Mistakes happen–and why they do.
Tuesday, June 29th, 2010I am fascinated by mistakes–and why we make them. What we do with mistakes is subject for another blog post, but I found some fun information from a couple of places on why we make mistakes. (No comments, please, on how often I make mistakes!) The following are from the book Don’t Believe Everything You Think, by Kida, and from a text on clinical informatics:
- We prefer stories to statistics. Hopefully, we physicians are more focused on evidence based medicine, but there is the adage that we are only as good as our last (complicated) case. This can also be described as availability–our estimate of the probability of something being true based on how well we remember similar events.
- We seek to confirm rather than to question our beliefs. In other words, we look for information that confirms we are right.
- We don’t appreciate that chance and coincidence can shape events. I don’t know how much this applies to medicine, as events usually aren’t coincidental.
- We over simplify. Who doesn’t? Everything is so darn complicated! I do wonder if there is so much information and options in medicine that it is too complicated for us to understand, so we instinctively try to simplify to better understand and manage problems. We arrive on a particular diagnosis and this is the anchor. We then adjust this anchor based on further information. However, we typically fail to reevaluate and underestimate the probability of a disease even when we have further information.
- We misperceive the world around us.
- Our memory is inaccurate. Medicine is very much based on what you know, which is based on what you remember. However, given the explosive rate of growth of information, perhaps as former Google CIO Doug Merrill points out, maybe it’s time we turn from memorizing information to becoming proficient on searching and finding relevant information.
- We classify data based on “representativeness”–e.g. in our experience, does this patient represent a patient with a certain illness? This works when a disease is common, but fails when the disease is rare, the patient is atypical, or our previous experience was atypical.
If, at the minimum, we are attentive to these filters we place on our thinking, our error rates should go down. And just for fun, check out the latest JAMA (June 23/30) article entitled “Adherence to Surgical Care Improvement Project Measures and the Association with Post0perative Infections.” This article shows some errors in common thinking–especially over simplification!


