Personal tools
You are here: Home → IMSG → RAG → Minutes → RAG minutes, 20 March 2002
 

RAG minutes, 20 March 2002

Minutes approved on: 14 November 2002

Present: Gerd Antes, Phil Alderson, Jacqueline Birks, Chris Cates, Mike Clarke, Katherine Deane, Monica Fischer, Sonja Henderson, Katherine Henshaw, Julian Higgins, Valeria Marinho, Heather Maxwell, Tess Moore, Rasmus Moustgaard, Elizabeth Pienaar, Carrol Preston, Mark Starr

In attendance: Jacob Riis (minutes)

Apologies: Clive Adams, Dymphna Hermans, Bill McGuire, Phil Wiffen

Contents:
1. Welcome
2. Approval of the agenda
3. Approval of minutes and matters arising from the previous meeting (13 December 2001)
4. Distribution of future agendas, minutes and background papers
5. Introduction of Generic inverse variance method to RevMan
6. RevMan 4.2: testing and release
7. Needs assessment survey
8. Needs assessment survey in developing countries
9. Withdrawn/suspended Cochrane protocols and reviews
10. Advice on reference handling in RevMan
11. Automatic labelling of references imported to classification pending
12. Design of the RevMan wishlist
13. Content of the RevMan wishlist
14. Membership of the RevMan Advisory Group
15. Input to the RevMan Advisory Group from outside the UK
16. Budget for the RevMan Advisory Group
17. Any other business
18. Date of next meeting

  1. Welcome

    2. Mike welcomed everyone to the meeting, especially the new members, Jacqueline Birks and Katherine Henshaw. Apologies for absence had been received from the people listed above.

  2. Approval of the agenda

    3. The agenda was approved.

  3. Approval of minutes and matters arising from the previous meeting (13 December 2001)

    4. The minutes of the previous meeting were approved.

    1. Archiving of The Cochrane Library (item 3.d)

      2. Mark clarified that Update Software has an archive of Cochrane reviews and pre-Cochrane Library reviews from the Pregnancy and Childbirth Group. People wanting a copy of these archived reviews should contact Update Software.

    2. Multi-user version of RevMan (item 3.e)

      3. Tess and Heather reported that they had not yet tested the multi-user version of RevMan 4.1. Sonja has been using it and had encountered some problems, which she had reported to Rasmus.

      The need for multi-user functionality was discussed, and Sonja explained the need for it in editorial offices that deal with a large number of draft protocols and reviews on a daily basis. She explained the importance of keeping all draft protocols and reviews, submitted for editorial consideration, in a single version of RevMan to ensure that staff who need access to these reviews are confident that they are accessing the correct version. Because RevMan does not currently have multi-user functionality, editorial offices have to find ways to cope with this limitation which may not be the most efficient way to work. For example, staff having to export individual reviews from a 'central RevMan' to their own version of RevMan to work on, ensuring that they then import their 'updated' version back into the 'central RevMan'.

      Rasmus explained that if a multi-user version of RevMan 4.2 is made, it will be specifically for CRGs. It would be produced by adding the RevMan 4.2 features to the code for version 4.1 multi-user. However, because of technical problems, it might not be possible to achieve full and reliable multi-user functionality without rewriting RevMan completely. Tess, Heather and Sonja agreed to test the multi-user version of RevMan 4.1 further.
      Action: Tess, Heather, Sonja

    3. RevMan 4.1 - RTF difficulties (item 5)

      4. Rasmus reported that he and Lois Sims continue to collaborate on solving a few problems for each issue of The Cochrane Library.

    4. Exporting reviews (item 7.c)

      5. Rasmus had included a draft export window in the pre-beta version circulated to the group.

    5. Internal/External vs Intramural/Extramural (item 7.d)

      6. Rasmus reported that the headings Intramural/Extramural had been changed to Internal/External in RevMan 4.2.

    6. Totals of events (item 8.e)

      7. Julian reported that this still awaits discussion within the Statistical Methods Group.

    7. Reference management files (item 9: Needs Assessment Survey)

      8. Rasmus reported that people willing to be responsible for the reference management files accompanying RevMan had been identified for the Reference Manager and ProCite files. Katherine Deane volunteered to help with the EndNote files.

    8. Validation of references in Cochrane reviews (item 10)

      9. Heather reported that the question of whether some of the reference formats in RevMan should be restructured would be discussed by the UK-based TSCs at their meeting on 22 March 2002.

