Welcome to Issue 28 of the Support Team Digest. The team are back from the Global Evidence Summit, and have lots of news to share. Keep in touch with us, and let us know of anything you'd like us to feature in future digests (cis-support@cochrane.org).
Cochrane networks
Structure and function of review groups
The report on the future structure and function of Cochrane review groups was a big talking point at the Global Evidence Summit.
The report on how the new structure will work was released in August (read it here). This was approved by the Cochrane Governing Board in Cape Town (key decisions are available here). The Support Team and the Information Specialists' Executive met with Karla Soares-Weiser (Deputy Editor in Chief), and Jo Anthony (Senior Media and Communications Officer) during the Summit.
They said:
There will be no changes to which groups are in which network in the foreseeable future
Your contract of employment will still be with your employing institution and your line manager will not change. Cochrane are not intending to make changes to your employment status, the responsibility for this will still be with your employing review group
Further announcements about what happens next will be made in the fortnightly digests issued by the Cochrane Editorial Unit, so keep an eye on these
Each review group has now received feedback on their performance. If you have not seen this report for your group, you are encouraged to ask your co-ordinating editor for a copy
The position of Senior Editor for each network will be advertised shortly
You are strongly encouraged to get in touch if you have concerns or want to raise questions
They plan to collate some FAQs, which they will make available via the web.
If you have any questions or concerns, you can contact the structure and review team directly at this email address: SFinfo@cochrane.org
Alternatively, if you would feel more comfortable contacting the Support Team or your Executive, you are welcome to do that instead. We can take your concerns to the implementation team and anonymise them, if you wish.
We are also happy to be contacted individually if you would rather do that.
Future directions for CISs
Feedback from the workshop at the Global Evidence Summit
There was a great turnout from Cochrane Information Specialists who came to the training event "Future Directions for Cochrane Information Specialists".
This was a whole day pre-conference workshop designed especially for CISs. The idea of the day was to give CISs practical ways to use and to try out the new tools that are available to them, and discuss how the information specialist role might change over the next few years.
Here's some information about the presentations, with a link out to the slides from each session:
Ruth introduced the day, giving us some clear questions to be answered: what tasks could we stop doing? How can new technology allow us to better support our author teams? How could new tools help us to improve existing products?
Doug and Anne framed the workshop by looking at the lifecycle of a study, and how we do things now compared with how we might do them in the future. Instead of going out to search for studies, we might now start to think about having studies come to us instead, via Cochrane Crowd, Classifier and the CRS.
Gordon and Ruth outlined the rationale for the Centralised Search Service, and updated us on which data feeds are currently coming into CENTRAL. PubMed, Embase, ClinicalTrials.gov and KoreaMed records are making their way into the CRS, and CENTRAL. CINAHL, LILACs and ICTRP are on the "to do" list.
Anna and Gordon took us through Cochrane Crowd and how it works, and looked at how it links in to the emerging Cochrane evidence ecosystem. Anna then set us going on a screening challenge - Lions versus Giraffes!
CRS Web (Gordon Dooley, no slides available)
Gordon demonstrated some new features of CIS that were released just in time for Cape Town. One of the features was the search for "included not in CENTRAL" records, which can now be added to the Dashboard. The new deduplication and merging screens were demoed, and we also looked at the triage screen. There was time to play with the new features.
Classifier (James Thomas, Sam Cox & Anna Noel-Storr)
The Classifier is the tool within CRS Web which uses machine learning and text mining to identify RCTs and records within a group's scope. Anna went through how the Classifier is helping to identify RCTs from the feed of records coming in from Embase. Sam then demonstrated how she uses the tool to help her screen the records for the ENT Group's specialised register.
Deirdre took us through the PICO annotation process, and introduced Confluence, a space where CISs can find information and store content related to their PICOs. There was then some dedicated time to annotate some reviews.
Gert from Cochrane's Information and Knowledge Management Department presented on how these tools all link together, and how the next challenge is integrating them with other Cochrane and related products (such as Rev Man and Covidence).
The day was wrapped up with a question and answer session with Chris Mavergames, Cochrane's Chief Information Officer.
If you attended the event, and have ideas about how we can improve it in the future, let us know! cis-support@cochrane.org
The enhanced Cochrane Library
Changes are coming to the Wiley platform...
A beta version of the new Wiley Cochrane Library platform was demonstrated at the Global Evidence Summit by Harriet MacLehose, Ruth Foxlee and Juliane Reid.
The look and feel of the Cochrane Library will change:
There will be links from Cochrane's reviews to the references within those reviews (and vice versa)
A new federated search means you can search the Epistemonikos database alongside the Cochrane Library
Cochrane Clinical Answers will be available via the Cochrane Library
Translations of Cochrane reviews will be available via the Wiley platform
We will say goodbye to the DARE and NHS EED databases, and the old Cochrane module content will shift to individual groups' websites.
How YOU can help...
