Domain Label Names & Variations

m Mike

on January 28, 2012

Hi, I've come across many submission packages, and it is quite interesting to see the many variations for domain names (not abbreviations or custom domains). I provided some examples across the board for this, just to exemplify how many variations exist. Based on CDISC input, the SDTMIG should be utilized and, at some point in the future, a correlating CodeList will exist for standard domains. Out of curiosity, why does OpenCDISC utilize the mentioned domain name over the SDTMIG? SDTM Terminology 2011-07-22 CT Code C61536 IE = Inclusion/Exclusion SDTMIG 3.1.2 IE = Inclusion/Exclusion Criterion not met OpenCDSIC (v. 1.2.1) IE = Inclusion/Exclusion Exceptions I’ve also seen… IE = Inclusion/Exclusion Criteria Not Met IE = Inclusion/Exclusion Criterion Exceptions IE = Inclusions/Exclusions Thanks

Forums: SDTM

a Anthony

on January 29, 2012

SDTMIG v3.1.1 has it as

SDTMIG v3.1.1 has it as "Inclusion/Exclusion Exceptions" (section 6.3.2, pg. 70). That's perhaps the reason you are seeing that variation.

At any rate, is there a definitive, referenceable source for dataset labels? For example, in SDTMIG v3.1.2, for LB, it is "Laboratory Test Results" in section 2.5, but "Labs" in section 6.3.3, under the section title. Throughout the IG, it seems to only emphasize on variable label, not dataset label.

In the grand scheme of things, a (slightly) different dataset label will unlikely hinder the review process. Consistency (across studies) is more important.

s Sergiy

on January 30, 2012

Anthony is right. "I/E

Anthony is right. "I/E Exceptions" label is SDTM 3.1.1 legacy issue, which should be fixed for OpenCDISC 3.1.2 specs.

Currently there are no OpenCDISC checks on datasets labels for the reasons Anthony already provided. Do you think we need to introduce them?

m Mike

on February 2, 2012

Yes, correct

Anthony/Sergiy,

After some external discussions, it seems that the dataset label is not of significance, that is, if the synonymy is in align with the domains purpose. Overall, it's most likely that standard domains will have controlled terminology applied to them.

Thanks again!

s Sergiy

on February 2, 2012

Current SDTM IG does not define or use dataset labels

There is no concept of dataset or domain labels in the latest SDTM IG. It has only "descriptions" of domain, which are not consistent across the document to use them as a source for a dataset label validation.