  4. Distribution of future agendas, minutes and background papers

    5. It was agreed that the distribution of all the documents necessary for the meeting in a single PDF file was helpful and that this practice should be continued.

    1. Minutes and papers on the web

      2. Monica and Jacob will discuss with David Booker whether his plans for developing the main Collaboration web site should influence the practices governing the storage and presentation of minutes and papers of the IMSG and its advisory groups, and report back to the RAG.
      Action: Monica, Jacob

  5. Introduction of Generic inverse variance method to RevMan

    1. Calculating the data

      2. It was noted that the IMSG had agreed that RevMan 4.2 should generate raw data for submission to the publisher, who should then recalculate the data for publication. This was endorsed by the RAG.

    2. Documentation

      3. Guidance on when and how to use the new method will be part of the revised statistical section of the Cochrane Reviewers' Handbook.

      It was agreed that a separate introductory document about the new method should be published to coincide with the release of RevMan 4.2. Julian agreed to oversee the preparation of this document.
      Action: Julian

  6. RevMan 4.2: testing and release

    7. The beta version of RevMan 4.2 will be completed in the next few weeks. The plan is to release this when the internal testing of RevMan Analyses by statisticians in Liverpool has been completed. It is expected that a date for the release of the RevMan 4.2 beta will be announced in early April.

    1. Feedback on pre-beta

      2. Rasmus requested that any additional feedback on the pre-beta version be sent to him by 2 April 2002.
      Action: Everyone

    2. Beta testing

      3. Monica described the plans for the beta test. The beta test will be scheduled to last 6 weeks, and testers will be encouraged to provide continuous feedback during the testing period. The testing will include formal testing both by a person at the Nordic Cochrane Centre and by a group of dedicated external testers. Dedicated testers will be registered and asked to complete a structured report on their testing. It was suggested that the dedicated testers could be recruited by asking RGCs to suggest people. Messages will also be sent to the RevMan, Adminors and TSCs email lists. A third, informal category of beta tester will be people who just download RevMan 4.2 beta to try it out, but who will also be encouraged to submit reports of any problems they encounter.

      Monica will circulate the feedback form and the schedule for the release and testing to the RAG when these have been finalized.
      Action: Monica

    3. Using beta with live data

      4. The importance of not using the beta version for 'live' data was stated. It was agreed that it should not be possible to submit live data reviews from the beta version. The warning screen in the beta version will include a warning against working on 'live' reviews in RevMan 4.2 beta and passing them through RevMan 4.1 for submission.
      Action: Rasmus

    4. Submissions using Revman 4.2

      5. Mark reported that Update Software plans to be ready for submission in RevMan 4.2 format for issue 1, 2003. It was agreed that testing should include a test of the submission phase.

  7. Needs assessment survey

    8. A report on the results of the Needs Assessment Survey prepared by Mike, Julian, Tess and Phil had been circulated as a background paper for the meeting. Mike explained that the report would be discussed by the IMSG later the same day and that, based on this and other information, the Cochrane Collaboration Steering Group are expected to agree on a set of recommendations for action, at their next meeting in April 2002.

    1. Replies from entities versus replies from individuals

      2. Sonja suggested that data from editorial bases should be extracted and reported separately and given more weight in setting priorities for action. Tess and Mike reported that there seems to be consistency between the replies given by entities and individuals, and that the overall conclusions were in keeping with the responses from editorial bases. Some overlap was also noted between this survey, the earlier tracking survey and the work of the Cochrane Collaboration's Quality Improvement Manger (Nancy Owens). Mike clarified that Nancy will discuss her preliminary findings with the Cochrane Collaboration Steering Group alongside this report, so her feedback will be part of the discussion.

    2. Possible follow-up

      3. Mark cautioned that there was a low response rate for individuals, and the possibility of a short follow-up survey aimed at reviewers was discussed. Chris suggested that the response rate for a follow-up survey may be increased by making the benefit of replying clear. It was agreed that RGCs should be part of any discussions about doing a further survey but that the development of the Information Management System should not be delayed while this was considered.

    3. Recommendations

      4. The 10 recommendations of the report were discussed, and it was noted that there was overlap between some of the items. It was also questioned whether there should be a division of priorities into two levels in the report, although it was noted that the division in the report accurately reflected the original data. The RAG made the following specific points about the recommendations:

      1. Item 1 - Contact details

        2. Item 7 should be integrated with this.

      2. Item 3 - Word processing

        3. The mention of Microsoft Word should be removed.

      3. Item 5 - ModMan

        4. The word "additional" should be removed since there currently is no training in ModMan.

    4. RevMan comments

      5. Rasmus reported that he had gone through all the free text comments that had been submitted about RevMan and classified them into three categories:
      • issues regarding functionality that already exists - documentation may need improving.
      • issues regarding functionality that will be available in RevMan 4.2.
      • issues that were added to the wishlist.

