CDR Tickets

Attachment TrialsForCTRP.xls has been added with description: Initial report for CTRP

BZDATETIME::2010-08-23 12:38:26
BZCOMMENTOR::Bob Kline
BZCOMMENT::1

[from separate email message sent by Lakshmi]

We should only include documents that are published to Cancer.gov. The records that are in process we will not have any abstracts to send. Also, I think that will address the transferred docs that still have an Inscope version.

Also, if the trial has a DCP ID, can you suppress the CTEP ID. In the past, before we had DCP ID and DCP had a Protocol Information Office and their own systems, the DCP studies came to us from CTEP and we had the CTEP ID. It does not imply that both DCP and CTEP are conducting the study.

Could we have one sheet for DCP, one sheet for CTEP, and the separate sheet for the transferred trials. Also, it may be helpful for my review to see the current protocol status.

BZDATETIME::2010-08-23 12:39:06
BZCOMMENTOR::Bob Kline
BZCOMMENT::2

(In reply to comment #1)

> We should only include documents that are published to Cancer.gov. The records
> that are in process we will not have any abstracts to send. Also, I think that
> will address the transferred docs that still have an Inscope version.

So we won't bother trying to include the transferred trials for which the latest InScopeProtocol (pre-transfer) version is in a different (blocked) document?

BZDATETIME::2010-08-23 12:41:15
BZCOMMENTOR::Lakshmi Grama
BZCOMMENT::3

(In reply to comment #2)
> (In reply to comment #1)
>
> > We should only include documents that are published to Cancer.gov. The records
> > that are in process we will not have any abstracts to send. Also, I think that
> > will address the transferred docs that still have an Inscope version.
>
> So we won't bother trying to include the transferred trials for which the
> latest InScopeProtocol (pre-transfer) version is in a different (blocked)
> document?

This is getting complicated for me to understand - do you have an example?

BZDATETIME::2010-08-23 12:51:50
BZCOMMENTOR::Bob Kline
BZCOMMENT::4

(In reply to comment #3)

> This is getting complicated for me to understand - do you have an example?

Sure. The second trial on the first sheet in the report is CDR65617. That's the InScopeProtocol document which we transferred last year, re-importing the CTGovProtocol as CDR653664 (the transfer was before we modified the process logic to retain the CDR ID of the InScopeProtocol document when importing the transferred trial back from NLM). Since CDR65617 is no longer on Cancer.gov (it's blocked), it would fall off the report. So the Transferred sheet of the report would only have some of the transferred trials. We could, with some more work, and a slightly more complicated report, track back to match up the CTGovProtocol and InScopeProtocol documents for transferred trials which didn't end up keeping the old CDR ID.

BZDATETIME::2010-08-23 14:16:46
BZCOMMENTOR::Lakshmi Grama
BZCOMMENT::5

Got it. I think for now let us go ahead and keep the blocked ones out. Anyway, we will need to reconcile trials they have with trials we are sending them, so I think we will get from them a list of IDs for which we have not sent them trials. We can then look those up and get them the trial records. I will send a cover email that explains some of this.

BZDATETIME::2010-08-23 14:36:57
BZCOMMENTOR::Bob Kline
BZCOMMENT::6

(In reply to comment #1)

> Also, if the trial has a DCP ID, can you suppress the CTEP ID. In the past,
> before we had DCP ID and DCP had a Protocol Information Office and their own
> systems, the DCP studies came to us from CTEP and we had the CTEP ID. It does
> not imply that both DCP and CTEP are conducting the study.
>
> Could we have one sheet for DCP, one sheet for CTEP, and the separate sheet for
> the transferred trials. Also, it may be helpful for my review to see the
> current protocol status.

For the third sheet, should I give both IDs if I have both a DCP and a CTEP ID? Or should we actually have four sheets, splitting sheet three into one for trials which don't have DCP IDs and one for trials which do have DCP IDs (dropping the CTEP IDs for those)?

BZDATETIME::2010-08-23 15:33:11
BZCOMMENTOR::Bob Kline
BZCOMMENT::7

(In reply to comment #1)

> Also, it may be helpful for my review to see the current protocol status.

