Does P21 crosscheck other ADaMs against ADSL?

Pinnacle 21 General General Discussion
1

Hi Suzy,

I discussed this amongst our ADaM team and confirmed the following:

We only cross-check a few specific variables in this case:

  • AGE
  • AGEU
  • SEX
  • RACE
  • SUBJID
  • SITEID
  • ARM
  • ACTARM
  • ETHNIC
  • TRTxxP → TRTP
  • APxxSDT, APxxSTM, APxxSDTM vs. their counterparts in non-ADSL
  • APxxEDT, APxxETM, APxxEDTM vs. their counterparts in non-ADSL
  • PHwSDT, PHwSTM, PHwSDTM vs. their counterparts in non-ADSL
  • PHwEDT, PHwETM, PHwEDTM vs. their counterparts in non-ADSL

At this time, we don’t have a generic check that says “If this variable in ADSL is also found in non-ADSL, check the value is the same”

So, assuming the variable you were asked to re-derive is not one of the ones listed above, you should be okay doing this. But it should be noted in the ADRG and called out that this scenario occurred. Because an agency reviewer may notice that variable ADSL.XXXXX has a value that differs in all other non-ADSL datasets.

Hope this helps!

Kind regards,

Justin

2

After a submission, we were informed that we needed to rederive a variable in ADSL that is used in a limited number of outputs. Since this is a huge deliverable, it was decided to create a 2nd ADSL dataset to use only for the relevant outputs (rather than rerunning/re-QCing everything). Since all the other ADaMs contain all of the ADSL variables, this will cause a mismatch for this variable between the new ADSL and the other datasets.

I’m not sure if this is something that P21 crosschecks (the value of a variable across multiple datasets). Are we OK renaming the new ADSL and submitting it with the other ADaMs as-is?