UDS FVP Error Checks Comments/Questions

[hewanadun]

In A5D2 and B9

• For variables that allow 777 (provided at previous) do we need plausibility checks to validate whether a valid value entered at prev_visit?
  • For example, in A5D2 HRTATTAGE variable, in IVP, it is possible for a user to select 0 or 9 for HRTATTACK and leave HRTATTAGE blank. And then in FVP, they can select 1 or 2 for HRTATTACK and enter 777 for HRTATTAGE. With current rules this response will be accepted, even though a valid value for HRTATTAGE is not provided in the IVP packet.

A4a:

• FVP PDF has only 7 treatment sections but Q&V and error checks have variables related to 8 sections (IVP PDF has 8 treatment sections, so the FVP PDF needs to be updated?)

A5D2:

• In Section 4, for head injury variables, want to double check whether the plausibility checks against values entered at [prev_visit] intentionally left out or need to be added?
  • Same question for Section 5, COVID19 variable.
  • Same question for Section 6, GENERALANX and PANICDIS variables
• a5d2-fvp-c-186 THYROID conformity => checks for 0-1 or 9, correct to 0-2 or 9
• typo in a5d2-fvp-m-352 test_name => should be changed to CANCERMETA, has CANCERACTV
• a5d2-fvp-m-304 OTHANXDISX => change to "If OTHANXDIS=1 or 2 then OTHANXDISX must be present" ?
• a5d2-fvp-m-357 OTHANXDISX => change to "If OTHANXDIS is 0 or 9 or blank then OTHANXDISX must be blank" ?
• a5d2-fvp-m-311 PSYCDISX => change to "If PSYCDIS=1 or 2 then PSYCDISX must be present" ?
• a5d2-fvp-m-358 PSYCDISX => change to "If PSYCDIS is 0 or 9 then PSYCDISX must be blank" ?
• NOMENSAGE, HRT, BCPILLS marked as required fields in Q&V and error checks, these should be optional since Section 7 does not applicable to all participants
  • Missingness checks for subquestions under HRT, BCPILLS should include blank for HRT and BCPILLS respectively
• a5d2-fvp-p-1093 HRT => this check doesn't align with allowed values for HRT, looks like a copy/paste mistake
• a5d2-fvp-p-1097 BCPILLS = > this check doesn't align with allowed values for BCPILLS, looks like a copy/paste mistake

B7

• Does I4 need to have similar plausibility checks against last UDSv3 prev_visit?

B8

• Modify below checks to account that neurological exam can be skipped
  • b8-fvp-p-1019 => Add "If NORMNREXAM is not blank, ...."
  • b8-fvp-p-1020 => Add "If PARKSIGN is not blank, ...."
  • I think I4 checks b8-i4vp-p-1022, b8-i4vp-p-1023, and b8-i4vp-p-1024 are also need to be modified?

B9

• For checks with A2 b9-fvp-p-1001 - b9-fvp-p-1003 need to check whether A2 is submitted => Add MODEA2 != 0
• COGAGE has range 9-110 in Q&V and 15-110 in error checks

D1a

• There are two FVP PDFs, please remove the incorrect one
• In PDF, for Q2b, for both answers (0 and 1) it says "END FORM HERE", I think we already discussed this in IVP and that is how it should be, but want to double check.
• We don't have any I4 plausibility checks against last UDSv3 prev_visit, I assumed that is due to these variables are new in UDSv4? (just double checking)

** Rest of the forms look good

[jculhane]

@kgauthreaux For reference, only A5/D2 and B9 (as far as I can tell) have 777s. These are both assigned to me.

Jessica: A1, A1a, A5D2, B1, B8, B9
Kathryn: A2, A3, A4a, B4, B7, D1a, D1b

[kgauthreaux]

@hewanadun @echeng06 for the B7 comment, see #485!

[kgauthreaux]

@echeng06 @hewanadun RE: A4a FVP PDF only having 7 treatment sections but Q&V and error checks have variables related to 8 sections - we removed one row in the FVP as it didn't fit with the new questions, and wanted to limit it to 2 pages rather than having one lonely row on pg. 3. They can still add the info for an 8th treatment when submitting data even if they didn't have the field on the PDF, and the FVP in REDCap has 8 options. RT kind of feels like it's fine that the PDF only has 7, but Aileen said she can add another row if it's a must

[kgauthreaux]

@echeng06 @hewanadun
D1a

• There are two FVP PDFs, please remove the incorrect one ✅
• In PDF, for Q2b, for both answers (0 and 1) it says "END FORM HERE", I think we already discussed this in IVP and that is how it should be, but want to double check. KG: Yes, this is correct
• We don't have any I4 plausibility checks against last UDSv3 prev_visit, I assumed that is due to these variables are new in UDSv4? (just double checking) KG: I uploaded some, but some of the vars are new to UDSv4 so you won't see all of the same ones as the FVP!

Pull request: #486

[hewanadun]

@echeng06 @hewanadun RE: A4a FVP PDF only having 7 treatment sections but Q&V and error checks have variables related to 8 sections - we removed one row in the FVP as it didn't fit with the new questions, and wanted to limit it to 2 pages rather than having one lonely row on pg. 3. They can still add the info for an 8th treatment when submitting data even if they didn't have the field on the PDF, and the FVP in REDCap has 8 options. RT kind of feels like it's fine that the PDF only has 7, but Aileen said she can add another row if it's a must

@kgauthreaux
That is fine, no need to change the PDF then. Just wanted to make sure additional variables are supposed to be there.