New CDISC Terminology

As you may have noticed CDISC release new terminology files on the 28th March so the first question is, what’s changed?

In my last post I posted a simple terminology comparison program and I use this for my day-to-day work. I have put a new version here. The update includes the new SDTM and QS terminology files (available here for SDTM – http://evs.nci.nih.gov/ftp1/CDISC/SDTM/Archive/ and here for QS – http://evs.nci.nih.gov/ftp1/CDISC/Questionnaire/Archive). The main results are again to be found in the “Summary” sheet. I have fixed a couple of bugs and added some better processing:

  1. Fixed a bug in the presentation of individual code lists to include the baseline version of the code list information (the sheets with the C codes as the name)
  2. Added the ability to recognise code lists that were retired and then have been re-introduced
  3. Added a highlight in the summary sheet for Code List Items were the SDTM submission value has changed over the history of the item

The changes that have caught my eye (this list is not exhaustive)  in the latest release are:

  1. C66727 – Completion/Reason for Non-Completion. Code List Item C48251 submission value has changed from “PROTOCOL VIOLATION” to “PROTOCOL DEVIATION”
  2. C66731 – Sex. Code List Item C45908 submission value has changed from “UN” to “UNDIFFERENTIATED”
  3. A number of code lists have had their role changed (Test Code swapped to Test Name and vice-versa). See C101811 and C101812 for example for Clinical Dementia Rating. This actually happened in the previous release but I did not notice it.
  4. Some code lists disappeared but have been re-introduced in the latest release. As an example see C101817 and C101818, the EQ5D assessment.

Hope the information is useful. Download the spreadsheet for the full information and please let me know if you spot any errors.

One comment on “New CDISC Terminology

  1. Jozef Aerts says:

    I also found these and other ones, but I stopped reporting them, as it doesn’t seem to help anyway. For example I found codelists where some of the coded values are “all uppercase” and others “mixed case”. This seems to be very illogical to me.
    I think I am “persona non grata” anyway since I proposed to discontinue the development of CT for LBTESTCD and UNIT (and use LOINC and UCUM instead).
    We really do need controlled terminology but it need to be governed well, and we need to avoid reinventing the wheel when good coding systems already exist. With the decision to publish new CT faster I am afraid that we will see even more differences between versions.
    On the long term this could even lead to dangerous situations when comparing studies. Or could it be (I sincerely hope not) that “CDISC controlled terminology may seriously harm you health”?

Leave a Reply

Your email address will not be published. Required fields are marked *