Data Files

To enable the tool’s effective categorisation and visualisation of protocol deviations, data must be prepared using specific spreadsheets. This section provides guidelines for formatting data to ensure it is usable within the tool, allowing for accurate NLP-powered analysis and insight generation. By following these instructions, users can optimise data entry, minimise processing issues, and unlock the tool’s full potential for detecting patterns and recurring issues. Please refer to the steps below on how to format your data accordingly.

Headers

The spreadsheet headers are summarised below.

Variable Name	Variable Label	Comment
SITEID	Study Site Identifier	Mandatory
STUDYID	Study Identifier	Mandatory
DVSPONDES	PD Descriptor Assigned by Sponsor	Mandatory
DVCAT	Category for Protocol Deviation	Non-Mandatory User can assign manually If not assigned manually then this field supports mandatory fields as NLP will assign a defined DVCAT using the DVSPONDES
DVDECOD	Protocol Deviation Coded Term	Non-Mandatory User can assign manually If not assigned manually then this field supports mandatory fields as NLP will assign a defined DVDECOD using the DVSPONDES
DVTERM	Protocol Deviation Term	Auto-populated