TSC Support Team digest
View this email in your browser

Issue 11 | April  2016

Welcome to our 4th digest of the year! In this issue we look at the recent text mining webinar, we answer a question about top up searches that are not incorporated in the review, and share a search tip. Our new email address is, but the old address will still work for the next few months. Don't forget to let us know of any Cochrane Information Support queries you might have, including CRS technical support.
Text mining webinars
The Support Team were delighted that so many of you joined the first of the webinars which we have arranged with Julie Glanville from the York Health Economics Consortium. Julie spent an hour and a half introducing various text mining tools, demonstrating them and giving real-life examples of how and when she had used them.

You can download the Blackboard session for yourself here to catch up if you missed it.

Three further webinars are planned, and there will be homework if you want to do it:

Monday 16th May (3pm BST, 2pm GMT)
Session 2: Identifying search terms
  • How do we typically identify search terms?
  • How do word frequency analysis tools help us?
  • Can MeSH finders help us?
  • How can we identify phrases?
  • How can we identify word adjacency?:
  • Questions and discussion
Monday 13th June (3pm BST, 2pm GMT)
Session 3:  Developing structured conceptual breakdowns (PICO, ECLIPSE, SPICE etc.)
  • How do we typically identify the concepts we are going to use in our search?
    • What are the challenges of choosing a conceptual breakdown?
  • What other approaches have you tried out, other than PICO?
  • Presentation of selected  non-PICO conceptual breakdowns
    • What are they?
    • What are they for?
    • How do they work?
  • Questions and discussion
Monday 11th July (2pm BST, 3pm GMT)
Session 4: Moving beyond structured conceptual breakdowns – searches with multistranded approach
  • What are the challenges we face when developing searches for complex topics?
  • What do we do when we can't fit our question into PICO or another conceptual breakdown?
    • Present examples of 'multi-stranded' approach
  • How do we know when we want to use this approach?
  • How do we describe and document this approach in a PICO-oriented world?
  • Questions and discussion
We have created a new page on the Information Specialists' Portal for the webinar series, and we'll be adding to this over the coming months.
References and RevMan

Getting references into RevMan

Graham Chan recently shared an improved EndNote filter for downloading references into RevMan, as the version available through Cochrane Tech was not handling reference identifiers properly. You can download Graham's improved version from the Portal, you can find it under the tab "Software & Projects>Review Manager". Graham adds the following note of caution:

"Neither the original nor mine outputs the Reference Type in the case required by RevMan, so RevMan imports everything as Reference Type Other... To fix this in EndNote would require extensive changes that might have unpredictable knock-on effects (I don't think anything can be done in the output style to fix it). Until the new, Web version of RevMan becomes available, if you don't want everything imported from EndNote to be Reference Type Other, and you don't want to manually edit every imported reference, I suggest you open the exported file in Notepad or Word before importing it into RevMan, and do a global Find/Replace to change the commonest reference types to the required case, e.g. change all "RT:Journal Article" to "RT:Journal article".

An alternative way of handling this is to import the references into the CRS, as you can then directly export them in the right format for RevMan. You need to make sure you choose the RevMan filter when you re-export the records from the CRS:

Updating searches

Question from the CIS Support mailbox: search dates

CIS Support have been asked about how to report search dates and prepare PRISMA flow diagrams when the results from a top-up search for a review have not been incorporated into the included and excluded studies.
Search dates:
The Date of Search has to reflect the date for which search results were fully incorporated into the review.  
So, if you run a top-up search but simply add the results to Studies Awaiting Classification instead of assessing them, you can put those new search dates into an appendix, but you can’t change the Date of Search or the search dates mentioned in the Search methods section in the main text of the review.  
The Search methods may describe the top-up search and state how many studies have been placed in Studies Awaiting Classification.
PRISMA flow diagram:

The PRISMA flow diagram should reflect the number of studies in the ‘Studies awaiting classification’ section.

The Cochrane Editorial and Publishing Policy Resource contains more information on reporting search dates in Cochrane Reviews.
Corrections to CENTRAL records

Using CRSO to change CENTRAL records

Thank you all for your contributions to tidying up CENTRAL records using the CRSO corrections mechanism. As some of you will be aware, it has proved difficult to manage this process. The Support Team, in conjunction with Metaxis, are dealing with the backlog of changes at present.
CENTRAL Corrections, via CRSO will be retired with the launch of CRS Web. There are a number of features of CRS Web that will make this facility redundant, including the HarmoniSR filter, and the option to make authority records, although there will still be a mechanism in CRSO to submit requests to remove records from CENTRAL
The facility to edit records in CRSO will remain until everyone is moved over to CRS Web, but we suggest this isn’t something that Cochrane information specialists should do as we anticipate that many of the changes that you would like to make to shared records will be done automatically in the future.
Search tip

Did you know about frequency searching in Ovid?

A recent discovery by the CIS Support Team is the ability to search for frequently occuring terms in Ovid. Records containing the term are retrieved only if that term occurs at least the number of times specified in your search statement. This could potentially help in a search where a higher level of precision may be important. In general, records that contain many instances of your search term are more relevant to your search than records that contain fewer instances. The frequency operator is particularly useful when searching a database’s Full Text field for a common word or phrase (the .tx. field). It can't be used on multiple fields (so for example, would not work using .ti,ab.) or with fields that consist of a combination of other fields (eg the Source field).

The syntax is:


So for example: toothbrush$.ab. /freq=2 would retrieve only those records where toothbrush appears at least twice in the abstract. Try it out for yourself if you're an Ovid user.
Style Guide reminder!

Cochrane Style Guide

A reminder that the Cochrane Style Guide is part of the Cochrane Editorial and Publishing Policy Resource (EPPR). The link to send authors to is now:
Contact us...
Don't forget, previous editions of the Cochrane Information Specialist Support Team digest are available on the Portal in the Support Team section (Archie login required)

Contact the Support Team for help with any Cochrane Information Specialist related issue (including CRS technical support):
The Cochrane Information Specialist Support Team:
Liz Doney I Sam Faulkner I Ruth Foxlee
Anne Littlewood I Doug Salzwedel
Cochrane Information Specialist Support Website
Cochrane Information Specialist Support Website

This email was sent to <<Email Address>>
why did I get this?    unsubscribe from this list    update subscription preferences
Cochrane TSC Support Team · Cochrane Editorial Unit · 57-59 Haymarket · London, SW1Y 4QX · United Kingdom

Email Marketing Powered by MailChimp