Skip to contents

Test Data Simulator

Overview

The gsm.datasim package provides functionality to produce sample data for use in the testing and development of applications and packages related to Risk-Based Quality Monitoring (RBQM) of Clinical Trials. Given inputs regarding the synthetic Study’s desired ID, Participant Count, Site Count, Snapshot Count, and the Key Risk Indicators to be constructed, the package generates data pertaining to:

  • Participant demographics
  • Adverse effects
  • Trial discontinuation
  • Lab results
  • Visit information
  • Study and site location
  • I/E criteria
  • Study metadata

By Default, the package constructs data that would create all KRIs that are available in the {gsm} package.

Installation *Temporarily broken due to repo permissions issue*

You can install the development version of gsm.datasim from GitHub with:

# install.packages("pak")
pak::pak("Gilead-BioStats/gsm.datasim@dev")

Sample Code 

 # Generate raw data using specified parameters
 data <- raw_data_generator(ParticipantCount = 100, SiteCount = 10, StudyID = "Study01", SnapshotCount = 5)

 # Generate raw data using specified parameters and Mappings
 data <- raw_data_generator(ParticipantCount = 100,
                            SiteCount = 10,
                            StudyID = "Study01",
                            SnapshotCount = 5,
                            workflow_path = "workflow/1_mappings",
                            mappings = c("Raw_STUDY", "Raw_SITE", "Raw_SUBJ", "Raw_ENROLL",)

 # Generate raw data using a template file (currently it runs on a smaller template but not always on a larger one),
 # this is our current goto sim process:
 data <- raw_data_generator(template_path = "~/gsm.datasim/data-raw/small_template.csv")

See inst/examples/run_gsm_workflows.R for a script to run all gsm workflows using data generated from raw_data_generator. Result will be all analysis and reporting data along with html report outputs.