SEARCHING FOR PREPARING SYSTEMATIC REVIEWS

1
SEARCHING FOR PREPARINGSYSTEMATIC REVIEWS

• Tehran university students scientific research
  center
• (TUSSRC)

Kazem Heidari
2
How to find studies?

• Very quick search on one electronic database and
  find a couple of relevant articles.
• Try to find every study thats ever been done
  addressing your reviews question.

3

• Problems
• Few studies
• Publication bias studies with dramatic results
  are much easier to find
• Less information
• Limit percision
• Restrict conclusion

4

• Problems
• Finding every study
• Not easy and might not be possible
• Many studies are never published
• Published , but not indexed
• Too much effort is needed

5
One of the major ongoing efforts of the Cochrane
Collaboration is to make the reports of relevant
studies easier to find.
6
Systematic reviews generally focus on reports
from randomized controlled trials (RCTs), when
such data are available because of the most
reliable estimates of effects.
7

• Search should
• Be sensitive
• look in a number of different places - not single
• Minimize bias
• Think about finding studies that arent in the
  major sources like MEDLINE.
• Be efficient
• start looking in the place you expect to have the
  highest yield.

8

• A comprehensive search for relevant RCTs, which
  seeks to minimize bias
• One of the essential steps in doing a systematic
  review.
• One of the factors that distinguishes a
  systematic review from a traditional review.

9
Where to search?

• Studies have shown that only 30 - 80 of all
  known published RCTs were identifiable using
  MEDLINE (depending on the area or specific
  question)Dickersin 1994
• So MEDLINE, is generally not considered adequate.

10

• comprehensive search is necessary because of
• Ensuring that as many studies as possible are
  identified
• Minimize selection bias for those that are found.

11
Bias

• The majority of the journals indexed in MEDLINE
  are published in English.
• The results of many studies are never published,
  and most of these probably remain unknown.
• Any summary of only the published reports may
  result in an overestimate of effectiveness due to
  a publication bias

12
SOURCES AND APPROACHES
13
Sources for searching

• Electronic databases
• Handsearching
• Checking reference lists
• Identifying unpublished studies
• Personal communication

14
1.Electronic databases

• CRG specialised register
• Cochrane library
• Other databases

15
CRG specialised registers

• Each Cochrane review group is building up a
  register of studies relevant to its scope.
• It would be more efficient to search these
  sources centrally for all trials relevant

16
The Cochrane library

• Primary source for clinical effectiveness
  information.
• Contains 4 databases several other useful
  sources
• Cochrane Database of Systematic Reviews (CDSR)
• Database of Abstracts of Reviews of Effects
  (DARE)
• Cochrane Controlled Trials Registry (CCTR)
• Cochrane Methodology Reviews (CRMD)

17
(No Transcript)
18
(No Transcript)
19
CENTRAL

• Contained just over 350,000 citations (Jan 2003)
  Includes citations to reports of controlled
  trials that might not indexed in MEDLINE, EMBASE
  or other bibliographic databases
• published in many languages
• citations that are available only in conference
  proceedings or other sources that are difficult
  to access

20
Cochrane controlled tials registry (CCTR)

• Ideally ,it should be a central place to put all
  the reports of controlled trials identified
  through the work of the Cochrane Collaboration.
• It contains the results of searching MEDLINE,
  EMBASE, some other databases and a long list of
  journals, books and conference proceedings.
• Its contents is updated each year

21

• The Cochrane Collaboration has been developing an
  electronic database of reports of controlled
  trials ("CENTRAL") that is now the best single
  source of information about records that relate
  to studies, which might be eligible for inclusion
  in Cochrane Reviews.

22
CENTRAL

• The US Cochrane Center and the UK Cochrane Centre
  have searched MEDLINE for publication years
  1966-2000 using Cochrane highly sensitive search
  strategy
• Updated each year
• Include citations (RCT CCT) if they meet the
  criteria.

23
CENTRAL

• Also the UK Cochrane Centre is retrieving records
  from EMBASE
• checking their titles and abstracts
• submitting these for inclusion in CENTRAL when
  appropriate.
• First run in 1999(include citations between
  1974-99)

24
CENTRAL

• Other general healthcare databases published in
  Australia, China, and Brazil are undergoing
  similar systematic searches to identify reports
  of trials for CENTRAL
• Each Collaborative Review Group (CRG) is
  responsible for the development of a subject
  specific specialized register of trials, which
  serves to ensure that individual reviewers within
  the CRG have easy and reliable access to the
  maximum possible number of studies relevant to
  their review topic
• Using similar methods
• The registers should, in turn, be submitted for
  inclusion in CENTRAL. Thus, records included in
  the specialized register of one CRG become
  accessible to all other CRGs through CENTRAL

25
(No Transcript)
26
(No Transcript)
27
Cochrane Database of Systematic Reviews (CDSR)

