[Skip to Navigation]
December 19, 2017

Guidelines for Statistical Analysis Plans

Author Affiliations
  • 1Department of Biostatistics and Medical Informatics, School of Medicine and Public Health, University of Wisconsin–Madison
JAMA. 2017;318(23):2301-2303. doi:10.1001/jama.2017.18954

The emergence of the randomized clinical trial as the gold standard for the evaluation of new clinical interventions has been met by the emergence of a host of guidelines for the design, conduct, monitoring, analysis,1-3 and reporting4 of randomized clinical trials including guidance from regulatory authorities reviewing pivotal studies to support approval of products, such as drugs and devices.5 Much of this guidance reflects the key scientific principle that comprehensive documentation should be prepared in advance, including a detailed protocol providing scientific rationale, clear statement of hypotheses, definitions of outcome measures, and description of trial design, as well as a statistical analysis plan (SAP) that provides technical detail about conventions and procedures for testing trial hypotheses using collected data.