Professor Peter Sandercock,

1
Part 2 How to do a Cochrane systematic review

• Professor Peter Sandercock,
• University of Edinburgh
• Co-ordinating Editor Cochrane Stroke Group

European Masters in Stroke. Krems April 2007
2
Why do a systematic review?(what does the author
get out of it?)

• An answer to your own burning question
• Education in research methods, EBM (critical
  appraisal), trials and statistics
• Publication and presentation
• at meetings
• electronic publication on Cochrane Library
• publication in print journal
• Travel to Edinburgh for mini-sabbatical

3
Outline

• Register with Cochrane Stroke Group
• Identify question / hypothesis
• Title register with Cochrane Stroke Group
• Protocol
• Review
• (Publish in paper journal, e.g. Stroke)
• Update review routinely, criticisms, new trials

4
Some facts about systematic reviews

• Theyre NOT all about meta-analysis and
  complicated statistics
• Its mainly a question of finding ALL the
  relevant evidence (and not just the one trial
  your boss always quotes, or the one you read
  about last week)
• Systematic reviews seek to reduce bias
• They seek to increase precision (i.e. overcome
  the fickle effects of the play of chance)

5
Getting started

• Register with Cochrane Stroke Group
• http//www.dcn.ed.ac.uk/csrg
• Talk with existing reviewers
• Collaborators?
• Contact others who are interested in the topic
• Plan time (sabbatical, after hours / hobby)
• Review manager 4.1 (PC, not Mac)
• download from web
• Attend training course

6
Title

• Check Cochrane Library (protocols, reviews)
• Titles not listed on Cochrane Library
• Register title with Cochrane Stroke Group
• Hazel Fraser email hf_at_skull.dcn.ed.ac.uk
• Title
• Authors
• Timelines
• Aim of system avoid duplication

7
Protocol

• Background
• Objectives
• Criteria
• Types of studies
• Types of participants
• Types of interventions
• Types of outcome measures

• Search strategy
• Comparisons
• Data
• References
• Protocol is then submitted for peer review
• Work on review cannot begin until protocol
  published on Cochrane Library

8
(No Transcript)
9
(No Transcript)
10
Review

• Searching
• - Own strategy Medline, Embase, ISI, non
  Cochrane reviews, trial reference lists
• - Stroke Group strategy electronic, hand
  searching (Brenda Thomas)
• Screen titles
• Get copies of papers likely to be eligible
  (irrespective of publication language)
• May need help from editorial base with
  translation/getting copies of obscure journals

11
Review 2 data

• Authors read each paper
• Decision include, exclude, pending, ongoing
• Collect information (protocol and data) from each
  trial publication independently
• Contact trialist if information details missing
• Enter trial information
• - Included, excluded studies
• - Data tables
• Check entered information

12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
Review 3

• Methods
• Description of studies
• Methodological quality of included studies
• Results
• Discussion
• Reviewers conclusions
• - Implications for practice
• - Implications for research
• Acknowledgements
• Conflict of interest

17
(No Transcript)
18
(No Transcript)
19
Writing up the results turning data into words

• Discussion
• Implications for practice
• Implications for research
• Plain language summary
• Please have a look at the worksheet which deals
  with the interpretation of a systematic review of
  the glycerol trials

20
Comments / criticisms system

• Maintain quality, balance, completeness
• Critic notifies Cochrane Criticism Centre (CCC)
• CCC notifies Comments Editor (CE)
• CE notifies authors
• Comments editor moderates authors response
• - Authors integrate response in review
• Response published on Cochrane Library

21
Updating a review

• Cochrane about most up to date source of
  information
• Triggers for updating
• - New study(s) to be included
• a) list of new studies from Brenda Thomas
• b) You identify from journal, conference,
  trialists
• - Comment submitted by reader
• Process can be very quick and easy

22
Cochrane other opportunities

• Get involved in Cochrane organisation
• Spend a sabbatical in Edinburgh to do review
• Attend local Cochrane Centre meeting
• Attend Cochrane Colloquium
• Join Cochrane steering group (get voted on)