Do you want that for the CTGovProtocols, too? Or just the InScopeProtocols?

BZDATETIME::2010-08-23 15:56:21
BZCOMMENTOR::Lakshmi Grama
BZCOMMENT::8

Just the inscopes

BZDATETIME::2010-08-23 16:04:36
BZCOMMENTOR::Bob Kline
BZCOMMENT::9

I went ahead and implemented the code to get the current status for both, so you can just ignore the ones for the CTGovProtocol when you review the report (it's being generated now).

BZDATETIME::2010-08-23 16:09:03
BZCOMMENTOR::Bob Kline
BZCOMMENT::10

I went ahead a did the 4-sheet version.

Attachment Report4896-1.xls has been added with description: New version of report

BZDATETIME::2010-08-23 16:15:23
BZCOMMENTOR::Bob Kline
BZCOMMENT::11

You'll notice that some of the "CTEP Trials" don't have CTEP IDs. That's because I included a trial if it had CTEP or DCP (or both) as a source, and then when I decided which sheet to put the trial on I put it with the DCP sheet if it had a DCP ID; otherwise I put it on the CTEP sheet. I suppose it's possible that there's a trial or two that ended up on a CTEP sheet because it didn't have either a CTEP or a DCP ID, and it had DCP as a source, but not CTEP. I suppose we could run the report again to refine the logic so that if it didn't have either flavor of the OtherID (CTEP or DCP) we decide which sheet to put it on based on source (or we could just adjust the sheets by hand if there are any cases like this).

BZDATETIME::2010-08-23 16:39:58
BZCOMMENTOR::Lakshmi Grama
BZCOMMENT::12

If it does not have either flavor of other ID, I would not include it on the spreadsheet, since there really is no hook to help them match to our trials then.

BZDATETIME::2010-08-23 16:42:46
BZCOMMENTOR::Bob Kline
BZCOMMENT::13

I went ahead and modified the report to handle the selection of which sheet to put a trial on if it has neither a CTEP ID nor a DCP ID. Probably more work than it was worth (in the sense that although it wasn't that much work to do it, the trials without either of those IDs won't be as much use for CTRP).

So: if a trial has a DCP ID, it goes on the DCP sheet. If it has a CTEP ID, it goes on the CTEP sheet. If it has neither ID and it has DCP as one of the sources, it goes on the DCP sheet. Otherwise it goes on the CTEP sheet.

I noticed that in the short time since the previous run some of the trials have migrated over to the Transferred sheets (presumably because CIAT is blocking and unblocking trials).

Attachment Report4896-2.xls has been added with description: Tweaked logic for trials without CTEP or DCP ID

BZDATETIME::2010-08-23 16:43:58
BZCOMMENTOR::Bob Kline
BZCOMMENT::14

Right. I figured that out, too (see my previous comment). Want me to run it again to get rid of those rows?

BZDATETIME::2010-08-23 16:46:59
BZCOMMENTOR::Lakshmi Grama
BZCOMMENT::15

Yes please.

Also, It seems that trials that are not published at all are showing up - this will confuse CTEP if CTRO asks for these trials- eg.,543780 was never published because it was disapproved by CTEP. It's not just the blocked trials but also the trials that were never published that we need to keep out of this list.

BZDATETIME::2010-08-23 17:03:28
BZCOMMENTOR::Bob Kline
BZCOMMENT::16

Attachment Report4896-3.xls has been added with description: Excludes unpublished trials and trials with CTEP or DCP IDs

BZDATETIME::2010-08-23 17:23:17
BZCOMMENTOR::Lakshmi Grama
BZCOMMENT::17

I still see trials that don't have CTEP or DCP IDs - am getting a little confused with the various versions here!

YOu said
Want me to run it again to get rid of those rows?

I said
Yes Please

Just want to make sure I am looking at the correct version.

BZDATETIME::2010-08-23 20:09:35
BZCOMMENTOR::Bob Kline
BZCOMMENT::18

Sorry, for the first pair of sheets I was inadvertently testing for DCP or NCT ID instead of DCP or CTEP ID. Too many IDs floating around!

