Personal tools
You are here: Home → IMSG → RAG → Minutes → RAG 5-4-2000
 

RAG 5-4-2000

RAG 5-4-2000

Minutes of RevMan Advisory Group (RAG) meeting held at the UK Cochrane Centre on Wednesday 5 April

Minutes approved on 25 May 2000.

Present: Clive Adams, Chris Cates, Mike Clarke (Chair), Monica Fischer (minutes), Julian Higgins, Sonja Henderson,  Rasmus Moustgaard, Mark Starr, Lindsay Stead, Paul Whatling

Apologies: Phil Alderson, Peter Langhorne, Heather Maxwell, Helen Smith, Phil Wiffen, David Wilkinson, Sue Wilson

In attendance: Philippa Middelton

Contents:
 Welcome
 Approval of agenda
 Aproval of minutes and matters arising not on this agenda from RAG meeting on 2 December 1999 
 Approval of minutes and matters arising not on this agenda from RAG meeting on 3 February 2000 
 Content of a Cochrane review on The Cochrane Library 
 Display of RevMan/MetaView graphs in The Cochrane Library 
 Display of additional graphs in The Cochrane Library 
 Running RevMan 4.0 under Windows 2000 
 Future plans for the Information Management System 
 Survey to investigate the needs for the next generation of the Information Management System 
 Progress on RevMan 4.1 beta testing 
 Web-registering for RevMan 4.1 on CD-ROM 
 Automatic inclusion of Conflict of interest 
 Transfer of contact details between RevMan, ModMan and HIREx 
 Wishlist for RevMan
 SPAM from RAG e-mail 
 Membership of RAG and creation of additional RAGs 
 Any other business 
 Date of next meeting 

  1. Welcome

    2. Everyone was welcomed to the meeting. Apologies were received from the people listed above.

  2. Approval of agenda

    3. The agenda was approved.

  3. Approval of minutes and matters arising not on this agenda from RAG meeting on 2 December 1999

    4. The minutes were approved for publication on the Internet.
    Action: Jacob Riis

    Item 3: Publication of minutes
    The web site where minutes of the meetings of the Software Development, RevMan Advisory and ModMan Advisory Groups are to be published, will soon be available for public access. Notification will be sent to the convenors of these Groups when this happens.
    Action: Monica, Jacob Riis

    Item 4.20: Training material for RevMan 4.0 and MetaView 4.0
    Preparation of this is pending.
    Action: Rasmus

    Item 7: Testing of software
    Phil Wiffen had provided a brief progress report by e-mail: extensive enquiries had been made to identify procedures for testing software but nothing of interest had been identified yet. There seems to be lots of opinions but no scientific approach. Phil will keep searching.
    Action: Phil Wiffen

    Item 9.C8: Truncated or wrapped column headings in Additional tables
    This still needs checking both for the Internet version and the CD-ROM version of The Cochrane Library.
    Action: Mark

    Item 11.a: Guidance on good outcomes
    Preparation of this is pending.
    Action: Chris

    Item 12: Licence agreement for RevMan 4.1
    Chris will send an example to Rasmus.
    Action: Chris, Rasmus

  4. Approval of minutes and matters arising not on this agenda from RAG meeting (3 February 2000)

    5. The minutes were approved for publication on the Internet. Mike will pass the latest version to Jacob Riis.
    Action: Mike, Jacob Riis

  5. Content of a Cochrane review on The Cochrane Library

    6. It has been brought to Mike's attention by Update Software, publishers of The Cochrane Library, that not all fields in reviews prepared in the new review format (introduced with the release of RevMan 4.0 in 1999) are currently being published. Mark explained the steps by which Update Software process the reviews when they have been recieved as part of the CRG modules submitted for publication in The Cochrane Library. The reviews are parsed into a database structure, and from this the various formats required by the different publishers are generated. During this process Update Software strips out the fields which are not for publication. A detailed description of which fields in a Cochrane Review should be published is therefore needed urgently both to guide Update Software, but also reviewers and editorial teams. Mike went through the list of fields contained in RevMan 4 and whether or not each of these should be published in The Cochrane Library was agreed. It was also agreed that a review should not be published if certain specified ("mandatory") fields were empty. The input received from members of the RevMan, RAG, Adminors and Allcoeds e-mails lists were taken into account during this discussion. It was agreed that it would not be possible to check automatically in RevMan that all mandatory fields have been completed. Some could be automatically checked by ModMan, others would need to be checked by the editorial team. It was suggested that the dates in the What's new section should be published as a property sheet for a review and could appear as a pop-up screen in The Cochrane Library rather than as part of the cover sheet screen.
    The recommended list of fields will be passed to the Software Development Group for discussion at their meeting and a decision on the next steps.
    Action: Mike

    No recommendation was made with respect to the publication of funnel plots. This should be considered once there is agreement on which graphical images, besides those produced by MetaView, should be considered for inclusion and publication in Cochrane Reviews, see Display of additional graphs in The Cochrane Library.

