Thursday, January 7, 2010

Electronic Medical Records: People are Now Thinking

tablets as to what an EMR, or EHR, is, others are now really starting to think about it. There is an interesting short piece by Kibbe and Klepper presenting some useful thoughts in terms of principles. Some I agree with others I have some issues, yet as we all know there is no general consensus and the Government mandates will not make it so. Just look at TSA, the Intel Community, and the like. No matter who runs the Government they never seem to get it right.

Yet I thought it would be useful to add some thoughts. For those of you wondering what I bring to the table I am now volunteering my time at the Brigham and Women's Radiology Department to try to sort through issues on the imaging side. At the other extreme I see on a day to day basis how clumsy the current system is. For I have digitized all family records and then print them out and hand carry between physicians. You see they do not use email, partly a HIPPA issue, partly an issue of added costs, partly a cultural issue.

So regarding EMR the following are my major principles:

1. Make it Patient Centric: The data should be in the hands of the patient, virtually as well as physically (electronically that is). To do this there must be a single registry. Well we have many of them as examples today, like Facebook and the like. Hundreds of millions are there already, even old folks.

2. Add Costs and Alternatives, or Feedback and Feedforward Work: Over the past twenty years we have observed in the literature and in the field two things. First if physicians are given costs information regarding alternative medications they generally choose the less expensive, subject to maintaining patient health. Second, if they are provided with alternative procedures, based on some agreed to set of standards, yes comparative clinical effectiveness, albeit "locally" sourced, say through their professional or local academic groups, they tend to reduce procedures. These are feedback systems reinforcing what works and feedforward systems making peer group "suggestions" or providing options. A simple example is do I get a CRP and an ESR, for a suspected thyroiditis and if not then which one?

3. Do Something: One of the problems of defining and then implementing the "right" system is that the result takes too long, it is filled with problems, it costs too much, and it is out of date when and if it is ever completed. Anyone who has ever been in the real world knows that. So the best approach is to just start with what is there now, say the Google or Microsoft systems, and then build upoon that. Yes the Kaiser system is great, expensive, and under a central control. It most likely will work no where else. But by just doing something and then reiterating one can in a Darwinian sense get somewhere. Namely bad ideas will not make it to fruition and stop the good.

4. Keep it Simple: This is the challenge. The success of the Internet was the success of TCP/IP over the IBM SNA world. One should keep the inner parts of the system as simple as possible and let the edges add the complexity, for they have it already.

5. Let it be Organic: This means we do not need a single God like dictator who in their own mind has the solution to everything. No matter how smart we think we are we always miss something. Back to my old MIT days of teaching, to test an idea I made a dicta in class and watched the little sharks tear it apart. The result was a good idea, not necessarily what I pontificated at the beginning. One must be open to criticism. Unfortunately physicians are frequently not prone to that.

Well just some thoughts.