SD0026 in FA domain: non-ordinal results can trip SD0026

Pinnacle 21 Support SDTM
1

Hi Joan,

We considered this options. It could handle your case. Unfortinately it does not add much value in majority of cases. Usually an issue is an opposite one to your example: --STRESN variable is used as coded value of --ORRES. E.g., --ORRES=“Choice A”, “Choice B” and --STRESN=1, 2.

Right now the SD0026 and SD0029 rules are under discussion by the FDA/Phuse CSS Workgroup 1. We are trying to figure out the best possible solution. This is a list of our current Action Items/ ideas:

" - Identify tests with no units and update rule implementation

a. Smart check based on list in define or TS of tests that do not have units. Requires TS domain.

b. Alternative, controlled term in that field that indicates no units so don’t have

to maintain table. This would impact listings

c. Add list of units and list of tests that have no units to define.xml

d. Add additional variable to the dataset that identifies the test as having no unit – NULL flavor. (this would be a record level not a test level soln.)

  • Propose solutions and determine pro/con of each before proceeding with solution."

Thanks,

Sergiy

2

FA non-ordinal results can trip SD0026 false positive

for the condition “Missing value for FAORRESU, when FAORRES is provided”.

In a study we have a test in FA domain for “Criteria for Staging”

with result values of: faorres = “1” fastresc = “Criterion 1 met”

faorres = “2” fastresc = “Criterion 2 met”

faorres= “1 and 2” fastresc = “Criteria 1 and 2 met”.

For all 3 values for this FATESTCD, fastresn is programmatically null, since the result is non-ordinal.
Is it possible then that the rule for units faorresu should be checked only when the numeric FASTRESN is nonnull?

Instead OpenCDISC assumes that the “1”, “2” values in faorres and fastresc should take a unit, even though the numeric is not populated.

Thank you.