Itching for Clarification

Humans and other primates are astonishingly good at carving the world up into useful bite-sized categories. We order the world from an early age: noun phrases, verb phrases; chairs, tables, stools, etc. In general, we are blissfully unaware of this task, our own ability, or how difficult it is to reproduce formally in, for instance, computer code. In dermatology we were once very proud of this ability: it is why we rightly revere the masters of the 19th century, who by one means or another were able to bring order to the universe of rashes they saw in their patients, but we have learned that it is very difficult to make many of these classifications explicit (although those of us charged with educating the next generation often pretend otherwise). If I sit on the table, is it now a chair? What are the boundaries that demarcate a stool from a chair?

What so often seems effortless and instinctive is difficult to explain. ‘Why is this a melanoma?’ asks the novice? ‘Because it looks like one’, says the expert. This last example perhaps brings home how important is our ability to categorize. Based on nothing more than appearance, we are able to order the world and say something about the future. Once we can recognize the features of a melanoma, we can make predictions that we find useful about the patient in front of us. What perhaps is most surprising is that we can make these predictions without any knowledge of biology, cell biology or genetics. Just by invoking morphology and the ability to classify the world, we can make predictions about the future.

Stander and her colleagues of the International Forum for the Study of Itch (IFSI) are attempting to address this problem head on. They want to divide the world of itch into groups, such that membership of a particular group confers useful information. “Useful” can be defined across a number of dimensions; aetiology, diagnosis, research or therapy. How well do they succeed? Well, in theory, at least, that is an empirical question. The test has to be what additional information comes from using the proposed classification. Just as writing computer code is the easy bit, debugging it takes longer and is more difficult. Therefore, like any classification system we have to “suck it and see”: does the world of itch make more sense with or without such classification?

Jonathan L. Rees

Section Editor