Here's a great, practical piece from Slate -- something many in health care would do well to read and understand better.
The public's general discomfort with numbers affects us more than we care to admit. People who understand numbers can take a data set and present it in ways that will be interpreted very differently and will prompt different decisions, a phenomenon known as "framing."
The point in this article is the difference between relative risk reduction and absolute risk reduction. A relative risk reduction of 25% is great if the number of people affected by an adverse event goes from 80/100 to 60/100 (an absolute risk reduction of 20 percentage points). A relative risk reduction of 50% is not so great if the number of people affected changes from 2/100 to 1/100 (absolute risk reduction of 1 percentage point).
Wednesday, September 27, 2006
Subscribe to:
Post Comments (Atom)

No comments:
Post a Comment