REDCapTidieR 0.3.0

REDCapTidieR v0.3.0 Release Announcement

We’re thrilled to announce the release of REDCapTidieR v0.3.0 on CRAN! REDCapTidieR provides a user-friendly way to import data from a REDCap project into the R environment.

You can install the current version from CRAN with:

install.packages("REDCapTidieR")

Major Update: Repeating Events Support! 🎉

REDCapTidieR introduces support for REDCap “repeating events,” enhancing its versatility in handling diverse data collection scenarios. A repeating event is a feature in REDCap that is often utilized in clinical trials for unscheduled study visits. It allows for the collection of the same data zero, one, or multiple times within a longitudinal project alongside regular scheduled study visit events. This feature enables researchers to efficiently gather data from the same set of instruments during various unscheduled time points or instances, while maintaining an organized data structure. With this update, users of longitudinal databases can now take full advantage of this feature, simplifying data import and analysis in R.

For example, consider the physical_exam table which contains data from both regular scheduled and unscheduled visits. These unscheduled visits occur under a single REDCap repeating event. Notice how subject_id 001-123-001 has data for both the “screening_enrollm” and “unscheduled” redcap_events. This demonstrates that you can work with data from both regular scheduled study visits and unscheduled visits in the same data tibble. Also note that the redcap_event_instance values identify the specific unscheduled event that the row is about. This makes it possible to link data in different data tibbles from the same unscheduled event.

library(REDCapTidieR)

read_redcap(redcap_uri, token) |> 
  bind_tibbles()

physical_exam |> 
  filter(redcap_event == "screening__enrollm" | redcap_event == "unscheduled") |> 
  rmarkdown::paged_table()

To support repeating events fully we updated the way REDCapTidieR creates identifier columns for a data tibble (see table below). A breaking change in version 0.3.0 is that we removed the redcap_repeat_instance and replaced it with redcap_form_instance for repeated instruments and redcap_event_instance for repeated events. If you explicitly refer to redcap_repeat_instance anywhere in your code you will update this to redcap_form_instance. Please note that REDCap does not allow having repeating instruments inside a repeating event.

Instrument Structure	Event Structure	Classic	Longitudinal, one arm	Longitudinal, multi-arm
Nonrepeating	Nonrepeating	record_id	record_id + redcap_event	record_id + redcap_event + redcap_arm
Repeating	Nonrepeating	record_id + redcap_form_instance	record_id + redcap_form_instance + redcap_event	record_id + redcap_form_instance + redcap_event + redcap_arm
Nonrepeating	Repeating	N/A	record_id + redcap_event_instance + redcap_event	record_id + redcap_event_instance + redcap_event + redcap_arm
Repeating	Repeating	Not Supported	Not Supported	Not Supported

Performance Improvements and Enhancements

• Better error messages
  • Added informative error messages for various REDCap API call-related conditions
  • Included helpful error messages across all exported functions
  • Provided guidance for submitting bug reports when encountering unexpected errors
• Added select_choices_and_calculations column to the metadata tibble

Bug Fixes

• Resolved an issue where the suppress_redcapr_messages argument of read_redcap() did not work
• Fixed incorrect parsing of REDCap choices for categorical variables without white space after comma
• Implemented a warning when duplicated choice labels are detected for categorical variables
• Addressed a bug causing empty rows in data tibbles for longitudinal REDCap projects
• Fixed an issue where the forms specification in read_redcap() did not work correctly