Attachment Report4896-4.xls has been added with description: Omitted trials with neither CTEP nor DCP ID

BZDATETIME::2010-08-24 12:34:33
BZCOMMENTOR::Lakshmi Grama
BZCOMMENT::19

Looks good. I think we can send this to charles

BZDATETIME::2010-08-24 13:25:00
BZCOMMENTOR::Bob Kline
BZCOMMENT::20

(In reply to comment #19)
> Looks good. I think we can send this to charles

I'm guessing that you're planning to do this, since you mentioned that you were going to provide them with explanatory background information when you sent it to them.

BZDATETIME::2010-08-31 09:50:57
BZCOMMENTOR::Bob Kline
BZCOMMENT::21

[offline follow-up; most of the thread has replies prepended above the previous message, so the thread has to be read (mostly) backwards; I assume the next steps will be for them to tell us which trials they need but can't find in the reports we've given them, then we'll settle on the set of trials they want us to export; I'll create an export structure (an expanded version of what we send NLM) and we'll send them a test batch; sound right?]

Hi Lakshmi,

We knew that matching the trials from PDQ to CTRP will be difficult since none of the IDs were loaded in CTRP when the initial load was done. My apologies for failing to mention that we needed the title in the initial list at our meeting last week. I take responsibility for this error. I understand your concern about Kim’s and her team’s time, and agree with it. My plan is to do the matching myself with the help of another analyst. Once we verify the matching, we will update the CTRP trial record with the proper ID from the PDQ list. Once that’s done and verified, the automated loader will then be able to match the trials from the PDQ xml with the proper CTRP trial record and perform the update.

To answer Bob’s question, either a spreadsheet or an xml document would be great for this initial list.

Thanks,
Charles

___________________________________________________________

Charles Yaghmour, PMP
ScenPro, Inc.
cyaghmour@scenpro.com

From: Grama, Lakshmi (NIH/NCI) [E] lgrama@mail.nih.gov
Sent: Wednesday, August 25, 2010 3:19 PM
To: Yaghmour, Charles
Cc: Beckwith, Margaret (NIH/NCI) [E]; Martucci, Joseph (NIH/NCI) [C]; Kline, Robert (NCI); Palaniappan, Ram (NIH/NCI) [C]
Subject: RE: [Follow up-2] CTRP import of PDQ data and Cancer.gov support

Charles

It would have been helpful if your team had checked to see if this would work.. While we can provide titles, there will definitely need to be manual work done to sort out some potential issues. Title matching is not a great way to do this. Hopefully, this will not take time away from actual processing work that needs to continue. I am particularly concerned that Kim’s group catch up on delays in processing.

Lakshmi

From: Yaghmour, Charles
Sent: Wednesday, August 25, 2010 2:22 PM
To: Grama, Lakshmi (NIH/NCI) [E]
Cc: Beckwith, Margaret (NIH/NCI) [E]; Martucci, Joseph (NIH/NCI) [C]; Kline, Robert (NCI); Palaniappan, Ram (NIH/NCI) [C]
Subject: RE: [Follow up-2] CTRP import of PDQ data and Cancer.gov support

Hi Lakshmi and Bob,

Unfortunately, none of the study IDs provided were loaded in CTRP when the initial load was done. Is there a way you could provide the title for those studies? I am trying to see if I can match on that. Not ideal, I know. We’ll see if we could make it work.

Many thanks,
Charles

___________________________________________________________

Charles Yaghmour, PMP
ScenPro, Inc.
cyaghmour@scenpro.com

From: Grama, Lakshmi (NIH/NCI) [E] lgrama@mail.nih.gov
Sent: Tuesday, August 24, 2010 4:44 PM
To: Yaghmour, Charles
Cc: Beckwith, Margaret (NIH/NCI) [E]; Martucci, Joseph (NIH/NCI) [C]; Kline, Robert (NCI)
Subject: RE: [Follow up] CTRP import of PDQ data and Cancer.gov support

Hi Charles

Attached is the spreadsheet with the CTEP and DCP trials.

The CTEP Trials worksheet includes trials from CTEP that have been published to Cancer.gov and in most cases to ClinicalTrials.gov.

Similarly the DCP Trials worksheet includes trials from DCP that have been published to Cancer.gov and CTGOV.

The developers should use this list to match trials in CTRP from CTEP and DCP.

In addition, there are two sheets that list trials from CTEP and DCP that have been transferred to the Responsible Party in ClinicalTrials.gov and reimported into PDQ as ClinicalTrials.gov trials. For these trials, we may not have all the information that we have in the trials that have not yet been transferred (since we have overwritten what we initially had).

Once your team has done the matching, you may want to run a report of what trials are missing and we can try to find those trials. In some very old trials we may not have CTEP/DCP Ids. But then again, I think those may be a handful of studies.

Let us know if you have questions

Lakshmi

From: Yaghmour, Charles
Sent: Monday, August 23, 2010 12:12 PM
To: Kline, Robert (NCI)
Cc: Grama, Lakshmi (NIH/NCI) [E]; Beckwith, Margaret (NIH/NCI) [E]; Martucci, Joseph (NIH/NCI) [C]
Subject: RE: [Follow up] CTRP import of PDQ data and Cancer.gov support

Great! Thank you, Bob.

Hi Lakshmi and Margret,

Do you have an estimate as to when you think you could review and forward the list?

Thanks,
Charles
___________________________________________________________

Charles Yaghmour, PMP
ScenPro, Inc.
cyaghmour@scenpro.com

From: Bob Kline bkline@rksystems.com
Sent: Monday, August 23, 2010 11:47 AM
To: Charles Yaghmour
Cc: 'Grama, Lakshmi (NIH/NCI) [E]'; 'Beckwith, Margaret (NIH/NCI) [E]'; 'Martucci, Joseph (NIH/NCI) [C]'
Subject: Re: [Follow up] CTRP import of PDQ data and Cancer.gov support

On 08/23/2010 11:20 AM, Charles Yaghmour wrote:

Hi Bob,

Just to follow up on our meeting last week. Have you had a chance to look into sending us a list of all DCP and CTEP trials that we will need to have in CTRP?

Hi, Charles. Yes, I've given Lakshmi and Margaret the results of my first stab at the report, and they're reviewing it to make sure it has what you'll need.

–
Bob Kline
http://www.rksystems.com
mailto:bkline@rksystems.com

Attachment trials-for-ctrp.xml has been added with description: Original protocol titles needed by CTRP for ID mapping

BZDATETIME::2010-08-31 09:52:37
BZCOMMENTOR::Bob Kline
BZCOMMENT::22

[another message in the offline thread]

On 8/25/2010 2:21 PM, Charles Yaghmour wrote:
> Unfortunately, none of the study IDs provided were loaded in CTRP when the initial load was done. Is there a way you could provide the title for those studies?

Lakshmi:

Would the original title be the best thing to provide?

Charles:

How about an XML file instead of a spreadsheet (titles are a little unwieldy on the user end in a spreadsheet)?

BZDATETIME::2010-09-09 11:02:50
BZCOMMENTOR::Bob Kline
BZCOMMENT::23

Attachment trials-for-ctrp-20100908.tar.bz2 has been added with description: First run of export program

BZDATETIME::2010-09-28 18:26:43
BZCOMMENTOR::Bob Kline
BZCOMMENT::24

CTRP identified 261 additional trials that they would like to have added to the export set we give them. I ran a report to identify why they weren't included in the sets we have been giving them up to now. The report is attached. Summary:

19 OK (status changed?)
130 CT.gov protocol (started that way or was transferred already)
26 Don't have source of 'NCI-CTEP' or 'NCI-DCP'
58 Not active any time between 2009-01-01 and now
8 Have neither a CTEP ID or a DCP ID
20 We don't have the trial at all

Do we need to relax some of our criteria?

Attachment Report4896a.out has been added with description: Disposition of additional trials

BZDATETIME::2010-10-04 10:55:47
BZCOMMENTOR::Bob Kline
BZCOMMENT::25

[From Charles Yaghmour, 2010-09-28]

Hi Bob,

The list below summarizes the changes we would like to receive in the next export. This is based on the email conversations we’ve been having over the last few days.

