The Cochrane Stroke Group Editorial Training Day

1
The Cochrane Stroke GroupEditorial Training Day
Identifying trials and reporting search
strategies

• Brenda Thomas

2
Plan for this session

• Update on trials identification
• Current Stroke Group practice
• Cochrane policy
• Main questions for Editors
• Define policy for acceptance of search strategy
  in final version of review

3
The quality and reporting of the search strategy
is important
We are judged by the Cochrane Quality Advisory
Group on the quality of the published search
strategies
4
Specialised Trials Register(RefTraK)

• Major resource for reviewers
• Considerable time and resources committed to
  development and maintenance

5
Bibliographic databases
GENERAL
SPECIALISED

• MEDLINE
• EMBASE
• BIOSIS
• Derwent Drug File
• SCISEARCH

• AMED (Allied and Complementary Medicine)
• CINAHL (Nursing and Allied Health)
• PsycINFO (Psychology)
• Digital Dissertations
• MANTIS (Manual, alternative and natural therapy)
• International Pharmaceutical Abstracts
• Occupational Therapy Journal of Research Index

(Complete to December 2002)
Plus some even more obscure ones!
6
Other electronic resources

• Trials registers
• bibliographic and ongoing
• Research registers
• NRR (UK), NIH (USA)
• Websites
• press releases
• drug company
• research centres
• Electronic journals

7
Handsearching
24 volunteers have contributed to the
handsearching of

• 50 specialist Journals (6 languages)
• 83 Books
• 64 Conference proceedings

8
Personal contact

• Trial information from colleagues
• protocols
• unpublished and confidential information
• Contact with trialists and drug companies
• unpublished data
• technical reports
• non-English language papers
• difficult to find publications

9
Translation of non-English trial reports

• 46 volunteers
• 15 countries, 17 languages
• 506 relevant trial reports added to Trials
  Register
• 50 Chinese or Japanese
• new on-line translation forms

10
Our searching is now very comprehensive
We hold a paper archive of all the trial reports
11
Stroke Interventions

• PHARMACOLOGY (65)
• Physiotherapy (10)
• Surgery (4)
• Complementary therapies (3)
• Psychological therapy (3)
• Service provision (3)
• Speech therapy (2)
• Nursing
• Occupational Therapy
• Diet
• Radiology/radiotherapy
• Other

10
Implications for reviewer search strategies
12
Growth of Trials Register
Approximately 900 trial reports identified per
year
13
The Stroke Group Specialised Register Today
6846 trial reports 3059 individual trials (793
different journal titles)
68 completed reviews 34 protocols
14
How does the Stroke Group help reviewers to
identify trials?
The ideal situation

• Reviewers contact Brenda at the protocol
  development stage and plans are made for the
  search strategy
• Preliminary search strategy is detailed in the
  protocol
• Searches implemented
• Adjustments can be made once the protocol is
  published and final search strategies documented
  in the review.

Reviewers need varying degrees of help
15
Practical help for reviewers

• Specialised Trials Register - RefTraK
• Developing and running electronic search
  strategies
• Advice on other appropriate trial identification
  strategies (eg specialist databases)
• Handsearching and translation
• Providing paper copies of difficult to find
  trial reports
• Contacting authors for information on trials with
  limited information

Avoid duplication!
16
Stroke Group general strategymodified for each
database
Highly sensitive Cochrane TRIALS strategy
STROKE specific strategy
?
Screened for stroke trials
17
Review-specificintervention-based strategy
Highly sensitive Cochrane TRIALS strategy
STROKE specific strategy
?
?
?
Review specific INTERVENTION-based strategy
Screened for additional relevant trials
RefTraK
18

• 1 exp cerebrovascular disorders/
• 2 (stroke or poststroke or cva).tw.
• 3 (cerebrovascular or cerebral vascular).tw.
• 4 (cerebral or cerebellar or brain or
  vertebrobasilar).tw.
• 5 (infarct or isch?emi or thrombo or apoplexy
  or emboli).tw.
• 6 4 and 5
• 7 (cerebral or intracerebral or intracranial or
  parenchymal).tw.
• 8 (brain or intraventricular or brain or
  cerebellar).tw.
• 9 (infratentorial or supratentorial).tw.
• 10 7 or 8 or 9
• 11 (haemorrhage or hemorrhage or haematoma or
  hematoma).tw.
• 12 (bleeding or aneurysm).tw.
• 13 11 or 12
• 14 10 and 13
• 15 1 or 2 or 3 or 6 or14
• 16 acupuncture/
• 17 exp acupuncture therapy/
• 18 electroacupuncture/
• 19 meridians/

Intervention-specific search strategy for MEDLINE
(Ovid) Should be copied and pasted into the
review to reduce error
19
The Trials Search Co-ordinator (TSC) and the
review process

• Guidance on writing search strategy section of
  protocol and review
• Editorial comment by TSC as part of review
  process
• Links with other review groups preparing
  stroke-related reviews

Currently TSC does not see revised submission
20
Cochrane guidelines

• Reviewers handbook
• lists resources for locating studies
• specific guidelines for documenting search
  strategy
• preferred format
• Style Guide
• consistency of database names etc
• Quality Advisory Group
• monitors reporting of search strategies
• aims to improve the quality of Cochrane reviews

21
The search strategy should be described in
sufficient detail in a review that the process
could be replicated - Cochrane Handbook

• How and when review group trials register was
  searched
• Main sources of trials
• Search dates and temporal constraints
• Language and other constraints
• Listing of each electronic search strategy

22
Workshop exampleProtocol Oral care
interventions for individuals post-stroke

• Protocol submitted no previous contact with
  Brenda
• proposed search strategy inappropriate
• Editorial comment
• new search strategy developed and recommended
• Revised protocol submitted
• lead editor checks comments have been responded
  to
• finds search strategy modified by reviewer
• further revision required

23
Other common problems

• Inconsistency between abstract and review
• Errors in search dates
• Inappropriate lines/terms added to search
  strategies
• Miss-spellings, although search strategies should
  be downloaded from databases
• Delay between searches and final review
  submission
• review out of date

24
What are the implications for Editors?

• Check comments from TSC have been addressed.
• Be aware of Collaboration guidelines (Handbook
  and Style Guide) and Quality Advisory Group
  recommendations
• new Cochrane guidelines for reporting search
  strategy awaited
• We need to be realistic
• technical difficulties
• limited resources
• focus on major searching issues

Contact Brenda with any specific queries
25
Discussion topicsWe need to answer the following
questions and decide on policy

• Do we need to change the Editorial process?
• How do we define mandatory vs non-mandatory
  comments?
• How far should reviewers comply with searching
  advice and Cochrane guidelines?
• Should we approve reviews for publication
• if search comments are not responded to?
• with out of date searches (gt1-2 years old)?