CDR Tickets

BZDATETIME::2011-01-25 19:22:43
BZCOMMENTOR::Volker Englisch
BZCOMMENT::1

I've talked to Victoria and she agreed with Robin that the 'dirty' can't be too dirty - Citations will have to be included into this new report.

BZDATETIME::2011-02-25 16:52:40
BZCOMMENTOR::Volker Englisch
BZCOMMENT::2

I've created a new version of our Denormalization filter which allows some of the link-elements to be incomplete (no cdr:ref attribute or a cdr:ref attribute linking to a non-publishable document).
At the moment I am only dealing with the elements
Diagnosis
SummaryFragmentRef/SummaryRef
Citation/CitationLink

For these elements I am allowing the filter to continue even if the element could normally not be denormalized but I am adding a new element <LinkError/>. This element will be used in the final filter to displaya message next to the incomplete element:

• • • Denormalization Error: [element-name] ***

I created a new filter set which includes this new Q&D denormalization filter and it's called:
QC QD Summary Set

We will need to discuss the following questions:
a) Should we provide this functionality for all link elements or only for a
selected group?
I've picked the three mentioned above because Victoria gave me a sample
document which had just these types included.
b) Should the Error message be displayed and if 'Yes' what would you like
to display?
c) Should this report mainly be used to allow board managers to submit
not-yet-completed summaries to board members or to allow CIAT to
identify link errors?
d) Do we want such a QD report only for BU, RS, both, other?

I've used the document CDR62901 (Gestational Trophoblastic Tumors Treatment) to work on this.

The following filters have been modified on FRANCK:
CDR589175 - Denormalization Filter: Summary - QD
CDR000114 - Module: Citation Denormalization - QD
CDR000079 - Health Professional Summary Report

The following filter sets have been created:
Denormalization QD Summary Set
QC QD Summary Set

BZDATETIME::2011-03-10 15:02:02
BZCOMMENTOR::Volker Englisch
BZCOMMENT::3

I had realized that I didn't give instructions on how to look at this report.

Follow the link:
http://franck.nci.nih.gov/cdrFilter.html

• Enter the document ID

• Remove the DocVersion

• For Filter ID enter
  set:QC QD Summary Set

• Click the 'Submit FilterRequest' button
  This should bring up the modified QC report.

BZDATETIME::2011-03-18 11:24:28
BZCOMMENTOR::Volker Englisch
BZCOMMENT::4

Here is a list of links being denormalized in the summaries:
TranslationOf
BoardInfo /last
LOESummary /lastp
BoardMember /last
Term /lastp
Diagnosis /lastp
Intervention /lastp
Drug /lastp
SummaryRef /lastp
SummaryFragmentRef /lastp
MiscellaneousDocLink /lastp
MiscellaneousDocumentText /lastp
MediaID /last
ProtocolRef
Board /lastp

Citation /lastp
CitationLink /lastp
ProtocolLink /lastp
Person /last
LeadOrganizationID /last
PublishedIn /lastp

BZDATETIME::2011-04-13 16:19:11
BZCOMMENTOR::Volker Englisch
BZCOMMENT::5

I've added the new option to the bottom of the QC Report options page which allows to run the Quick & Dirty report on MAHLER.
While doing so I realized that the new filters were only available on FRANCK but not yet on MAHLER. The necessary filters and filter sets have been copied from FRANCK to MAHLER as well.

BZDATETIME::2011-04-13 17:32:58
BZCOMMENTOR::Robin Juthe
BZCOMMENT::6

I met with Victoria today to discuss link elements and then with Volker to ask some questions about this report. I'm going to recap where things stand now below.

This report will be dual-purpose. It will be used in emergencies to:

1. allow Board managers to run a quick (but not so dirty..) report to send to a Board member when the usual QC report will not run.

2. help CIAT identify validation errors (when the usual QC report will not run).

Based on our discussions today, it seems to make sense to include "checks" for common link elements in summaries rather than include everything under the sun (e.g., link elements used in protocol documents, just in case they happen to be used in a summary someday).

So, I propose that we "check" for denormalization errors in the following elements:

TranslationOf
LOESummary /lastp
Diagnosis /lastp
SummaryRef /lastp
SummaryFragmentRef /lastp
MiscellaneousDocLink /lastp
MediaID /last
ProtocolRef
CitationLink /lastp
GlossaryRef
LOERef

William, given that this report will be used to help locate validation errors, it would be great to get your input on this list of elements (and also perhaps someone from the summaries group?). Thanks!

