Hi Jozef,
Study Day variable is not useful in Medical History data, because in most cases only a year of diagnosis is known. SD1087/FDAC0128 check is not assigned to MH domain.
Kind Regards,
Sergiy
Thanks Sergiy,
It then looks as there is an error in the by the FDA published Excel worksheet with rules. For the "domains" for rule FDAC0128 it says "Interventions, Events, Findings, SE, SV". It does not say anything about MH being an exception.
The more I am looking at this worksheet (published by the FDA), the more I am doubting they had a serious QC and QA on it. This is a pitty, as this once again means that the rules are open for interpretation and self-decision. But we cannot blame OpenCDISC for that.
Best regards,
Jozef
I had another look at the wording of Rule FDAC128 "Study Day of Start (--STDY) variable should be included into dataset, when Start Study Date/Time (--STDTC) variable is present".
The more I look at it, the more I think that it means that it is sufficient to have the variable in the define.xml and as a column in the dataset, even when not populated.
Can you comment?
Is Rule FDAC0128 (Missing --STDY variable, when --STDTC variable is present) implemented in OpenCDISC Community 2.0?
I was testing using an MH dataset (the one that came with define.xml 2.0 spec) which has a MHSTDC column, but no MHSTDY column.
In some cases MHSDTC is a partial date, sometimes a full date.
However, I do not see any errors about FDAC0128 in my output.
The rule itself is a bit unclear to me. Does it mean that when there is a column MHSTDTC is present, there must a column MHSTDY? Or does it mean that when there is a value for MHSTDTC, there must also be a value for MHSTDY? In the latter case the rule would be wrong, as there can be partial dates for which no MHSTDY can be calculated. In the former case I expect that I would have got an error.
Or did I oversee something