Return to Discussion Forum

Title: Patient Level Data Management Requirements

On the bottom of this page, you will find the topic for discussion and the name of the contributor.

Please add comments and then click on the "Add comment" button.

%COMMENT{type="belowthreadmode"}%

Jon Shuster - 28 Oct 2011

This is from a webinar. I have heard informally that a mandate is coming, but probably not be 3/12. Nonetheless, we should be thinking of getting ahead of the curve to make this not only a burden, but also a research opportunity. For example, maybe we can assemble a pilot consortium to develop and test a system, and assess cost/benefits, perhaps based on REDCap. Supplemental funding would be sought.

Best wishes,

Jon

Peter Kaufmann - 27 Oct 2011

NHLBI routinely creates a de-identified, limited public use data set for its trials, usually through the Data Coordinating Center. Others may also have done this. These could be used as a model for the process, estimating cost and other issues.

Best,

Peter

ChrisLindsell - 27 Oct 2011

This conversation is very timely and very pertinent to our world. I wonder if this fits directly within the general frameworks of the reproducible research and the data management and quality working groups headed by Drs. Harrell and Pollock, respectively. I would like us to consider the discussion in those contexts. I would also encourage us to consider the entire timeline of a data element from when it is measured/collected to when it is locked in a dataset, and subsequently how it propagates through the statistical analyses. These areas are so important for us to be gaining controlling of, and it is a great plug for people to get involved in some initiatives BERD is starting to roll on.

Talk soon Chris

KnutWittkowski - 27 Oct 2011

The March 3, 2003 NIH Data Sharing Polic is already setting the stage.

Knut

Jon Shuster - 27 Oct 2011

Dear Colleagues:

I wonder if we should be proactive in this regard, and develop a pilot consortium project on putting out de-identified data for members of the pilot for their single institutional trials of members of this pilot consortium. The project would need funding to initiate, and the overarching goal would be to develop something we all could subscribe to.

Best wishes,

Jon

Bruce Barton - 27 Oct 2011

I think that some elements of this are already in place. For one of my studies that recently published the main results, a subcontractor (RTI) asked for my programs and analysis files to verify the analysis. They checked the analysis files against the public access data sets and verified the programs. As a bare minimum, we should all be prepared for this level of scrutiny post-Duke.

Bruce\x85

Jon Shuster - 27 Oct 2011

The recent rule making for adding clinicaltrials.gov to consent forms may be another hint that this is coming. I do know it is being considered after the Duke scandal.

Jon

Marc Walton - 26 Oct 2011

Obviously, I did not hear the presentation, and do not know precisely what was said. But if they said that it is decided that patient level data will be required, you may want to ask them where they got that information from, and if it is still current. If their presentation was from February of this, perhaps they misunderstood some early discussion that they heard. No requirement of that type has been issued by NLM.

Marc

Jon Shuster - 26 Oct 2011

It was not a typo. Huron was the presenter, and although they are consultants to the fed, they do not make things up that have not been vetted by the fed. I believe this is coming, if not for trials opened on or after 3/7/12, it will come soon thereafter. It is the will of congress.

Jon

Marc Walton - 26 Oct 2011

I also want to emphasize what Paul Wakim said in his email. The detailed patient-level primary data is not going to posted on the Clintrials web site. There is no requirement for this in the statute.

I do not have the slide-deck that is being referred to, and so cannot comment on any particular slide; and I also do not know who made those slides.

I hope that it was from someone who understands the law and it's requirements; Perhaps it is either a typo in the slides, or that in presenting the slide there would be clarification of what the bullet point means.

Marc Walton

Jon Shuster - 26 Oct 2011

Dear Paul:

This is coming in March, 2012. Did you see the enclosure from the Government Consultants?

This is what is coming in 3/12. See the last line. This may slide, but the intent is to have open access patient level data in the system.

The rules are expanding as we speak! It will have motr potential impact upon the BERDs than the new IRB rules.

Best wishes,

Jon

From Slide 12: There will be retroactive new requirements.

Compliance required on March7,2012 for all trials initiated on or after March7,2011.

Registration (From Slide 15)

\x96Trials in serious,life-threatening conditions

\x95Ongoing trials:27Dec2007

\x95Newtrials:within21daysofFPFV

\x96All other applicable trials

\x95Ongoing trials:27Sept2008 (This is what you were talking about JJS)

\x95New trials:within 21days of FPFV

FDAAA Basics: Milestones

\xA7Results

\x96Basic results: 27 Sept 2008

\x95Demographic, baseline characteristics

\x95Primary, secondary outcomes

\x95Point of scientific contact

\x95Investigators ability to discuss results

\x96Serious/frequent adverse event: 27 Sept 2009

\x96Expanded results rules:2011? 2012?? (Beware: JJS)

\x95Lay and technical summary of results?

\x95Posting trials of unapproved drugs?

\x95Statistical analysis plan?

\x95Full protocol?

\x95Patient level data? (This is the potential heavyweight!!! JJS)

Paul Watkins - 26 Oct 2011

Dear CTSA BERD members,

Just want to clarify that there is no provision for patient level data in ClinicalTrials.gov, and it is not mentioned in FDAAA. Results reporting involves \x93summary data\x94 only.

Thanks,

Paul

This topic: CTSpedia > WebHome > DiscussionForum > DiscussionBERD002
Topic revision: 09 May 2013, MaryBanach

---