Understanding the statistics you read

Understanding the statistics you readStatistical analysis is a vital part of good medical research. But it’s equally vital that patients and caregivers not get misleading impressions from the statistics and numbers they read.

First, realize that statistics apply to populations, not individuals. 32% of humans have blue eyes and 25% have brown, but that doesn’t mean that your two eyes are 32% blue and 25% brown – you’re an individual, not the population.

Second, many studies report “median survival time,” which can scare the pants off you. For my profile, median survival time was 24 weeks. This does not mean “they gave me 24 weeks to live.” The median is the middle person in a study, which tells you nothing about either the best or the worst cases. In my case, some people are still alive 13 years after the study was done.

The classic layman’s article on this is Stephen Jay Gould’s The Median Isn’t the Message. His cancer’s median survival time was 8 months; he lived 20 years.

As it happens, I was further thrown off by the fact that the study I found was done before any of today’s treatments existed. But without even knowing that, I did the right thing: I asked “What can I do to put myself in the group that did better than the median?”

e-Patient Dave

Healthcare blogger "e-Patient Dave" deBronkart works in high-tech marketing in the Boston area. His interest in healthcare ramped up rapidly in January 2007 when a routine shoulder x ray revealed kidney cancer that had spread throughout both lungs.

Note: This Perspectives Blog post is written by a guest blogger of DrGreene.com. The opinions expressed on this post do not necessarily reflect the opinions of Dr. Greene or DrGreene.com, and as such we are not responsible for the accuracy of the information supplied. View the license for this post.

Get Dr. Greene's Wellness Recommendations

Sign up now for a delightful weekly email with insights for the whole family.

Got an idea, tip or a comment?

Your email address will not be published. Required fields are marked *