How many mice? Design considerations for murine studies

Murine experiments, like human experiments, require thoughtful design and sample size calculations. The statistical approaches to design are quite similar to those used for clinical trials, but the differences to be detected are larger, and the sample sizes consequently smaller, in the mouse experiments. As in human studies, it is important to articulate the end points of interest and how they will be measured, and to identify the additional sources of variability within the experiment, to assure the appropriate statistical approach for both design and analysis.

Correspondence: Donna Neuberg, Biostatistics and Computational Biology, Dana-Farber Cancer Institute, LC-1052, 450 Brookline Ave, Boston, MA 02215; e-mail: neuberg@jimmy.harvard.edu.