To the Editor.—
A "Peanuts" cartoon has Charlie Brown exhorting his baseball team to greater efforts, saying "All right. We've heard the report from our statistician... Both our hitting and our fielding averages were down this year... So you all know what we have to do next season." In chorus, the team responds "Get a new statistician!!!" The article by Spitznagel and Helzer1 proposing a solution to the base rate problem in the k statistic has much the same flavor. The k statistic indicates that the validity and reliability of diagnostic tests is population-specific and tends to be low in extreme-risk populations (low or high base rate or prevalence). What to do about it? Get a new statistic!It is, of course, a fact that k is population-specific.2 However, the situation is even more problematic than has been suggested, for two different clinical populations having similar base rates
Kraemer HC. Charlie Brown and Statistics: An Exchange. Arch Gen Psychiatry. 1987;44(2):192–193. doi:10.1001/archpsyc.1987.01800140104018
Customize your JAMA Network experience by selecting one or more topics from the list below.
Create a personal account or sign in to: