This week I will be giving a series of talks at the Peking Union Medical College Hospital, and even more important, learning how internists in that institution see their route to bridging patient-centered care and medicine’s evidence base. I hope to write about my visit daily, if not necessarily to post (that depends on internet connections and whether I can reach social media). Photos will come later.
It’s also an excuse to improve my execrable Mandarin. I know it is bad, despite the unfailing support and friendliness with which many Chinese greet my halting attempts at the language.
I love learning languages, though my success has been varied. I am old enough to remember what it was like to learn a language before Google Translate. An estimable tool, to be sure, one that facilitates looking up those fiendish arthropodian characters with 15 strokes.
Yet we all know Google Translate has problems. Sometimes, with all its fantastic power to make an educated, database-driven guess as to what the source text must mean, the translations stink. No human being fuent in that language would produce such a sentence.
Which is why Google Translate still needs supplementation by dictionaries. Of course, today’s dictionaries use much the same technology as GTrans: databases, search strategies. But they are curated, assembled by teams of lexicographers who are able to bridge the native expressions of one idiom with those of another.
I never shy away from analogizing, and you might have guessed where I am going with this.
Evidence-based medicine produces a number of studies. The best guidelines generate recommendations based on large datasets requiring considerable computational power, in order to model the interaction between various variables and the outcome of interest.
Yet we doctors and patients know that sometimes these guidelines, even those produced by the best science, come out with suggestions that bear faint resemblance to the options available to real human beings living complicated lives – the same way that the frictionless surface beloved of physicists is a crude approximation of sublunar life.
By now it is a cliché to say that the art of medicine is a language in itself. That’s not quite what I am saying here. Only patients are fluent in their own languages: idioms of body, society, family, daily life. Together, providers and patients can function as an expert lexicographic team, bridging the ever-improving, but still sometimes outlandish recommendations of Google Medicine, with the diverse speech of real human beings.
Dr. Google, meet Professor Wittgenstein.