    For clarification Mike explained that the RAG provides advice on the specification of a Cochrane Review. This is then passed to the Software Development Group (SDG) to act on as appropriate. The editors of the Cochrane Reviewers' Handbook in turn follow guidance provided by the SDG.

    Philippa informed the meeting that the newly established Quality Group, convened by herself and Pim Assenfeldt, will look at what constitutes a model protocol. It was agreed that this Group and the Cochrane Library User Group should also be asked to advice on the order in which the various sections of a Cochrane review should be published. Advice is also needed on what, if anything, should be published if a field in a review is left empty by the review group.
    Action: Mike, Philippa

    The following wish list items emerged during the discussion of this item:

    • It should be possible to have a corporate author for a review, e.g. a Collaborative Group of trialists
    • The dates under the What's new section in RevMan should be redesigned to allow for the history of the review to be captured, rather than the review's development from one publication to the next. In the meantime, the RevMan User Guide should be updated to encourage reviewers to describe the review's history in the Notes field. The guideline should make the difference between the use of the Notes field and the What's new section more clear.
    • An ability to switch numbers for outcomes on and off on the graphs

    Action: Jacob Riis

  6. Display of RevMan/MetaView graphs in The Cochrane Library

    7. Unlike RevMan 3.x/MetaView 3.x, RevMan 4 allows the reviewer to specify which graphs the reader should see as the default when calling up MetaView 4. However, it will not be possible to display graphs in MetaView 4 format in The Cochrane Library that is distributed on CD-ROM for Windows 3.x (16 bit systems). This is because MetaView 4 has been developed for Win9x/NT (32 bit systems). Four solutions had been suggested at the previous meeting of the RAG (see minutes of 3 February 2000):

    1. Include graphic images on the CD-ROM as for the Internet version.
    2. Modify MetaView 3.1 to allow reviewers to specify the statistical settings used for each outcome.
    3. Display a warning message in The Cochrane Library if Windows 3.1 users run MetaView 3.1 for a review where some outcomes are set to anything other than the Peto Odds Ratio or WMD (which are the defaults in MetaView 3.1).
    4. Place warning messages in all reviews for which some outcomes are set to anything other than Peto Odds Ratio or WMD.


    Since options 1 and 2 are not possible, option 3 was favoured. Mike, in his role as the liaison between the Cochrane Collaboration and Update Software, will write an official request to Update Software about this. In addition it was agreed that stronger advice should be included in the Cochrane Reviewers' Handbook asking reviewers to specify the methods used to calculate the results reported in the text of the review.
    Action: Mike

    Mark reported that for the Internet version, MetaView 4.1 will be offered as an option. Otherwise, images will be included as specified by the reviewer. As soon as MetaView 4.1 is ready, it will be put on the CD-ROM for 32-bit systems and offered to the Internet users.

  7. Display of additional graphs in The Cochrane Library

    8. Mike tabled a "work-in-progress" paper prepared by Jon Deeks and himself. The paper should be regarded as a starting point and not as a final recommendation. Feedback should be sent to Mike and Jon by the end of April. A revision will be circulated by e-mail to the Group during May. A final paper is expected for wider circulation and consultation in June. It was suggested that the paper should also include specification of figures and graphs to be published with a Cochrane review (existing as well as additional) but this will require input from the Statistics Methods Group.
    Action: Mike, Jon Deeks and Julian

  8. Running RevMan 4.0 under Windows 2000

    9. Rasmus reported several problems with running RevMan 4.0 under Windows 2000 (e.g. printing text of review, saving as HTML and automatic update of links does not work). This is because Windows 2000 handles RTF text (in which the core text of a review is stored) differently from Windows 95/98. Basically, it introduces new codes in the text that RevMan 4.0 is not set up to deal with. RevMan 4.1 will be modified to make it compatible with all existing versions of Windows but until then a patch for Windows 2000 users of RevMan 4.0 is needed. It was agreed that a message about this patch should be sent to the review group coordinators (through the adminors discussion list) and the RevMan discussion list, and that the problem should be noted on the RevMan web page.
    Action: Rasmus

  9. Future plans for the Information Management System

    10. Monica thanked members of RAG for their helpful input on the draft project proposal. Useful advice had also been obtained from people with expertise in funding during her visit to the United States. However, the project proposal will not be further developed until it is clear which funding body(ies) should be approached. A new angle on the project, which may help to make it more attractive to information technology or knowledge management funders, would be to introduce the notion 'tacit knowledge'. Tacit knowledge is also known as implicit knowledge, e.g. knowledge in our heads that makes us act in a certain way. There is a lot of tacit knowledge stored in the various sub-systems of the Information Management System and, by providing suitable information technology tools, this could be made available and shared across the Collaboration. If this aspect is included in the project it will become more than just a new technological infrastructure, it may also have an innovative aspect which may help secure funding from, for example, the European Union. The RAG supported the idea of including exploitation of the tacit knowledge within the Cochrane Collaboration in the project proposal.
    Action: Monica

