Your Easy Guide to Clinical Studies

A Statistical Analysis Plan (SAP) is an independent document (i.e. separate from the study protocol) written by the study statistician. The aim is to provide a more in depth supplement to the existing statistical section of the study protocol, where the statistical analysis plan is described in less detail.

 

The SAP provides a statistical roadmap for statisticians and researchers, outlining a detailed description of the (statistical) methods used to analyse the study outcome(s)/endpoint(s). It provides assurance that data analysis is conducted in a transparent, rigorous, and pre-defined manner.

 

Deviations from an approved SAP (e.g., modification of the analysis method or the analysis of additional outcomes not defined in the study protocol) must be clearly documented and justified. Some changes require a protocol amendment that must be approved by the Ethics Committee (EC) and, as applicable Swissmedic (e.g. introduction of additional Interim Analysis (AI) or the omission of a planned IA)

As a SP-INV:

• Discuss with a statistician the necessity (or not) of having an SAP for your study. An SAP is usually required for high-risk studies using complex study designs and statistical methods, including the requirement of  IA(s)
• Once the protocol is finalised, ask a study statistician to write a SAP prior to any data analysis
• Date and sign the SAP together with the statistician. Once approved, the SAP becomes an official document that must be complied with during data analysis, and the implementation of IAs

References

ICH Topic E9 – see in particular

• 4.1 Trial monitoring and interim analysis
• 5.1 Prespecification of the analysis