• Full-text of systematic reviews undertaken by
  Cochrane Collaboration
• Protocols
• Updated reviews

28
(No Transcript)
29
Database of abstracts of reviews of effects (DARE)

• Provided by CRD (Centre for Reviews and
  Dissemination)
• Database of quality assessed reviews
• Has structured abstracts of systematic reviews

30
http//nhscrd.york.ac.uk/darehp.htm
31
Other databases

• e.g Medline and EMBASE
• Some other bibliographic and fulltext databases
  that are available
• Needs an information specialist who is familiar
  with the searches needed for systematic reviews

32
Medline

• By US National Library of Medicine
• Type of Database Bibliographic
• Over 11,000,000 citations of both clinical and
  preclinical studies.
• complementary database known as PreMEDLINE
  includes citations and abstracts for studies that
  have been published recently but not yet indexed.
• Subject Coverage All Specialties of Medicine
• 60 of References contain Abstracts

33
Medline

• Relatively comprehensive coverage of medical
  journals and because it is readily accessible.
• Free of charge using
• Pubmed
• Internet Grateful Med .(not available yet)
• Commercial vendor
• OVID
• Knowledge Finder
• Silver Platter

34
(No Transcript)
35
(No Transcript)
36
EMBASE

• Elsevier Science Publishers, Netherlands
• Type of Database Bibliographic
• Numbers of Journals 3500
• Bias European in focus
• Thesaurus Uses own thesaurus
• Subject Coverage Strong on Pharmaceutical and
  Clinical Medicine
• Materials Indexed Letters, Editorials, Research
  Articles from 1974 or 1981 onwards (depending on
  method of access)
• Overlap with MEDLINE estimated at between 25 and
  40
• 75 of References contain Abstracts
• Update Updated Weekly and is about 4 weeks behind

37
(No Transcript)
38
Scisearch

• Indexes the references cited in articles as well
  as usual author ,title ,abstract and citation of
  articles
• By web of science (from ISI) that is a citation
  database
• From 4500 major scientific and technical journals
  and the articles that cite them
• Useful for finding follow up work done on a key
  article and for tracking down the address of
  authors.

39
Scisearch

• Types of Searches
• General Search Search on authors, titles,
  keywords, abstracts, sources, institutions, other
  topics.
• Cited Search Search on specific paper or author
  to see who has cited that paper or author in
  journal articles. Cited items can include
  dissertations, books, book chapters, and any
  journal. Weekly Alerting

40
Scisearch

• Can be used to
• identify studies for a review by identifying in
  the database a known relevant source article
• checking each of the articles citing the source
  article, to see if it is also relevant to the
  review
• searching forward in time from the publication
  of an important article
• includes reference lists for records it indexes.

41
(No Transcript)
42
(No Transcript)
43
Print versions of electronic databases

• Index Medicus is the print version of MEDLINE
• Excerpta Medica is the print version of EMBASE
• Science Citation Index is the print version of
  SciSearch

44
Electronic databases

• Generally begins with health-related electronic
  bibliographic databases.

45
Bibliographic databases

• Hundreds of electronic bibliographic databases
  exist.
• Easiest and least time-consuming
• Example MEDLINE and EMBASE (include abstracts
  for the majority of recent records)
• Often a searcher can determine an articles
  relevance to a review based on the abstract
• If the reported study is clearly not eligible for
  inclusion, retrieving the full journal article is
  necessary.

46
Bibliographic databases

• They can be searched electronically
• Title
• Abstract
• Standardized subject related indexing terms
• Example MEDLINE indexing term RANDOMIZED-CONTROLL
  ED-TRIAL (Publication Type) was introduced in
  1991 and allows a user to search for articles
  describing individual randomized trials

47
TYPES OF BIBLIRAPHIC DATABASES

• Cover all areas of health care and index journals
  published from around the world.
• MEDLINE/PubMed
• EMBASE

48

• Index journals published in specific regions of
  the world.
• Australasian Medical Index
• Chinese Biomedical Literature Database
• Latin American Caribbean Health Sciences
  Literature (LILACS)
• Japan Information Centre of Science and
  Technology File on Science, Technology and
  Medicine (JICST-E)

49

• Focus on specific areas of health
• Cumulative Index of Nursing and Allied Health
  (CINAHL)
• AIDSLINE

50
2.HANDSEARCHING

• Definition Handsearching involves a manual
  page-by-page examination of the entire contents
  of a journal issue to identify all eligible
  reports of trials.

51

• whether they appear in
• Articles
• Abstracts
• News
  columns
• Editorials
• Letters
• Conference
  proceeding
• other
  text..

