[Skip to Navigation]
Oct 2011

Visualizing Bayesian Analysis Using a Spreadsheet Geometry

Author Affiliations

Author Affiliation: Department of Dermatology and Clinical Informatics, Marshfield Clinic, Marshfield, Wisconsin.

Arch Dermatol. 2011;147(10):1225-1227. doi:10.1001/archdermatol.2011.295

Bayesian analysis is a useful way to look at evidence. A Bayesian analysis was recently published by Rhodes1 on the prognostic value of sentinel lymph node biopsy (SLNB) for melanoma based on the depth of the tumor. His analysis relies on the crucial summary statistics of sensitivity, specificity, positive predictive value, and negative predictive value. These statistics are derived from a population of patients that is first subdivided by a binary end point (eg, dead or alive) and then further subdivided by a binary predictive test (eg, positive or negative SLNB results). Based on these subdivisions, each patient in the population can be counted in one of the cells of a 2   ×  2 table.

Add or change institution