RevMan 5 Conversion Tips

RevMan 5 Conversion Tips

Tips for authors using RevMan 5 for the first time

Here is a quick set of tips for manoeuvring through Archie and RM5 (the term ‘review’ refers to all stages – title, protocol and review)

Download RevMan 5 from http://www.cc-ims.net/RevMan/. You will notice on the panel on the left side of the screen, a link for ‘authors’, which will take you here http://www.cc-ims.net/authors. This is an excellent resource page that gives you information on Archie, RevMan 5, GRADEProfiler and how to start using them. Take the time to complete the tutorial on RevMan 5, located under the ‘help’ button at the top of the screen. Although the basics are easy to learn, RevMan 5 has lots of new and time-saving features that will only be realized if you take the time to find out about them.

Before you can start to work on your review in RevMan 5, you need a user account. With this user account, you can update your own contact information and access reviews and files for which you have access permission.

All files must be started by the Editorial Office. This is done once the title is registered and the review team members are established.

Check the review out using RevMan 5, not Archie (the Archie route is technically possible, but is preferable). Go to File > check out > click on the title you want > OK.

TIP: do not choose the option ‘download a copy of the review without locking it for others’ – your co-authors will not know you are working on the file and may check it out to make some of their own edits > RevMan does not merge files, so someone will end up copying/pasting their work!

When you start working on the RevMan 5 file, use the ‘save as’ function and save it to whatever drive/folder you ordinarily store your reviews – thereafter you can just ‘save’ and ‘close’ without losing it somewhere on your hard drive! However, files should only be saved temporarily on your own computers – checking them back into Archie ensures you will always edit the most current version.

The same person (i.e. same user account) must check the review out of Archie and back in (but you can do it from different computers if you have saved the file on your travel drive).

Remember that while you have the file checked out, nobody else can access it.

TIP: don’t store the file for a long time on your own computer – check it back into Archie. The most recent version will always be available and you won’t lose data when your home computer crashes!

TIP: delete all RevMan 5 versions of the review from your computers when you check it back in to Archie, so that you do not inadvertently work on the wrong version. Archie will only allow you to check in the version you checked out.

TIP: sharing the file: depending on the size and proximity of your team, you can decide whether it is more expedient to share the file through email or have the members of your team check it out/into Archie (in which case everyone will need a user account). Depending on your co-authors’ comfort with RevMan 5, you may find it easier to share the html or word document via email. You can only export the review from RevMan 5
as an html file – remember that some firewalls will block html files. In that case, you may be better off to export it and save it as a word document.

Checking the review back into Archie before it is finished:

When you check in the file, you will be asked if you want to check it in as a ‘draft’ or for ‘editorial approval’. If the team is still working on it, enter a description of the version (e.g. ROB tables finished, data entered) so that you and others will remember/know what you did. You will get a message saying it was successfully checked in.

If you want Vicki to review it prior to completion, check it in as a draft and send Vicki an email letting her know.

Checking the review back into Archie for editorial consideration:

Go to file > reports > validation report. Run the validation check and fix all the ‘errors’ and ‘warnings’ you can. You can check a file in to go through the editorial process with ‘warnings’ but not ‘errors’.

Check it in for ‘editorial approval’ rather than as a draft – this will generate a message to Vicki – you can type a message in the text box, thus alleviating the need to send another email, unless you have an attachment you wish to send.

Q & A on the conversion to RevMan 5

Q. I am half way through doing an update of my review in RevMan 4. Can I continue to finish in RevMan 4?
A. Yes please — continue in RevMan 4. We will convert it once you submit it (we can do that at the editorial office, it's just that the authors cannot).

Q. When do I start using the new risk of bias assessment criteria?
A. If you have finished the quality assessment of all your included studies already, don't worry about the new guidelines. However, if you have not yet reached that point of your review or update, then please follow the new guidelines for the risk of bias assessment for all the newly included studies. You can go back and extract supporting text for the other included studies if you wish, but we don't require it at this point.

Q. How should I share the RevMan 5 document with the review team?
A. There are a few ways you can circulate the manuscript:

export the text as an html document; circulate amongst the team and then import the same document back into RevMan 5 — this only works if the major headings are not altered in any way and must be checked carefully, since sometimes it doesn't import perfectly, so you might have to copy and paste some sections. You also have to remember that some firewalls are set to reject html attachments.

export the text as an html document, and then save it as an MS Word document. You will lose some formatting and will not be able to import it back into RevMan 5 (converting it back to html and then importing it is possible but again, not perfect), but can copy and paste the areas that were changed.

each person can download and learn RevMan 5, and then check out the review, suggest their changes in track changes and check it back in for the next person (remember that only one person at a time can check the review out).

you can export the file into RevMan 5, save the file when you are finished and send the RevMan 5 file to your team — but you must be the person who checks in the same file you checked out, or it will not be accepted, and you will have to ‘undo’ the checkout (or get the editorial office staff to help you). This can be done, but you lose all changes in this process. You also run the risk with this option of the RevMan 5 extension being stopped by firewalls — which is what happens at our end! (and we have to have our IT department ‘release’ it)

Personally, I suggest you consider option 2 — I suspect that is the way most folks handled RevMan 4 files and it will make more sense to do the same with RevMan 5. Some small and savvy teams will choose option 3, but everyone has to learn the software or you will end up with a very interesting file!

Q. I finished with the review and I'm ready to have it verified by the editors. But, I can't seem to check in the review. I get an error message that says "The server is unable to perform this action (Document exception: Document has an invalid version number)." Any ideas what I should do?
A. I think that it means you checked in a different file than you checked out, or you are trying to check the same file in for a second time — Archie is very specific about it. In order to make changes, you have to check the file back out again, make the changes, and check it back in. You can't check it in and then continue working on the version you saved in your e-files, which is what it sounds like you did — so Archie sees that you are trying to check the same version in again and won't allow it.