I think the error message is fine as is, but it would be nice if there was an option to hide the messages for a "cleaner" version of the "dirty" report.

For the time being, we are only adding this report to the HP RLSO report menu. However, it would not be a huge volume of effort to add this option to the HP B/U QC report or to the patient summary QC reports, so, Margaret, we should discuss with other Board managers as well as patient summary users to see whether such a report is desired.

BZDATETIME::2011-04-15 09:14:59
BZCOMMENTOR::Robin Juthe
BZCOMMENT::7

(In reply to comment #6)
> it would not be a huge volume of effort to add this option to the HP
> B/U QC report or to the patient summary QC reports, so, Margaret, we should
> discuss with other Board managers as well as patient summary users to see
> whether such a report is desired.

Per the discussion in yesterday's status meeting, we will add this report to each of the QC report interfaces - HP B/U, HP RLSO, Patient B/U, and Patient RLSO. It is not intended to be used frequently, but it will be there if we need it.

BZDATETIME::2011-04-15 14:55:59
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::8

(In reply to comment #6)
> William, given that this report will be used to help locate validation errors,
> it would be great to get your input on this list of elements (and also perhaps
> someone from the summaries group?). Thanks!
>

The summaries group at CIAT is OK with your selection of the elements to check for denormalization errors.

BZDATETIME::2011-07-21 16:45:02
BZCOMMENTOR::Volker Englisch
BZCOMMENT::9

I am currently handling three link types for HP summaries (Diagnosis, SummaryFragmentRef/SummaryRef, Citation/CitationLink). Since these are the three most commonly failing links we've discussed to put the changes in production now and add to this in stages rather than waiting until every link type has been addressed. We also decided earlier to make this report type available for HP and Patient reports, so I'm going to make the Q&C option available to both, HP and patient summaries for now.

BZDATETIME::2011-07-26 17:39:40
BZCOMMENTOR::Volker Englisch
BZCOMMENT::10

The first installment of the Q&D report has been copied to FRANCK and BACH. Before doing so I ran diff reports for the Summary QC output. There were no differences found.
At this point the three link elements are being resolved

• Diagnosis
  If a Diagnosis element is being entered without linking to a CDR document
  an error message will be displayed in its place.

• SummaryFragmentRef/SummaryRef
  If a SummaryRef element is being entered without linking to a CDR document
  an error message will be displayed in its place.

• Citation/CitationLink
  If a Citation link is entered but not linked this is OK but if a link is
  made to a non-publishable document the report won't fail and the created
  reference will be blank in the reference section.

I've created the following filter sets:
Denormalization QD Summary Set
QC QD Summary Patient Set
QC QD Summary Patient Set (Bold/Underline)
QC QD Summary Set
QC QD Summary Set (Bold/Underline)

I've created/updated the following filters:
CDR000079 - Health Professional Summary Report - R10129
CDR434822 - Module: QC Filter Templates - Summary - R10129
CDR699063 - Denormalization Filter: Summary - QD - R10129
CDR699064 - Module: Citation Denormalization - QD - R10129

I've updated the following program:
QcReports.py - R10130

Please note that some changes made for OCECDR-3372 are already included in this version even though the changes for the filters are not yet ready.

This is ready for review on BACH.

BZDATETIME::2011-07-28 09:41:52
BZCOMMENTOR::Robin Juthe
BZCOMMENT::11

(In reply to comment #10)
> The first installment of the Q&D report has been copied to FRANCK and BACH.

Verified on BACH for HP and patient QC reports. Victoria even had a use for it this week!

BZDATETIME::2011-08-04 18:15:57
BZCOMMENTOR::Robin Juthe
BZCOMMENT::12

Marking this P6, as decided in today's status meeting, since we have a working version (which resolves some of the most common link elements) available on BACH.

BZDATETIME::2011-12-19 09:02:32
BZCOMMENTOR::Robin Juthe
BZCOMMENT::13

We decided to close this issue in the last status meeting since a report has been implemented and is working fine as is. We will reopen this issue if we determine there are other link element errors that should be resolved in the report.