In order for the links between references and reviews to appear in the Library, there is a job of work to be done by Cochrane Information Specialists. Here's how you can make sure that all of your included studies are appearing in the Cochrane Library with the links enabled...
1. Log in to the CRS
2. From the dashboard, click "Included not in CENTRAL". If this option doesn't appear on your Dashboard, click on the Configure tab to the left, and you can add this option from the drop down menu to your Dashboard from there.
3. This will give you a set of records which are included in your reviews, but haven't been added to CENTRAL. After you've clicked on the dashboard item, these records will appear in the record listing
4. Deduplicate those records, so that you can ensure that you are not adding duplicate records to CENTRAL. You can do this by clicking on deduplication, where you will get to choose the deduplication filter that you prefer.
5. Go through the identified duplicates, and merge any that that are the same.
6. When you are happy that there are no duplicates in the list, you need to check the records over to make sure they are compliant with the HarmoniSR guidance (you can find that on the CIS Portal here, please note that if you click on "guidance" in any record in the CRS, you can also find it there)
7. If all the records in the listing are eligible for CENTRAL publication (ie are RCTs, CCTs, Interrupted Time Series or Controlled Before and After Studies) then click "Add to" on the menu bar above the record listing, and select "Add Records to CENTRAL".
If you have any questions about any aspect of this, contact us.
Webinar
Webinar on PICO Annotator tools
Confluence and Slack are two of the tools that are used in PICO annotation. A drop in webinar is taking place on 12 October at 2pm BST, for those CISs who have questions regarding Confluence or Slack.
If you have a question or an issue, register here.
After registering, you will receive a confirmation email containing information about how to join.
Useful papers
Predatory journals
The Information Specialists' meeting at the Global Evidence Summit featured a discussion on predatory journals, and two papers were highlighted by Liz Stovold which may be of interest.
Predatory journals have been defined as: "an exploitative open-access publishing business model that involves charging publication fees to authors without providing the editorial and publishing services associated with legitimate journals (open access or not)." (Wikipedia, link here)
These journals are becoming an increasing issue in academia, and the discussion centred around the role of the information specialist in identifying these journals and highlighting them to authors. Those who were at the meeting concluded that although publication in a "predatory journal" would not exclude a study from a Cochrane review in and of itself, there may be a responsibility for us to warn authors about these journals.
Anna Noel-Storr and colleagues have recently published an article on machine learning in the Journal of the American Medical Informatics Association. The article looks at a hybrid approach combining machine learning and human effort (via crowdsourcing):
Discuss and share knowledge with other Cochrane Information Specialists
The Information Specialists Forum is now up and running. Go to http://forums.cochrane.org to sign up. You can join existing conversations or propose your own topics.
We're hoping that the Forums will be a great addition to our other ways of communicating. While the mailing list is great, it doesn't allow us to archive or tag content so that we can search for it again. The forums will also make it easier to follow a discussion thread.
Some topics have already been set up, including:
CRS Web - software updates
PICO and Confluence
HarmoniSR standards
Study based registers
Covidence
Archie for CISs
Journal Club
First meeting of the journal club
The inaugural meeting of the journal club took place on 22 August via the web. It was a great session hosted by Justin Clark of the CIS Exec.
The study addressed a clearly focused question: is an objective search better than a conceptual one?
Discussion summary
The recruitment of participants was not considered representative of the general systematic review population, but the group acknowledged the difficulty of recruiting and were complementary of the fact the authors did manage to recruit five organisations to be involved in their study. The measurement of the effects were free of bias as it was an objective measure, did they find all the studies and how many irrelevant studies did they find?
The authors did not take into account the many confounders that can impact their results, primarily the fact that the authors used the same method, in the same location each time while the comparison group was different each time and the methods could also vary widely. This was a major flaw in the study and was considered too large a problem to make the conclusions drawn believable.
There was not enough differentiation between the objective and conceptual approaches. Most of the attendees at the journal club already use most of the techniques outlined by the authors, so there was some uncertainty as to whether it was a fair test as normally a new method is tested against the current best method, and in this study this was not the case.
The outcome of the study was that the objective approach, the approach favoured by the authors, was superior the approach used by the comparison group. The journal club attendees, although not faulting the objective approach, considered the study did not prove this point. Therefore this article will not impact on the current practices of the group. The attendees agreed that we would benefit from a detailed introduction to the objective approach from a member of the IQWiG team.
Get involved!
The Exec will be holding another journal club soon, and all CISs are welcome to join. Contact Robin Featherstone to be added to the list.
Contact us...
Don't forget, previous editions of the Cochrane Information Specialist Support Team digest are available on the CIS Portal
Contact the Support Team for help with any Cochrane Information Specialist related issue (including CRS technical support): cis-support@cochrane.org
The Cochrane Information Specialist Support Team:
Charlene Bridges | Sam Cox I Ruth Foxlee
Anne Littlewood I Doug Salzwedel
