DMC Modifications Tasklist (Overview task)

[marshalc]

Following the meeting on 30th Nov 2017, the following (reduced) overview was minuted (see email at 14:48 for full text):

SK: I met with @AllyBradley , @marshalc and LD this morning in order to better understand the WP4 data, issues and come up with a way forward to create robust DMC reports and trial outcome data moving forwards. We spent over 3 hours looking over the data and reports and have come up with a (hopefully foolproof) plan.

The first thing that is evident is that there are different data sources being used. Unless data has been directly entered into the database or received on a paper CRF or AE form, it cannot be considered in the trial.

We currently have database records for XXX randomised donors (YYY kidneys). This will form the starting point for all analyses/reporting. CM will then be able to provide data for all kidneys that have been excluded, and the reasons for these exclusions, to give us our final number of included kidneys to be analysed. He will report the overall number of kidneys, and the number of complete pairs available.

Potential reasons for exclusion after randomisation are:

• Not meeting inclusion criteria (no consent, multiorgan/dual transplant)
• Kidney not allocated to a site
• Kidney discarded after allocation
• Allocated to non-participating site
• Reallocated to a non-participating site
• No evidence of transplant available in the database
• No timings available (knife to skin, anastomosis start, reperfusion, operation concluded) AND no f/u forms received

It is possible that the final group (no evidence of a transplant in the database) includes some kidneys that were transplanted but we have not received data. Once we have this list, we will provide it for the PI to investigate whether these kidneys were indeed transplanted, and if so, why we have no data available. If necessary, CM can provide trial IDs for each of these groups so that the master database can be compared to any other data sources that you may be keeping.

The consort flowchart for the final DMC report will then follow the template attached. LD will produce this once CM has provided the data.

We have agreed the following action points:

1. CM will merge AB's records of received CRFs to the database, so that we can easily identify which paper forms we have received, where data were directly entered into the database, and where follow-up is missing. Moving forward, when paper CRFs are transcribed or data are entered into the database, a field will be set to define the data source so that we can report missing data by visit and by patient.
2. CM will write rules to define excluded kidneys and the reasons for exclusion, so that LD can easily see which kidneys are to be included for analysis.
3. CM will flatten the data into a single spreadsheet for LD to analyse. This spreadsheet will have YYY rows - one for each kidney randomised - and include (where applicable/available)
   • Trial ID
   • Included/excluded and reason for exclusion
   • All donor demographics
   • All procurement detail
   • All recipient demographics
   • All transplant procedure data
   • All follow-up data for each timepoint
   • Total number of AEs/SAEs for that kidney
   • Recipient death and date of death (for included kidneys) - defined by death reported on an SAE form
   • Permanent graft failure (for included kidneys) - defined as report of permanent failure on an SAE form, or graft failure/graft removal on a follow-up form
   • Delayed graft function (for included kidneys) - defined as dialysis on days 1-7 post-transplant or selection of a reason for dialysis (but not if graft reported as removed on initial follow-up form)
4. LD will use this dataset to produce the DMC report. This should follow the format used by WP3, and does not need to repeat protocol information.

AB/SK will review the report and identify any discrepant data. These will be chased and corrected in the master database prior to submission of the report to the DMC.

[marshalc]

1. [DMC Updates] Add paper form received field to FollowUp models #310
2. [DMC Updates] Create Import function for AB paper check spreadsheet #311
3. [DMC Updates] Add Field to FU forms stating the source of the input #312
4. [DMC Updates] Add field to Organ for Included status referencing #313
5. [DMC Updates] Statistician New Data Extract for Organs #314
6. [DMC Updates] Organ Flow Chart #315
7. Update Documentation
8. [DMC Updates] Set inclusion / exclusion for analysis #318

These tasks will all take place on a new Dev 0.9.0 branch, to be merged into Testing, and then Production.

[marshalc]

Progress so far (all in dev, not live):

1. Fields added to Follow Up and Organ for FU Forms and T Form data
2. Spreadsheet data cleaned; import function created + data normalisation; data transfer functions to new fields written
3. Not yet
4. Field added, though not clear how to populate it yet
5. New extract not started.
6. Not started.

[marshalc]

Task 7 (tbc): Refresh the documentation

[marshalc]

Partial release of the 0.9.0 branch being done with task 5 complete - need to test it on the live server though because of performance concerns.

[marshalc]

Partial release has brought the server process to a halt, so website offline at present. Investigating.

[marshalc]

ISE #317 has been resolve, server is active again.

Also, the Simplified Data Extract ran in under 10 seconds on the server! Performance difference between systems is now somewhat stunning and unpredictable.

Progress so far:

1. Fields added to Dev, Staging, and Production
2. Spreadsheet data imported in Dev, and Production. Staging doesn't have a copy of the data to work with so the spreadsheet is redundant there
3. Not done yet
4. Field added, though process for populating it needs defining
5. Extract built, and available on Dev, Staging, and Production
6. Not started yet
7. Documentation not updated yet

[marshalc]

Task 8. Write a function that will automatically set the Inclusion field based on the set of rules outlined in the OP. Add another field to store a note containing the reasons for exclusion so that lookup speed is not affected.

[marshalc]

Task list updated with 7 & 8 (though 7 not linked to an issue yet)