So, I'm sorry to say, you will have to check the review out, copy and paste in your changes, and then check it back in to Archie — if it is ready for us to look at, then click on ‘editorial approval’. The good news is that copying and pasting is easier in RevMan 5 than it was in RevMan 4 — you can copy whole paragraphs/sections, as long as you don't try to copy over one of the headings.

Q. I'm finding it confusing to determine which outcomes to enter into my GRADE Table — there are many that have similar titles and its not until I ‘view’ the Summary of Findings table that I realize I don’t have the information I thought I would have. Any tips?
A. It is important to determine a priori which comparisons (called profiles in GRADEPro) and which outcomes you will include in your Summary of Findings (SOF) table. Each comparison generates a new SOF table, because it is a separate clinical question. Remember that GRADEPro is a separate software from Revman 5 — they communicate, but only RevMan 5 is stored in Archie.

Tips for getting started with your SOF tables

create a file title in your e-files for your GRADE/SOF tables. This title can be the same as your review title and will be a spot to keep your working tables/files.

check the review out of Archie; make sure you have completed the analyses in RevMan 5 and saved the file in your e-file.

open GRADEPro and import the data from RevMan – data from all analyses will be imported, therefore, you must select the comparisons/outcomes you wish to include and delete the rest.

TIP: in Archie, select your review > go to ‘history’ > click on the most recent version > choose ‘view’ > choose ‘view graphs’ > view graphs > print all the graphs so you can refer to and select the correct ones. If you have a pdf writer program, you can print/save it in pdf to share with co-authors.

click on the comparison on the left of the screen to bring the data onto the screen.

delete the comparisons/outcomes you do not wish to include; add comparison/outcomes that you identified as important but for which you could find no data – this also gives important information in the SOF table.

complete the information for all fields (double click on an outcome to open up the data file – toggle between both ‘go to quality assessment’ and ‘go to summary of finding’ and complete data in both views.

TIP: in RevMan 5, you may find it handy to print your Risk of Bias summary table so you have the info in front of you to complete the ‘quality assessment’ view.

view the SOF table – if it has all the info you wanted, then export it as a SOF table and save it in your e-files.

TIP: save each SOF with the name of the comparison.

open the RevMan 5 file > Tables > Summary of Findings table > add a summary of finding table (wizard will open) > find the file in your e-files (.sof extension) and open it (follow instructions on screen). Repeat with each SOF table you wish to include. Link in the text.

Tips for converting RevMan 5 files into Spine journal format

1. Text of review: you can export the full review in html format, save it, then open and save it as a word document. It will include all tables, graphs and figures, which can be copied and pasted from the file, but not edited.

2. Figures: will be saved as individual .png files. You can open them in Microsoft Office Picture Manager, where you can edit them if needed and export them as a .tif, .jpg or .gif file. According to the Spine web site, these are acceptable formats in which to include your figures.

“To ensure a quality experience for those viewing supplemental digital content, it is suggested that authors submit supplemental digital files no larger than 10 MB each. Documents, graphs, and tables may be presented in any format. Figures, graphics, and illustrations should be submitted with the following file extensions: .tif, .eps, .ppt, .jpg, .pdf, .gif.” (http://www.editorialmanager.com/spine/, accessed 27 October 2008)

3. Risk of bias (ROB) Summary Figure: Since this figure is automatically generated as studies are added/subtracted in the Characteristics of Included studies table, it must be edited in the same way. If you have included additional ‘arms’ or ‘comparisons’ of an original study, you will have to remove them in order to remove them from the ROB summary figure.

In RevMan 5, right click on the name of the incomplete study in the tree view on the left of your screen and delete it. Once you have deleted the unwanted and incomplete ‘studies’, you should have a ROB Summary Figure that only lists the included studies (e.g. Maison 2007).
Double-click on the ROB Summary Figure in the tree view on the left of your screen, which will activate a ‘save’ icon on the top right-hand corner of your screen. Click on the ‘save’ icon and save the ROB Summary Figure as a .eps file. According to the Spine web site, this is an acceptable format in which to include a figure.

4. GRADE or Summary of Findings Tables: These will be exported as part of the html format and may be edited as per #2 above. However, if you need more editing than this option allows, you will have to export them from GRADEPro as a word document.

Note: If you have created the GRADE or Summary of Findings tables using GRADEPro software yourself, you should export them as an image and save them as .jpeg, .gif, .bmp or .png files. RevMan 5 requires imported figures to be in .png or .jpg format. If you save them in .jpg format, they will be available for both RevMan 5 and Spine (and you will save yourself a step!).

5. Forest plots: In RevMan 5, open the analysis for the comparison and outcome of interest. Remember that in RevMan 5, you now have both the data and the forest plot in one screen – a series of icons will be activated in the top right corner of the screen.

Click on the ‘forest plot’ icon – it will open into another window.
Click on the ‘save’ icon in the window and save the image as a .eps file.
Do this for each forest plot you wish to include.

TIP: I would strongly suggest that you check out the published version of your review in RevMan 5, save it in an e-folder ear-marked for your Spine submission, and then check it back into Archie before doing any exports or edits. That way, you will have the published copy of your review available on the Cochrane central server for any updates, etc and have a ‘working’ copy that you can edit for other formats – without having to worry about confusing the two.

TIP: Remember to keep the number of tables and figures to a minimum. Spine states it will include five tables and eight figures. Do not present the data as (for example) both a GRADE table and a forest plot.

Note: Remember that any submission specifications of print journals are the responsibility of the authors.