Usage of statistics in the surgical literature and the 'orphan P' phenomenon.

The statistics used in 240 surgical publications were reviewed. Basic parametric statistics were used in 60% of the publications; 21% of publications failed to document a measure of central tendency, 11% of publications contained an undefined '+/-' notation, and 10% of publications did not state the type of evaluative statistic that was used to calculate a P value, that is, an 'orphan P'. These results indicate the need for wider education about the use of descriptive and basic parametric statistics. It is impossible to evaluate the surgical literature critically without these skills.