It would be great if you could regenerate the export file for all trials with those updates, provided you don’t see any show stoppers.

Thanks in advance,
Charles

1. Study Status Date – provide the lead organization’s trial status and status date
2. Study Start Date Type – Provide whether it is Actual or Anticipated
3. Responsible party – provide an indicator to distinguish between a person and organization responsible party
4. Eligibility – Provide the <eligibility><criteria><textblock> tag value in marked-up format
5. Participating Sites – provide the role for the person to identify whether s/he is the PI or just a site POC
6. Person information: The goal is to be able to either locate the person in CTRP, or add it. For CTRP to add a person it requires first and last names, full address, and email address. So if you don’t have a ctep_id for any person it would be great if you could provide us with that additional information, if you have it. If you send us the CDR ID for the person we will be able to store it in CTRP, only if we ended up adding that person
7. Lastly, and thanks to Ram, we identified several trial that we have in CTRP, and were identified by Kim as DCP/CTEP trials, however, they were not on the list you sent us. Is it possible to include those as well? Files are listed in the attached file

BZDATETIME::2010-10-04 10:57:16
BZCOMMENTOR::Bob Kline
BZCOMMENT::26

[Offline reply to Charles, sent today]

On 09/28/2010 04:49 PM, Charles Yaghmour wrote:
> Person information: The goal is to be able to either locate the person in CTRP, or add it. For CTRP to add a person it requires first and last names, full address, and email address. So if you don’t have a ctep_id for any person it would be great if you could provide us with that additional information, if you have it. If you send us the CDR ID for the person we will be able to store it in CTRP, only if we ended up adding that person

This item on the punch list is the one that's holding us up. As I tried to explain when this request came up earlier, there are several complications:

1. the persons in our repository can have more than one set of contact (address/phone/email) blocks;
2. much of the contact information for those persons is stored in linked documents (e.g., Organizations), not directly in the Person documents; and
3. the information in the person documents is no longer maintained, so much of what we would give you could be obsolete

This is why I was saying that the best we could do might be to give you the CDR ID for the persons (when we have an ID at all) and provide a way for you to get to the denormalized Person documents, with the understanding that what you would be getting would not be current information.

Lakshmi/Margaret:

Is there a better approach to this problem than what I've been able to come up with?

In another thread you (Charles) identified 261 additional trials that you would like to have added to the export set we give you. I ran a report to identify why they weren't included in the sets we have been giving you up to now. Here's a summary of what I found:

19 OK (status changed?)
130 CT.gov protocol (started that way or was transferred already)
26 Don't have source of 'NCI-CTEP' or 'NCI-DCP'
58 Not active any time between 2009-01-01 and now
8 Have neither a CTEP ID or a DCP ID
20 We don't have the trial at all

For the first category (the 19 which – I assume – would have been included had we generated the spreadsheet from which we're working today), I see no problem including them. For the largest group (the ones for which we have CTGovProtocol documents instead of InScopeProtocol documents for the trials), I don't see what we could do that would be useful for you. Some of them will be trials for which we have older InScopeProtocol documents representing the information before ownership of the trial was transferred, but I wouldn't think you'd want stale information. For the 20 trials we don't have at all, again there's not much we could do that you couldn't do yourself (i.e., retrieve the information from NLM directly, assuming it's available). For the other three categories the question would be: were the original criteria we used to select the trials incorrect? Do we need to relax some of our criteria? Should we go back and generate a new spreadsheet with different constraints?

BZDATETIME::2010-12-21 15:05:20
BZCOMMENTOR::Robin Juthe
BZCOMMENT::27

Added [CTRP] to issue title and added me as a cc.

BZDATETIME::2011-01-25 10:00:14
BZCOMMENTOR::Bob Kline
BZCOMMENT::28

Created another, larger set at Charles's request:

http://pdqupdate.cancer.gov/larger-sample.tar.bz2

[or]

http://pdqupdate.cancer.gov/larger-sample.zip

BZDATETIME::2011-03-21 10:45:48
BZCOMMENTOR::Bob Kline
BZCOMMENT::29

========================================================================
Feedback from Charles:
========================================================================

Hi Bob,

As we were testing our upload procedure for PDQ trials, we encountered some data anomalies that I like to get your opinion on. If there is something that could be done about those on your end it would be great. If not, we will have to find a workaround on our end.

