Transcript Abstract Presentation

The Cochrane Stroke Group
Helping reviewers
to identify trials
Brenda Thomas
Alison McInnes
Peter Sandercock
Role of the
Information Specialist
• Plan, implement and co-ordinate a comprehensive
trials identification strategy
• Assess and code relevant trials
• Develop and maintain a specialised register of
randomised trials and controlled clinical trials
• Support reviewers
• Contribute to the editorial process
Considerable time and resources are committed to
maximising trial identification
A systematic approach to
trial identification and management
• Multiple overlapping search strategies
– electronic resources
– handsearching and translation
– personal communication
• Trials register: relational database where the trial is the
unit of data
–
–
–
–
facilitates preparation and updating of reviews
reliably tracks trial information
identifies areas where reviews are needed
monitors review quality: ensures reviews are kept
up-to-date
– Used to influence future trials searching activities
Search Strategies
•
•
•
•
20 electronic bibliographic databases
Handsearch and translation programme
More than 500 relevant electronic journals
Web-based international clinical trials and research registers
(ongoing studies)
• Websites: e.g. press releases, research centres
• Contact with trialists and drug companies (unpublished studies)
Supported by the considerable resources of
The University of Edinburgh and NHS Scotland e-library
Bibliographic databases
GENERAL
• MEDLINE, EMBASE, BIOSIS, Derwent Drug File, SCISEARCH
SPECIALISED
• AMED, CINAHL, PsycINFO, Digital Dissertations, FROSTI,
International Pharmaceutical Abstracts, MANTIS, Occupational
Therapy Journal of Research Index, SPORTDiscus, Wilson Social
Sciences Abstracts
Many thousands of references to be screened
Handsearching and translation
• 26 volunteer handsearchers
• 48 specialist Journals (6 languages)
• 147 books and conference proceedings
• 47 volunteer translators from 19 countries
• 700 non-English trial reports (20 languages)
• 56% Chinese or Japanese
Contribution of volunteers has been crucial
Growth of Trials Register
9000
8000
Cumulative number of trial reports
7000
6000
5000
4000
3000
2000
1000
0
1995
1996
1997
1998
1999
2000
2001
2002
2003
Year of entry to Trials Register
Approximately 1000 trial reports identified per year
2004
The Stroke Group
Trials Register Today
8137 trial reports
3598 individual stroke trials
Found in
900 different journal titles
121 conference proceedings
143 books and dissertations
130 electronic sources
Included in the Cochrane Central Register of Controlled Trials
Stroke Group Reviews
More than 250 International Stroke Group
collaborators based in 22 countries have
contributed to the production of
77 Cochrane Systematic Reviews
Example: new review title
Acupuncture for stroke
rehabilitation
Reviewers based in China
How can the Stroke Group
help identify trials?
• Regular searches of Stroke Group Trials Register
• Developing search strategies and running searches of
bibliographic databases and trials registers
• Advice on other appropriate trial identification
strategies e.g. specialist databases
• Handsearching specialist journals and conference
proceedings
• Translation
Avoid duplication!
Searches of trials registers
• Cochrane Stroke Group Trials Register
• Cochrane Central Register of Controlled Trials
• Cochrane Complementary Field Registry
– http://www.compmed.umm.edu/Cochrane/registry.html
• National Center for Complementary and
Alternative Medicine
• The Centralised Information Service for
Complementary Medicine (CISCOM)
• Chinese Stroke Trials Register
• Chinese Acupuncture Trials Register
REVIEWER
Intervention-based searches of
bibliographic databases
•
•
•
•
MEDLINE
EMBASE
CINAHL
Allied and Complementary Medicine Database
(AMED)
• China Biological Medicine Database (CBM-disc)
Avoid overlap with Stroke Group searches
Stroke Group
general strategy
modified for each database
STROKE
specific
strategy
?
Highly
sensitive
Cochrane
TRIALS
strategy
Screened for stroke trials
Review-specific
intervention-based strategy
STROKE
specific
strategy
?
?
?
Screened for
additional
relevant trials
Highly
sensitive
Cochrane
TRIALS
strategy
Review specific
INTERVENTION-based
strategy
Other help finding trials
• Sourcing paper copies of “difficult to find”
trial reports
• Contacting authors for information on trials
with limited information
• Arranging translation of non-English
language trial reports
The Information Specialist
and the review process
• Advice on trial identification methods
• Guidance on writing search methods
section of protocol and review
• Editorial comment as part of review
process
Updating reviews
• Searches of Stroke Group trials register on
demand
• Update searches of bibliographic databases
• New search sources can be identified
Active support can help motivate reviewers
Reviewers need varying
degrees of help
•
•
•
•
Access to databases
Previous literature searching experience
Availability of local librarians
Problems accessing journals and
conference proceedings
• English language limitations
Summary
• The Stroke Group provides considerable support for
reviewers searching for trials
• The major resource is the Stroke Group Trials
Register which is comprehensive and rapidly
expanding
• The Information Specialist devises additional
complementary strategies to maximise trial
identification
• The quality of the search strategy section of reviews is
monitored as part of the review process