Gilead-BioStats / gsm

Good Statistical Monitoring R Package
https://gilead-biostats.github.io/gsm/
Apache License 2.0
36 stars 8 forks source link

QC: Fully document dfGroups data mode in wiki/vignette #1649

Closed jwildfire closed 2 months ago

jwildfire commented 2 months ago

QC Details

Specify which Params are expected for each GroupLevel in dfGroups

Additional Comments

jwildfire commented 2 months ago

Study Params with nice labels in Report_StudyInfo

lStudyLabels <- list(
      StudyID = "Unique Study ID",
      protocol_title = "Protocol title",
      nickname = "Protocol nickname",
      site_summary = "Sites (Enrolled / Planned)",
      participant_summary = "Participants (Enrolled / Planned)",
      status = "Study Status",
      product = "Product",
      phase = "Phase",
      therapeutic_area = "Therapeutic Area",
      protocol_indication = "Indication",
      protocol_type = "Protocol type",
      protocol_row_id = "Protocol row ID",
      protocol_product_number = "Protocol product number",
      est_fpfv = "First-patient first visit date (CTMS)",
      est_lpfv = "Last-patient first visit date (CTMS)",
      est_lplv = "Last-patient last visit date (CTMS)"
    )
  }

Summary Derivations:

 lStudy$site_summary <- glue("{round(as.numeric(lStudy$enrolled_sites))}  / {round(as.numeric(lStudy$planned_sites)} )")
  lStudy$participant_summary <- glue("{round(as.numeric(lStudy$enrolled_participants))}  / {round(as.numeric(lStudy$planned_participants)} )")
lauramaxwell commented 2 months ago

@jwildfire as I'm working on this, I see some discrepancy between the dfGroups output from the data-reporting.yaml and what is expected(?) in ReportStudyInfo according to this list. Is this list intended to just be a list of common params? Are any required?

Also, participant_summary and site_summary are not created anywhere in the reporting pipeline as far as i can see. Should I add them to Report_Setup.R if the necessary columns exist in the dfGroups data?

lauramaxwell commented 2 months ago

remove most of lStudyLabels defaults (keep SiteCount, ParticipantCount and Status) and add that mapping to {rbmPipe} in some post {gsm} v2 work. This is too gilda specific at the moment, so moving this out of the open source package makes sense. As such, the params documentation in DataPipeline.Rmd will be limited to these basic params.