Attached is a spreadsheet with a list of the latest trials you sent us on 25-JAN-2011. Trials with data issues were marked accordingly. Here is a brief summary:

1. Primary completion date and type are missing
2. Primary completion date seems incorrect when compared to the latest study status date, i.e. completion date is marked anticipated, however, it is prior to the current trial status date
3. Missing Responsible party information
4. Responsible party phone number format is inconsistent
5. Missing masking roles for double blind studies
6. Missing participating site information

Please let me know what you think.

Thanks,
Charles

========================================================================
Feedback from myself:
========================================================================
Since these issues are driven by the data, and not by the software exporting the documents, I'm passing the questions on to Margaret and William. This is a forward instead of a reply because William wasn't on the original distribution list so he wouldn't have seen the attachment.

Bob

========================================================================
Draft response from William:
========================================================================
Hi Margaret & Lakshmi,

I have recorded my findings for the data anomalies below. I thought you would like to review them before I send it to Charles.

Thanks,

William

[1. Primary completion date and type are missing]

Studies that do not fall under Food and Drug Administration Amendment Act, 2007 (FDAAA)

-Biomarker/Laboratory analysis studies

Studies which have a Study Category of Biomarker/Laboratory analysis type do not fall under FADAA so we don’t collect FDA Regulated information for those. [Examples 271322 ]. Type is an attribute of the Completion Date element so if a trial is missing the Completion then it will obviously be missing the type element as well.

[2. Primary completion date seems incorrect when compared to the latest study status date, i.e. completion date is marked anticipated, however, it is prior to the current trial status date]

Our protocol schema purposely has no restrictions regarding this. Site and status information for certain trials (cooperative trial, for example) are maintained programmatically through the RSS service. While the completion date information is maintained manually. A trial status update can be done independently of a completion date update and vice versa. If there are restrictions, this will create a lot validation problems.

[3. Missing Responsible party information]

Like #1 above studies that do not fall under FADAA do not require this information so this information will be missing for these trials.

[4. Responsible party phone number format is inconsistent]

i. We collect or store this information for clinicaltrials.gov registration and updates but clinicaltrials.gov does not have any specific format restrictions for phone numbers.

ii. Some of the anomalies are due to data entry errors which can be fixed

[5. Missing masking roles for double blind studies]

It looks like the trials you marked as missing masking roles (examples : CDR0000384087 and CDR0000331829) have this information captured under the Protocol design element. It seems the problem is that that we call this information differently. Let me know if this is not the information you are looking for.

[6. Missing participating site information]

Trials may not have participating site information because

i. The trial is closed or completed to accrual [i.e., completed or closed to accrual].

The existence of participating sites partly depends on status of the trial and partly on the way the trial is updated in the CDR. For Cooperative group trials (RSS service), when the trial is not accepting patients (Approved not yet active, Temp. Closed, Closed, Completed), sites are not provided through the RSS service. If the trial is accepting patients then sites are provided through the service.

For some other trials (non cooperative trials), the sites data is maintained manually so it is possible for a trial to be closed or completed and also have the site information. In such cases, the site status should reflect the protocol status.

ii. Approved not yet active trials

These trials are not yet open for accrual and may not have participating sites for that reason.

BZDATETIME::2011-03-24 11:41:56
BZCOMMENTOR::Margaret Beckwith
BZCOMMENT::30

It seems like CTRP is going to have to review all of these trials and add any of the missing information they need. That is part of the clean up we talked about when we first discussed this with them. My understanding is that we are basically doing them a favor by giving them what information we have for the trials, which is more up to date than what they have. We definitely aren't going to collect any of this data. Lakshmi--am I off base here?

BZDATETIME::2011-03-28 10:27:42
BZCOMMENTOR::Bob Kline
BZCOMMENT::31

During last week's conference call Lakshmi and Margaret let CTRP know that they'll need to fill in existing gaps in the data themselves.

