AD0138: CRITy is populated and CRITyFL is not populated - this is not a rule

Check seems to be correct - ADaM rule should be changed

I think about this ADaM rule and I think the rule should be changed. If a result is missing the CRITyFL must also be missing. Why CRITy should be filled when the result is MISSING?

I often use the Y/N flag instead of Y/null because when all results do not fulfilled the criteria I need the text in CRITy to create requested tables. But I don't need tables for MISSING values.

What you think about this ADaM rule?

Perhaps it will be nice if you can forward this to the ADaM team.

Best Regards,
Matthias

Hi Matthias, we will be

Hi Matthias, we will be submitting this and all feedback to the ADaM team.   BTW, do you have my original reply to your comment here regarding AD0138 in an email or something that you could send to me ? 

no email

hi mdigian, I don't know what you mean. You didn't send an email to me. I replied on my own question.

kind regards,

Matthias

Reply

No problem.  I realized last week I didn't respond.  I had formed this detailed response and i must have lost my connection or perhaps hit preview only, followed by a close of my browser.   here's what I intended to send

I agree that "Missing values are allowed if the result is missing" as in example 2 ADaM IG 1.0", page 77, Table 4.7.1.2.    However because these were published checks in the ADaM Validation 1.0 document, we implemented them as is.   We did try to supplement the Validations with info from the Standard and IG.   But as you correctly indicate here, we sometimes missed the contradictions between rules and IG. 

CRITy: text defining conditions needed to satisfy presence of the criterion (Systolic Pressure >160)

CRITyFL:  was CRITy met (Y or N). 

CRITyFL must be present on the dataset if variable CRITy is present

CRITyFL  can be set to “Y” or “N” or null.  In Example 2, CRIT1FL is set to null for Subject 1005 because the criterion is unevaluable due to missing input(s).

The ADaM team may come back and say that the AD0138 is necessary to actually alert the user to the possibility of these missing inputs in addition to the CRITyFL variable not being populated.   I'll update this post once I have their feedback

AD0138

Hi Matthias.   In the CDISC ADaM Validation Checks v 1.2, this check is being removed.  I will put a ticket in to the OpenCDISC community to also have it removed, unless you feel there is some merit in keeping a modified version of this rule.  Then we can take that into consideration.  there has even been a comment to populate it as U for unknown/unable to assess

Thanks.

Hi Michael, many thanks for

Hi Michael, many thanks for this. Removing this role is fine for me.

Best regards, Matthias