  10. Survey to investigate the needs for the next generation of the Information Management System

    11. It was agreed that a Collaboration-wide survey should be carried out, and initially that it should be limited to a few broad questions. Once the results of this survey are known, one or more targeted surveys can be carried out. Experienced advice on how to carry out surveys should be sought and a pilot survey should be carried out in advance of surveying across the whole Collaboration. It was agreed to set up a small working group, headed by Mike Clarke, to carry forward the survey(s). Other members of the group should include: Julian Higgins, Philippa Middleton, Paul Whatling, someone from the ModMan Advisory Group (or another representative of review group coordinators), a reviewer from a developing country and a reviewer who writes reviews in a word-processing package rather than in RevMan. Suggestions for these two latter positions should be forwarded to Mike. RAG members input would be from the angle of producing documents (reviews), whereas the representative of ModMan users would add the perspective of managing documents. The pilot survey should be carried out by this year’s Colloquium, which would allow a discussion of the preliminary results. The working group could also consider putting on a workshop at the Colloquium.
    Action: Mike

  11. Progress on RevMan 4.1 beta testing

    12. The Nordic Cochrane Centre had carried out a small survey of the people who have downloaded RevMan 4.1 beta. Of the 48 people who were surveyed, 28 responded. It was agreed to extend the beta testing period until the end of April. The Centre should follow up with a phone call to the people who had registered as dedicated beta testers (and had, thereby, committed to spending at least 8 hours testing the software) to find out if this extension would be sufficient. It was also agreed to identify a few Windows 2000 users, who would be willing to test a revised beta version where the Windows 2000 problems had been addressed (see Running RevMan 4.0 under Windows 2000).
    Action: Jacob Riis, Rasmus

  12. Web-registering for RevMan 4.1 on CD-ROM

    13. Monica briefly presented the draft web-page which is set up to register for a copy of RevMan 4.1 on CD-ROM. A few modifications were suggested. The web-page will be made available when RevMan 4.1 is ready for distribution.
    Action: Jacob Riis

  13. Automatic inclusion of "Conflict of interest" information on the Permision for Publication Form

    14. The Conflict of interest statement in the text of the review is for the review in general and not for the individual reviewers, so the current form cannot be used. It was agreed to add this to the wishlist for consideration in RevMan 5.
    Action: Jacob Riis

  14. Transfer of contact details between RevMan, ModMan and HIREx

    15. This item was referred to the ModMan Advisory Group.
    Action: Sonja

  15. Wishlist for RevMan

    16. The wishlist continues to grow, especially because there is a tendency to request changes rather than report bugs when testing software. The content of the current wishlist should be fed into the needs assessment for the next generation of the Information Management System. Jacob Riis should send the list to Mike for further action. The meeting was also informed that a special e-mail address (revman-wishlist@cochrane.dk) had been set up, to which users should send proposals for changes.
    Action: Jacob Riis

  16. SPAM from RAG e-mail list

    17. It was agreed that the RAG e-mail list should be moderated to avoid the distribution of SPAM.
    Action: Monica

  17. Membership of RAG and creation of additional RAGs

    18. There is still a need to expand the membership of RAG to include more reviewers and Mike encouraged current members to propose additional ones. Centres have been encouraged to establish RevMan advisory groups in other parts of the world to allow for a more diverse input to the development of RevMan (e.g. from low-income countries). CRGs and reviewers who are interested in becoming involved in the work of the RevMan advisory group, but who are unable to attend the meetings held in the UK, should encourage their local centre to establish a group.

    The remit of the RevMan Advisory Group will be published on the SDG Internet site, along with minutes of the RAG meetings. Potential new members, and centres considering supporting the establishment of new groups, can refer to this site to find out what is involved.
    Action: Monica, Jacob Riis

  18. Any other business

    19. It was suggested that the RAG should request the aid of the Cochrane Collaboration Secretariat for taking minutes at its meetings. This proposal will be discussed at the Software Development Group meeting.
    Action: Monica

    Chris provided a brief update on the status of the Comments and Criticisms System. A suggestion had been put forward from the Publishing Policy Group that a BMJ-type comment/criticisms system should be considered instead of the current system. A working group headed by Chris had been set up to investigate this further. It was suggested that this group should contact someone from the BMJ to get their input. Groups and criticisms editors should continue to use the current system until further notice.

  19. Date of next meeting

    20. It was decided that the next meeting should be held late November or the second week of December. Mike will canvass members for possible dates.
    Action: Mike

RAG minutes | MinMan home

Document Actions