BZDATETIME::2011-04-18 10:22:37
BZCOMMENTOR::Bob Kline
BZCOMMENT::32

Created the full set of converted trial documents for Charles:

http://pdqupdate.cancer.gov/trials-for-ctrp-20110418.tar.bz2

BZDATETIME::2011-07-20 10:04:32
BZCOMMENTOR::Bob Kline
BZCOMMENT::33

New twist: they've decided they need more recent trials after all. New proposed logic is to send them all published InScopeProtocol documents which were created on or after Jan. 1, 2009, and which have values of 'NCI-CTEP' or 'NCI-DCP' in the ProtocolSource/SourceName element, regardless of trial status. Lakshmi will get back to confirm that this is what we should do.

BZDATETIME::2011-07-21 14:29:39
BZCOMMENTOR::Margaret Beckwith
BZCOMMENT::34

Here is the email from Kim outlining the additional requests from CTRP:

Hi all-

CTRP would like to explore the import of additional records from PDQ aside from the ~826 records Bob is currently providing. Please see the three data sets below and share your thoughts on the possibility of providing this information. (I have spoken to Bob and Margaret and CTRP leaders on a high level about this request.)

1 – The data set we are currently receiving is DCP/CTEP trials that met publication and status requirements at a specific period in time last year. All new CTEP/DCP trials that meet those criteria (namely, new submissions), are not included. Can the list be updated/refreshed to include those new trials, or can a second list be generated?

2 - CTRO staff are exploring the best workflow for handling dual entry in CTRP and PDQ. Would it be possible to receive the same XML as we have received for the other submissions for newly processed trials going forward? We are looking at an average for 4-5 trials, and could receive one load per week. We could provide the CDRID if that helps. This would last for the period of time that we are completing dual entry, about 3 months.

3 – CTRP has received many registrations for trials that are already registered in CTGov, many originating from PDQ (and are not CTEP or DCP trials). If we provided a list of CDRIDs, would it be possible for us to receive XML files for these trials so we may import and use the data that is already public in our abstraction rather than abstracting from scratch? This would be a onetime request.

Please let me know if you have any questions about the requests!
Thanks,

Kim

At our CDR meeting we proposed that we will implement these using the current process and will not put in any logic to hold back trials; we will send one large feed and it can be de-duped on the CTRP end. Did I get this right Bob?

BZDATETIME::2011-07-21 15:13:39
BZCOMMENTOR::Bob Kline
BZCOMMENT::35

(In reply to comment #34)

> At our CDR meeting we proposed that we will implement these using the current
> process and will not put in any logic to hold back trials; we will send one
> large feed and it can be de-duped on the CTRP end. Did I get this right Bob?

Pretty much. There will be some modifications to the current program, but nothing as extensive as we'd need in order to track which ones we've already sent.

BZDATETIME::2011-07-21 17:48:49
BZCOMMENTOR::Bob Kline
BZCOMMENT::36

I have implemented the logic to pick up the new CTEP/DCP trials for the first of the three requests in comment #34. At this moment, that picks up and additional 221 trials not included by the existing logic. I expect that number to fluctuate up/down as new trials are added to PDQ and existing trials are converted to CTGovProtocols. I can fold this into the program whenever you tell me to.

For the second and third requests, they'll be providing lists of CDR IDs. I'm going to implement the program to read whatever files I name when the program is invoked, so if they want the trials for the second request to be sent only once, they can send lists that aren't cumulative.

BZDATETIME::2011-07-29 09:38:30
BZCOMMENTOR::Margaret Beckwith
BZCOMMENT::37

Adding Kim to issue.

BZDATETIME::2011-08-02 11:54:00
BZCOMMENTOR::Bob Kline
BZCOMMENT::38

(In reply to comment #36)
> I have implemented the logic to pick up the new CTEP/DCP trials for the first
> of the three requests in comment #34. At this moment, that picks up and
> additional 221 trials not included by the existing logic. I expect that number
> to fluctuate up/down as new trials are added to PDQ and existing trials are
> converted to CTGovProtocols. I can fold this into the program whenever you
> tell me to.

