almost impossible. It seems that the physicians Bob Williams was talking about are not so interested in completing that piece of the puzzle. As I commented earlier this week, this is one area where physicians have a real opportunity to show great leadership. Millenson doesn’t think that they have done so far.
One project was the tracking of myocardial infarction patients. A nurse goes through all charts on the telemetry and coronary care units, checking to see whether patients are being started on the medications recommended by the American College of Cardiology. We placed on all the pre-printed admission order sheets spaces for beta blockers, ACE-I’s, statins, and aspirin, along with. Just that little reminder alone got the docs improved compliance in writing the orders, along with the parameters for holding the medications.
So there you have it. A little hint of the guideline via the information system (in this case a paper order sheet) and the guideline begins to get followed. Similar examples are all over the country, and have been for some time. But the question is when does the tipping point occur?