  8. Needs assessment survey in developing countries

    9. Elizabeth reported on the preparations for the survey of reviewers in developing countries. The overall plan can be divided into two phases, where the first phase is identifying the needs of reviewers, and the second is defining solutions. Preparation for this new survey should include looking at the answers from the completed survey to identify any questions that may have lead to misunderstandings. This will also involve looking at all the free text replies. A number of suggestions for additional questions for the survey were made:
    • Do people have access to statistical support?
    • What language do they write in?
    • How do non-English speakers feel about copy editing done by the editorial base?
    • How important is the language of software?

    As well as targeting people in developing countries, it was agreed also to try to get people from developing countries who currently live in a developed country to complete the survey.

    1. Access to The Cochrane Library for developing countries

      2. Mark reported that all countries on the WHO list of developing countries now have free access to The Cochrane Library, and he agreed to circulate a message within the Cochrane Collaboration about this.
      Action: Mark

      It was suggested that information about free access to The Cochrane Library should be advertised in the document acommpanying the survey for developing countries, and announced on relevant web sites.

  9. Withdrawn/suspended Cochrane protocols and reviews

    10. The background paper containing the Handbook Advisory Group's recommendations to the Cochrane Collaboration Steering Group was discussed. The RAG agreed with these recommendations:
    • Withdrawn reviews will appear in The Cochrane Library with only title, authors and published notes.
    • The entry for withdrawn protocols will only appear for one issue, and would thereafter be noted in the Review Group's module. It was agreed that information about the withdrawal of the protocol should be included in the module at the same time as the withdrawn protocol is published in The Cochrane Library.
    • It was agreed that split reviews should be linked through the reference to the other published version.
      Action: Mark
    • It was noted that there was a need for clear guidance on the things to keep in mind when splitting reviews. The Handbook Advisory Group will be asked to prepare this guidance.
      Action: Mike

    1. Archiving protocols

      2. Mark confirmed that all published protocols will be archived.

  10. Advice on reference handling in RevMan

    11. The issues relevant to this item were discussed at other points in the meeting.

  11. Automatic labelling of references imported to classification pending

    12. Phil had provided a background paper arguing that the reference IDs for references imported into Classification pending are not strictly necessary, and instead of the IDs generated for those, it would be more helpful to be able to view the whole reference. It was agreed that this should be added to the wishlist, with a high priority.
    Action: Rasmus

  12. Design of the RevMan wishlist

    13. Rasmus reported that a search function had been added to the wishlist site.

  13. Content of the RevMan wishlist

    14. It was agreed that items which have already been implemented should not be kept on the wishlist.

    1. Wishlist items


      2. Item Title Priority Notes
      84 Future of 'other data tables' Rejected
      85 Cancelling shut down High
      87 Display RevMan citation more prominently in documentation High Will be implemented in Revman 4.2
      109 Calculated data High
      110 References to Cochrane reviews High
      111 Number of decimal places High Should not truncate exact 0s
      112 Undo delete High Should only work for most recent deletes
      113 Print classification pending references Already possible
      114 Folders in tree view High
      115 Sort descending (Z-A) Low
      116 Sort options same as ModMan High ModMan has sort on list of reviewers

  14. Membership of the RevMan Advisory Group

    15. Mike noted that the number of members had increased, but that it would still be good if more reviewers could be recruited.

  15. Input to the RevMan Advisory Group from outside the UK

    16. There had been interest from a few Canadian reviewers to join the RAG and Mike had encouraged the Canadian Cochrane Centre (CCC) to set up a RevMan user group. The Group supported the idea and Mike agreed to follow-up with Arne Ohlson, director of the CCC, who himself had expressed an interest in the Group's work.
    Action: Mike

  16. Budget for the RevMan Advisory Group

    17. Monica reminded everyone to send in their claims forms as soon as possible, as the Cochrane Collaboration's financial year ends on 31 March 2002.
    Action: Everyone

  17. Any other business

    18. none

  18. Date of next meeting

    19. It was suggested that this should be in the week beginning 11 November 2002. Monica and Mike will discuss this in relation to the IMSG meeting that will take place alongside that RAG meeting.
    Action: Monica, Mike

RAG minutes | MinMan home
Document Actions