I have been invited here to try to give an historical and practical perspective. Incidentally, I am in the enviable position of not having to pull my punches. I don't have a classification to sell. I don't work for the federal government so I don't have to be polite to either the public or the private sector. Indeed, I don't have a job to protect at all. I intend, therefore, to speak my mind. I sense there are people in the room who think I never pull my punches. I am the gadfly who can be pushed into saying things that others would like to say but who are too diplomatic to do so.
I have lived and worked through three revisions of the International Classification so I have a certain amount of experience when it comes to preparing for and implementing changes. The first thing to remember is that though there will be weeping and wailing and gnashing of teeth and prophecies of doom, sooner or later it gets done and everyone comes out the other side only a little the worse for wear.
The next thing to remember is that a vocabulary is not a classification—repeat, a vocabulary is not a classification. A vocabulary is something you use for input, a classification is something you use to organize output.
At the last meeting one of the presenters displayed a chart which ostensibly showed how inadequate the current classifications were in exhibiting what he said were the characteristics of the ideal classification. It is, of course, elementary to select attributes that your competitors don't have when deciding which is best. I am most expert in ICD and I know that the attributes he thought were missing in ICD were due to his own misapprehension of the classification. ICD does display synonymy, it is concept oriented (at least as concept oriented as medicine or the human body is), the semantics are clear to the trained eye, it does exhibit granularity (that's the thing a hierarchical statistical classification does best), etc., etc. Knowing this makes me a little skeptical about his evaluation of the other classifications. I worked with the Read system in Ireland and I know it's not the Jim Dandy he seems to think.
I'm sure everybody knows by now, if they didn't know before I even opened my mouth where my bias lies. As far as I'm concerned, the strongest argument for continuing with the International Classification (aside from the fact that not using it puts us out of step with the rest of the planet) is its familiarity. Even a new revision of it will look familiar to the current users. We know how to use it, we know how to teach it, we were instrumental in its development. We have spent more than forty years in this country developing educational programs on it, learning how to devise multiple hierarchies using it, creating meaningful groupings with it. Sure, we have some trouble explaining some of the rules and regulations of the classification, but we have a firm base from which to work. If you really want to create an immense administrative burden, try introducing a whole new concept in diagnosis coding (the industry is going to have enough trouble with ICD-10-PCS without further ordeal).
I am not such a hidebound conservative on the classification that I think we should not implement ICD10, however. On the contrary, I am extremely chagrined that we haven't done it already. I consider it a national embarrassment that the richest country on earth whines that it doesn't have the wherewithal to implement a classification already implemented in other countries—it's almost as bad as us ducking our UN dues.
I am not only an old fogy where the classification is concerned—I'm also an old fogy where computers are concerned (I've been working with them since the late fifties), but I know that we should be working toward a totally computerized medical record and that we will have it one day. I just think the day is not yet. I know it looks imminent when you're seeing it from the viewpoint of Mayo Clinic or Columbia Presbyterian Hospital or the University of Michigan Hospital or the National Library of Medicine, but it looks pretty far away when you're looking toward it from Grand Island, Nebraska or Maurice, Louisiana or even Placerville, California. This is especially true if you're looking at it from the perspective of a doctor's office or a school infirmary or a factory clinic. Until we have the facility for generating medical records that develop diagnoses from physical findings and diagnostic tests without needing the intervention of a coder, I think we are going to need a visible classification with pigeon holes that people can understand and not some nebulous vaporware that they can't get their minds around.
And not only do we not have universal computer power to implement the electronic medical record universally, we also have to deal with the fact that physicians are taught to think taxonomically. And physicians are the ones generating the medical record. So the bottom line, in my opinion, is we'll have to wait for meaningless (or context-free) codes until everybody has as much computer access as the National Library of Medicine does and until all the physicians taught to think taxonomically have retired.