Kim:

If you confirm that you want this modification to the export program, I'll go ahead and incorporate the new logic into it.

BZDATETIME::2011-08-02 12:06:48
BZCOMMENTOR::Kim Eckley
BZCOMMENT::39

(In reply to comment #38)

> Kim:
> If you confirm that you want this modification to the export program, I'll go
> ahead and incorporate the new logic into it.

Yes, Bob, please incorporate the new logic.

Thanks!

BZDATETIME::2011-08-02 17:33:33
BZCOMMENTOR::Bob Kline
BZCOMMENT::40

(In reply to comment #39)
> (In reply to comment #38)
>
> > Kim:
> > If you confirm that you want this modification to the export program, I'll go
> > ahead and incorporate the new logic into it.
>
> Yes, Bob, please incorporate the new logic.
>
> Thanks!

Done. Results are at http://bach.nci.nih.gov/trials-for-ctrp-20110802.zip

The set contains 1054 trials.

BZDATETIME::2011-12-21 11:29:10
BZCOMMENTOR::Bob Kline
BZCOMMENT::41

Here are some of the exports we've provided to them (including a new set this morning):

http://bach.nci.nih.gov/trials-for-ctrp-20100908.tar.bz2
http://bach.nci.nih.gov/trials-for-ctrp-20110627.zip
http://bach.nci.nih.gov/trials-for-ctrp-20110628.zip
http://bach.nci.nih.gov/trials-for-ctrp-20110629.zip
http://bach.nci.nih.gov/trials-for-ctrp-20110714.zip
http://bach.nci.nih.gov/trials-for-ctrp-20110721.zip
http://bach.nci.nih.gov/trials-for-ctrp-20110802.zip
http://bach.nci.nih.gov/trials-for-ctrp-20110901.zip
http://bach.nci.nih.gov/trials-for-ctrp-20111014.zip
http://bach.nci.nih.gov/trials-for-ctrp-20111024.zip
http://bach.nci.nih.gov/trials-for-ctrp-20111031.zip
http://bach.nci.nih.gov/trials-for-ctrp-20111117.zip
http://bach.nci.nih.gov/trials-for-ctrp-20111221.zip

BZDATETIME::2011-12-23 09:38:02
BZCOMMENTOR::Bob Kline
BZCOMMENT::42

Processed another export request:

http://bach.nci.nih.gov/trials-for-ctrp-20111223.zip

BZDATETIME::2012-01-19 16:35:15
BZCOMMENTOR::Bob Kline
BZCOMMENT::43

Processed another export request:

http://bach.nci.nih.gov/trials-for-ctrp-20120119.zip

BZDATETIME::2012-01-30 11:39:30
BZCOMMENTOR::Bob Kline
BZCOMMENT::44

Got another request from CTRP to export a document which had already been converted to CTGovProtocol. Reminded them to keep Margaret on the distribution list for these requests.

BZDATETIME::2012-03-01 09:01:44
BZCOMMENTOR::Bob Kline
BZCOMMENT::45

Processed another export request:

http://bach.nci.nih.gov/trials-for-ctrp-20120224.zip

BZDATETIME::2012-04-27 10:33:55
BZCOMMENTOR::Bob Kline
BZCOMMENT::46

http://bach.nci.nih.gov/trials-for-ctrp-20120427.zip

BZDATETIME::2012-06-11 08:35:21
BZCOMMENTOR::Bob Kline
BZCOMMENT::47

http://bach.nci.nih.gov/trials-for-ctrp-20120611.zip

BZDATETIME::2012-08-17 14:30:45
BZCOMMENTOR::Bob Kline
BZCOMMENT::48

http://bach.nci.nih.gov/trials-for-ctrp-20120817.zip

BZDATETIME::2012-08-17 14:33:16
BZCOMMENTOR::Bob Kline
BZCOMMENT::49

http://bach.nci.nih.gov/trials-for-ctrp-20120713.zip
http://bach.nci.nih.gov/trials-for-ctrp-20120801.zip

BZDATETIME::2012-10-22 11:31:04
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::50

We decided to close this issue at last Thursday's CDR meeting. Issue now closed.