Standardization in health informatics is one of the courses in Term 2 of the MSc in Health Informatics at KI. In this course you learn a new set of abbreviations for standards (such as SNOMED, LOINC, HL7, OpenEHR, ICD10 etc) that helps health information systems developed by various groups “talk” to each other. The need for a common language is borne out of the all too human phenomenon called the unknowable other – where systems made by different people use different definitions and where one writes tomato and the other says it with a “may” before the “toe”, and where both mean “the part of the alimentary canal which connects the throat to the stomach” but I write esophagus and you write oesophagus. Confused? Me too. This course highlighted our constant struggle to understand each other and how well-meaning ways to do this may turn out pretty cumbersome in the end.
The course reminded me of an article I read for an eHealth policy paper I was writing a couple of years ago. Written by Trisha Greenhalgh, Jill Russell, Richard E Ashcroft, and Wayne Parsons, the paper “Why National eHealth Programs Need Dead Philosophers: Wittgensteinian Reflections on Policymakers’ Reluctance to Learn from History (available here: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3250633/) is about how the ideas from Ludwig Wittgenstein could explain why England’s $20.6 billion National Programme for Information Technology (NPfIT) failed and how to avoid mistakes similar to the ones committed by policy-makers can be avoided. While this paper tackled how policy makers, software developers and users generally ignored each other and went their merry way to the tune of billions of dollars in losses and limited benefits to patients in the end, the analysis framework could also be use to understand the need for standards in health informatics.
My recollection of Wittgenstein and his language games is quite vague (it has been more than a decade since I last took a course in philosophy) but I venture to say that the need for standards could be to let everyone in on the rules of one language game (how we define medical terms, how we define concepts, how we call one procedure in one system and map the same in another) so that the fancy applications, EMRs, EHRs, EPRs we created could understand each other. It was an interesting course and could prove useful down the line but I couldn’t help but ask myself if there wasn’t a better way to do this.