Tuesday 16th October, 2012
11:10am to 11:25am
Mark Twain once said “it ain’t what you don’t know that gets you into trouble. It’s what you know for sure that just ain’t so.”
Knowing things ‘for sure’ is hard, especially in healthcare, as the WSJ noted in recent articles (one drawing attention to the irreproducibility of clinical trial results, calling it one of medicine’s dirty secrets, and another critical of an increasingly common type of medical investigation, the observational study).
These are big, big issues. We want medicine to be scientific, because we believe that science works (and we want, with all our hearts, that which works). But as Kathryn Schulz points out, being wrong is hard, especially when it turns into an inflammatory statement (money and reputations are at stake, after all). And yet, as the WSJ points out, we often are.
To address the issues we face in healthcare we must be able to be wrong. We must quell our (false) certainties and, rather than “look for what we’re looking for”, allow ourselves to be surprised. How can we prepare to be surprised?
Big Data can help. By letting go of old paradigms that start by assuming only certain variables are key and end by correlating only these, we can stress-test all possible hypotheses and apply methods that give us more than correlation, which say that “A causes B”, (rather than just “A and B are correlated”). After all, don’t we really want to know, if we possibly can, what causes what?
Sign in to add slides, notes or videos to this session