Nomenclature vs. Classification – Part 3

Previously on Imelda’s World: You invented SNOFOO to standardize the nomenclature of footwear, so your servants can talk about your (by now 15,000) pairs of shoes unambiguously via their EFRs. You invented the ICF-9 classification to assign each pair to one of 256 mutually exclusive and exhaustive cubbyholes. When that became too restrictive for your growing collection, you got 2048 new cubbyholes and invented a new and better classification, ICF-10. Because people were using the cubbyhole number as a shorthand for a partial description of the types of shoes found in a cubbyhole, you invented the CHEMs to help them go back and forth between the two classifications.

Now you are wondering whether the CHEMs were such a good idea. People (other than cubbyholers) seem to think that ICF-10 is just a simple expansion of ICF-9. Now that you have eight times as many cubbyholes, they think you just took each of the original 256 and neatly subdivided the shoes in each cubbyhole, distributing them in the new, roomier structure. How many times, you wonder, do you have to explain to them that, while this is very often true, you also took into account changes in shoe fashion, not to mention changes in your own opinions about the best way to organize? Some categories disappear, some are severely reduced, new categories are introduced, and some whole chapters in the ICF-9 book are completely reorganized (for example, maternity shoes).

But many people refuse to get it. So instead of giving up ICF-9 and learning to think in ICF-10, they try to use ICF-9 to get them halfway and the CHEMs the rest of the way. Some see cosmic significance in the interrelationships and, like astrologers, draw elaborate CHEMs diagrams linking the cubbyholes in ICF-9 to their candidate cubbyholes in ICF-10. The CHEMs were supposed to help ease the transition, not make it more complicated. In frustration, you have a huge banner printed and hung over the shoe wing of the palace: “LOOK IT UP IN THE BOOK.”

You particularly regret making the 9-to-10 CHEM. What you should have done, you realize belatedly, was just make a table of ICF-9 codes not used in the 10-to-9 CHEM. This happens when the number of cubbyholes for classifying a particular type of shoe was actually reduced instead of expanded, so there is no way to correlate a less specific ICF-10 shoe cubbyhole to any single more specific ICF-9 cubbyhole.

Why does this matter? Because people almost invariably use the 9-to-10 CHEM when they should be using the 10-to-9 CHEM—particularly the people you were trying to help the most by building the CHEMs, people with ICF-9–coded data, like sales records. They think, “I have ICF-9 data, and I’m going to need ICF-10 instead after the transition, so I have to use the 9-to-10 CHEM.”

No, you keep telling them, they will have an ICF-10 cubbyhole after the transition, and they will need to compare with the ICF-9 data they already have, so they need the 10-to-9 CHEM. They collected the data in 256 categories, so every comparison they make will have to be in those 256 categories until the old data isn’t interesting any more. They can’t make 2048 distinctions out of 256 without guessing. The result would be fiction, not analysis. Nevertheless, they keeping asking you to pick one ICF-10 code for them for each of the 256 ICF-9 codes, and they start listening to astrologers when you say no.

The final straw comes when the EFR people approach you. Since a cubbyhole number is “just another way of talking about a pair of shoes,” all the ICF codes, both 9 and 10, have been added to their SNOFOO-based systems, each with its particular FID. Okay so far, but then some EFR designers started using the ICF-9 FIDs as the basis for pick lists (i.e. choosing shoes by their cubbyhole number instead of their more fundamental attributes) and now they want those lists automatically updated to ICF-10. They are disappointed when the CHEMs (which you had designed for a different purpose) do not magically accomplish this. So they come to you asking for even more maps.

You’ve had enough. You give away all the shoes, leave Ferdie, and start a new life and a new career in health information management, where, you hope, things will be saner and more rational.

Ron Mills is a Software Architect for the Clinical & Economic Research department of 3M Health Information Systems.

5 responses to “Nomenclature vs. Classification – Part 3

  1. Reblogged this on سليمان العمران and commented:
    Part 3 of :
    Nomenclature vs. Classification

  2. “..but then some EFR designers started using the ICF-9 FIDs as the basis for pick lists..”

  3. Pingback: Nomenclature vs. Classification – Part 2 | 3M Health Information Systems

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s