Issues in CO domain

COVAL in "Comments" domain

Dear Able, I checked in the SDTM-IG (as you maybe also did already) but found that the guidance there is inadequate.
For your first question, it is clear that splitting should be between words.
For the two other questions, here is the way it is done in our popular SDTM-ETL software:
- The labels for COVAL1, COVAL2, ... are "Comment1", "Comment2", ...
- If I remember well, we always trim, so that the first character in COVAL, COVAL1, COVAL2, is never a blank (but I need to check). Your assumption that it will be difficult to combine in such case is incorrect, as there will be trailing blanks (SAS Transport 5 is a fixed field-length format) in the COVALn-1 field.

Who is still saying that SAS Transport 5 is a good and "easy to use" format?
Here is an interesting article about this: http://www.phusewiki.org/docs/Denmark%202014%20SDE%20Presentations/Sven%20Greiner%20presentation.pdf

Good luck with the "stone age" format!
 

Issues in CO domain

Thanks for you so much.

But in SDTM IG 3.2,
we can see "When the comment text is longer than 200 characters, the first 200 characters of the comment will be in COVAL, the next 200 in COVAL1".
It seems Ithat we split long comment text with splitting a word.

spliiting between words or not

You are right.
On the other hand, section 4.1.5.3.2 states "When splitting a text string into several records, the text should be split between words to improve readability".
It is however not very clear whether this also applies to COVAL, COVAL1, ...
It doesn't harm at all to split between words also in COVAL, COVAL1, ..., as a simple "trim()" in the software that reads these records takes care of making it right.
If you just split in the middle of a word, this looks also OK, at least if the reviewer is able to see COVAL and COVAL1 at the same time.
 

Hi Able,…

Hi Able,

1.      A recommended option is not splitting words across variables COVAL, COVAL1-COVALn

2.      We recommend using unique Labels (“Comment”, “Comment 1”, etc.) to avoid potential confusion

3.      No. You should not delete a leading space character in COVAL1-COVALn variables

Regards,

Sergiy, Pinnacle 21 team

Thank you for your reply so…

Thank you for your reply so much. especially Sergiy.
For CO domain, I have no any issues now.