Guidance Document Process

From QIBA Wiki
Revision as of 18:01, 20 August 2020 by Kevino (talk | contribs)
Jump to navigation Jump to search

DRAFT - wording and content both open to change - if reading for the first time, skip over the italics

The QIBA Community may periodically find it necessary to publish a document that is not a Profile but does provide expert guidance from QIBA on how to perform imaging that is conducive to quantitative image analysis. <Should we highlight this is secondary to QIBA's primary mission and thus should be relatively rare?>

<Q. When/why QIBA would benefit from doing one? We aren't trying to encourage anyone who has a paper and wants to up its profile by getting a QIBA Guidance sticker...>
<Q. What are we trying to achieve - Should QIBA get into general imaging guidance or should that be left to others?> Technical Performance, imaging related to an existing biomarker, "good" data, don't step on existing standards, appropriateness criteria, people publishing papers alleging to report quantitative information from imaging but may not have been managing their acquisitions/image "quality" adequately, comparability of results needs comparable imaging, building databases of imaging to then build AI algorithms, (but no QIB for COVID), image quality issues you should pay attention to, this kind of thing might not happen a lot - new disease calls for guidance but existing disease might be more stable/covered, other entities cover "how to do imaging" but we can help with quantitative(compatible) data, even just formatting/data handling, people often don't publish around our profiles and Guidance can serve as a flag and pointer to our Profiles.

A QIBA Guidance document does not include a performance claim or formal conformance requirements and is not constrained to the common structure of a QIBA Profile. In fact, it is probably best if Guidance uses different formatting to avoid confusion with QIBA Profiles.

A QIBA Guidance document does provide recommendations/best practices related to imaging procedures and associated patient handling, device qualification, etc.

  • <There is a risk that readers won't understand/recognize the difference between a QIBA Guidance and a QIBA Profile. Don't want to detract from Profiles. Should we draft a boilerplate disclaimer/differentiation statement sentence/paragraph explaining the difference? E.g. no groundwork per se, no consensus process, no technical confirmation process, etc.>
  • Note some Guidance documents could serve as an "entry point" to a Profile that is a more rigorous coverage

The intended audience of a QIBA Guidance document includes:

  • Clinical radiologists and their supporting staff (physicists, techs) - looking to good quality for a new disease
  • Treating clinicians - ordering appropriateness, patient handling, quality expectations, improve, achieve diagnostic need
  • Vendors - want to provide reference protocol but harmonize with others, general consistency
  • Imaging Centers - educational materials, eg to support clinical trial population
  • Clinical Trials/CROs who want "good imaging" even if there is no quantitative claim
  • (Guidance may have broader target, both broader audience and broader target like any F-18 PET) -
  • Hosp Admin/Dept Heads - how to make good use of resources - adopt imaging for COVID - novel disease -

A QIBA Guidance document is developed by a Task Force formed by the Steering/Executive Committee.

  • The original idea/proposal may come from a Modality Coordinating Committee or within the Steering/Executive Committee.
  • Task Force membership is recommended to include a breadth of expertise appropriate to the subject and scope of the guidance (i.e. similar to a Biomarker Committee)
  • consider input from radiologists, clinical specialists, physicists, equipment/SW vendors, technologists, statisticians, ...
  • to represent consensus guidance, consider multiple representatives from key stakeholder groups
  • When deciding on whether to pursue a Guidance document, the Steering/Executive Committee may consider:
  • the resources the Task Force would require and whether those are new resources or borrowing existing resources
  • the priority of the Guidance document relative to the other Profiles in progress
  • the potential "marketing" benefit to QIBA of having such Guidance published

A QIBA Guidance document is not required to have a Public Comment period, although it may be beneficial and may choose to do so, or may solicit targeted review by specific affiliated organizations with an interest, or that fill a gap in the expertise of the Task Force. Similarly, it is not expected to track and publish comment resolutions.

A QIBA Guidance document is approved before publication by the Steering/Executive Committee.

  • For modality-specific Guidance, the Steering/Executive Committee may request the a review/recommendation from the corresponding Modality Coordinating Committee.
  • For non-modality-specific Guidance (e.g., trans-modality papers, metrology papers, etc.) the Steering/Executive Committee may convene an ad hoc review group of members from relevant QIBA Committees.
  • In some circumstances, QIBA Guidance may need to be issued quickly, in which case the Steering/Executive Committee may choose to shorten typical ballot/review periods.
  • Note: if "rejected", the authors may still choose to remove the "QIBA Guidance" branding and publish the content as a paper without presenting it as the QIBA position.

A QIBA Guidance document may be published on the QIBA/RSNA websites and/or may be submitted to a journal.

Typically the Task Force is dissolved when the document has been approved and published and no near term maintenance/updates are expected.

Note: This process is not intended to apply to papers written by any group of expert authors that refer to QIBA or its activities in the text, but are not presented as guidance from QIBA.

See Also

  • AGREE is intended to help:
  • Assess the quality of guidelines;
  • Provide a methodological strategy for the development of guidelines; and
  • Inform what information and how information ought to be reported in guidelines.