1777 Sentry Parkway West
Building 17, Suite 405
Blue Bell, PA 19422
+1 (888) 507-2270
TI domain 2 versions vs IETESTCD in IE domain
Hi Sasdue - You are…
Hi Sasdue -
You are seeing this rule because the IE follows the model: One record per inclusion/exclusion criterion not met per subject. There is no reference to version.
For this message to go away, you will need to implement naming of IETESTCD consistent with the IG. Most of the rules about IETESTCD naming can be found in the Trial Inclusion/Exclusion Criteria (TI) section. It states that "Individual criteria do not have versions. If a criterion changes, it should be treated as a new criterion, with a new value for IETESTCD. If criteria have been numbered and values of IETESTCD are generally of the form INCL00n or EXCL00n, and new versions of a criterion have not been given new numbers, separate values of IETESTCD might be created by appending letters, e.g. INCL003A, INCL003B."
In your case, you may want to implement new naming something like the below:
ETEST IETESTCD version
Abc EXCL2 2
Def EXCL3 2
Ghi EXCL4 2
Jkl EXCL5 2
IETEST IETESTCD version
Abc EXCL2 4
Def EXCL3 4
Ghi EXCL4 4
Xxx EXCL4A 4
Jkl EXCL5 4
I hope this help! Thanks! Amy
TI domain versions
In my opinion, it is always the decision of the sponsor what the structure of the dataset is. For example, for LB, the SDTM-IG states "One record per lab test per time point per visit per subject", but if you don't have time points, you can also use "One record per lab test per visit per subject", and if you have a remote clinical trial without visits ...
The decided structure is then declared in the define.xml, using the "KeySequence" attributes on the "ItemRef" elements within the ItemGroupDef.
Similarly, if in TI, for your trial, you set the structure to "One record per IE criterion per protocol version (by also giving TIVERS a "KeySequence" in the define.xml), the validation software should pick that up and use that.
I will also discuss this case with my colleagues in the new ORCS team.
Jozef Aerts
Define-XML team
Hi Amy,Thanks for the input…
Hi Amy,
Thanks for the input. However, with the naming you've suggested below, the problem is for version 4, EXCL5 <- is actually Exclusion Criteria 6 in the Protocol, so it looks inconsistent with IE domain versus what's in Protocol. In some listings, we've to display what Exclusion criteria subject was excluded on, and if we display EXCL 5 (which in fact is 6 in the protocol), doesn't that look inconsistent? i believe sponsor won't like seeing that..
ETEST IETESTCD version
Abc EXCL2 2
Def EXCL3 2
Ghi EXCL4 2
Jkl EXCL5 2
IETEST IETESTCD version
Abc EXCL2 4
Def EXCL3 4
Ghi EXCL4 4
Xxx EXCL4A 4
Jkl EXCL5 4
Hi Sasdue -I agree is…
Hi Sasdue -
I agree it could look inconsistent but individual criteria do not have versions. If a criterion changes, it should be treated as a new criterion, with a new value for IETESTCD. In your case, only one criteria was added so only one new IETESTCD will be made. It might be worth investigating the listing to display the actual IETEST value instead of the IETESTCD value or add a footnote to explain why the numbering isn't the same as the protocol. I would also point out that IETESTCD doesn't have to match the protocol numbering, its just a convention.
Thank you Amy, What i've…
Thank you Amy,
What i've decided to do is, add a Protocol version before each IETEST so that each IETEST looks unique with different versions, that way i do not have to change the numbering of the IETESTCD and it'll be consistent with the protocol as well. The listings require to display the number of the exclusion criteria, for example: EXCL6, sponsor doesn't like it to display it as EXCL5 if we change the numbering trying to fix the errors.
As shown below, i've added "Protocol V2:" preceding IETEST for version 2 and "Protocol V4:" preceding IETEST for Version 4, now each IETEST has one to one relationship. This should solve the error i was getting with Pinnacle and also makes sponsor happy and everything is consistent with the protocol.
ETEST IETESTCD version
Protocol V2:Abc EXCL2 2
Protocol V2:Def EXCL3 2
Protocol V2:Ghi EXCL4 2
Protocol V2:Jkl EXCL5 2
IETEST IETESTCD version
Protocol V4: Abc EXCL2 4
Protocol V4:Def EXCL3 4
Protocol V4:Ghi EXCL4 4
Protocol V4:Xxx EXCL5 4
Protocol V4:Jkl EXCL6 4
Thank you,
Hi, I've two versions of the protocols in my study where one of new the exclusion criteria was added in the last version of protocol. All other Exclusion criteria were exactly same except for the new addition in the last protocol.
With that now Pinnacle gives me Error saying Inconsistent value for IETESTCD within IETEST. I guess this is due to the fact that the IETEST were similar in both versions but the TESTCD changed, for example EXCL4 became EXCL5 in version 2 due to addition of one extra exclusion criteria in version 2. It looks like pinnacle validation doesn't look for the version of the protocol. ie. (TIVERS variable in TI domain ?) is there a way to fix this ?
Here's an example that may help understand what's going on.
IETEST IETESTCD version
Abc EXCL2 2
Def EXCL3 2
Ghi EXCL4 2
Jkl EXCL5 2
IETEST IETESTCD version
Abc EXCL2 4
Def EXCL3 4
Ghi EXCL4 4
Xxx EXCL5 4 -new criteria added here
Jkl EXCL6 4 this IETEST was EXCL5 in version 2, now it became EXCL6
Pinnacle Error: “ Inconsistent value for IETESTCD within IETEST”
Any information would be appreciated.
Thanks,
Sasdude.