Hi,
When the length of CM.CMCLAS exceeds $200, the data should be truncated within $200.
This results in an SD1344 (Value for CMDECOD not found in WHODrug dictionary) validation finding.
Could you please adjust the validation rule so that this error does not occur when the length is truncated to $200 or less?
Thank you.
Best regards,
MinJi
Hi MinJi,
Thank you for this valid suggestion. We also understand its potential benefits. However, there are implementation challenges prevented us from the refinement of this rule.
How can the rule algorithm know if submitted value in CMDECOD was truncated? Any ideas?
Thanks,
Sergiy
Hi Sergiy,
Thank you for your opinion.
After investigating this issue, I confirmed that finding is detected when the data does not match the WHODD dictionary entries.
From=“FILE:TAB-HEADERLESS:%System.TerminologyDirectory%/data/WHODD/%System.WHODD.Version%/DD.txt” Search=“V1 @eqic [%Domain%DECOD]”
Additionally, splitting the data by semicolons resolved this error.
However, according to the SDTM IG V3.4 4.5.3.2 Text Strings Greater than 200 Characters in Other Variables section, when splitting a text string into several records, the text should be split between words to improve readability.
Rather than splitting only the coding dictionary data by semicolons to eliminate the validation finding, it would be better to maintain consistency by splitting dictionary values longer than 200 characters into word units of fewer than 200 characters.
It seems there might be a way to either merge the data from SUPPCM.CMDECOD# and compare it or split the WHODD data into words of fewer than 200 characters and compare them that way.
Best regards,
MinJi
Unfortunately, now P21 Engine cannot merge data from different sources.
Also, formally SUPPCM.CMDECOD# is not standard variable and cannot be considered as reliable source of information.