A little weekend reading. Yves Smith points us to a piece by Jerome Groopman in the NYRB.
A fascinating and somewhat disturbing article at the New York Review of Books by Jerome Groopman looks at what counts for progress in medical diagnosis and finds it to be more of a mixed bag than most readers would assume. This won’t come as much of a surprise to those who know a bit about the field (one of my colleagues who worked at the National Institutes of Health called it “a medieval art”). But what is a tad disconcerting is that the efforts to make medicine more scientific may not in fact be a plus.
That may sound simply bizarre to readers. Isn’t evidence based medicine a good thing? Well, maybe not.
One of the reasons this piece struck a chord with me is that some of the efforts to make medicine more scientific parallel, in their negative aspects, the push to make economics more scientific. In medicine, this means developing more rules and tools for diagnosis; in economics, the course chosen was to impose more “rigor” which meant make greater use of mathematical exposition (proof-like theoretical papers) and to have “empirical” papers centered around statistical analysis of data sets.
Now while this all may sound well and good, in fact, both are methodological choices that limit investigation. For instance, evidence based medicine seeks to gather symptoms and then use that to determine what the ailment might be. Well, the problem is these protocols have been developed from people with only one thing wrong with them. Many people who show up in doctor’s offices have multiple pathologies. So a lot of effort is being expended to develop an approach that has limited value in the field, and worse, doctors are increasingly expected to conform to it.