52
Need for handsearching

• Not all trial reports are included on electronic
  bibliographic databases
• For example Conference proceedings are
  important to handsearch because individual
  conference. These abstracts are not included on
  MEDLINE and are not usually included in other
  databases.
• 2. Even when they are included, they may not be
  indexed with terms that allow them to be easily
  identified as trials

53
OTHER REASONS

• Some reports that were published in
  MEDLINE-indexed journals do not have
  corresponding MEDLINE records
• 1- published during a year when the journal was
  not indexed
• 2- report is of a type not generally indexed
  separately by NLM
  (e.g., abstracts from conference proceedings)
• 3- a report or issue may have been omitted by
  mistake (VERY RARELY)

54

• Research has shown that electronic searching for
  controlled trials on some databases retrieves
  only about half the relevant available studies.

55

• Benefits
• Full Identification of all reports (as much as
  possible)
• Reducing bias
• most important Published reports tend to
  show positive effects, but grey literature in
  general has been found to be more likely than
  health care journals to contain negative
  reports (McAuley 2000) PUBLICATION BIAS
• Harms
• Time consuming

56
HANDSEARCHINGinCochrane Collaboration

• For complete identification of published reports,
  there appears to be no alternative to a
  page-by-page search of the literature
• Handsearching within the Cochrane Collaboration
  refers to the planned searching of a journal page
  by page, including editorials, letters, etc., to
  identify all reports of randomised controlled
  trials and controlled clinical trials.

57
HANDSEARCHINGinCochrane Collaboration

• Cochrane Collaboration has encouraged people to
  register their handsearching and look for all
  reports of controlled trials.
• Then collected and put onto the CCTR so that no
  one else has to handsearch that source.
• List of journals being searched can be downloaded
  from the internet.. MASTERLIST

58

• Overall coordination of the Collaborations
  handsearch of the worlds medical literature is
  managed by the US Cochrane Center, which oversees
  on the Master List of Journals being Searched
• Each CRG is responsible for coordinating
  searching activities of specialist health care
  journals and to ensure that all RCTs and CCTs
  within these journals are identified as
  completely as possible and registered

59
MASTERLIST

• Almost 3700 journals have been, or are being,
  searched within the Collaboration, and are
  included in the Master List.
• Conference proceedings being searched are also
  included.
• Managed by US Cochrane Center
• Updated weekly
• Downloaded as an excel file

60

• IN SUMMARY MASTER LIST IS THE MOST IMPORTANT
  RESOURCE FOR HANSEARCHED JOURNALS
• AND
• THE BEST WAY OF AVOIDING THE DUPLICATION OF
  EFFORTS.

61

• Identify a search coordinator.
• Determine which journal(s) to search.
• Register the intent to search a journal with the
  New England Cochrane Center at Providence, using
  the Journal Hand Search Registration Form
• Train and test searchers

62

• Search journal(s)
• Examine each issue, page by page, to identify ALL
  reports of RCTs and CCTs, whether they appear in
  articles, abstracts, news columns, editorials,
  letters or any other text, and regardless of the
  relevance of the subject matter to the searcher.
• Photocopy the first page of each report
  identified, as well as any subsequent pages
  requested by a group's search coordinator.
• Annotate each photocopy, on either the front or
  back, with complete citation information,
  wherever necessary.
• Classify each report as RCT or CCT (or other as
  relevant to the needs of the search group) and
  annotate each photocopy with the classification
  codes. Highlighting (e.g., with a yellow marker)
  or underlining the key terms which justify the
  classification may be required by a group's
  search coordinator.

63

• Forward citations to the New England Cochrane
  Center at Providence
• Whenever such citation records exist, download
  the citation records corresponding with the
  search results directly from MEDLINE
• Transfer (or enter, when there is none to
  transfer) citations into Pro-Cite and add
  classification codes to any specified, unused
  field.

64

• Forward citations to the New England Cochrane
  Center at Providence. Include annotated
  photocopies of any reports without a
  corresponding MEDLINE citation record.

65
3.Checking reference lists

• Following up references from one article to
  another
• Looking for previous reviews of the topic and
  checking their reference lists
• Efficient means of identifying studies for
  possible inclusion in a review
• Reference lists should never be used as a sole
  approach to identifying reports for a review
  (because of publication bias)

66
Checking other reviews

• Existing reviews are some of the most convenient
  and obvious sources of references to potentially
  relevant studies.
• Obtaining copies of previously published reviews
• Checking for references to the original studies.

67
Checking other reviews

• Sources
• CDSR
• DARE (provides information on previously
  published reviews of the effects of healthcare)
• MEDLINE (the most appropriate review articles
  would be indexed under the Publication Type terms
  META-ANALYSIS and REVIEW, ACADEMIC)
• EMBASE
• Enhanced search strategies

68
Print versions of electronic databases

• Searching the earlier printed subject indexes may
  be worthwhile, especially if there is reason to
  believe that there were early studies of the
  intervention being reviewed.

69
4.Identifying unpublished studies

• What Is Grey Literature?
• Literature that is not widely published
• Dissertations
• Theses
• Government reports
• Ongoing studies
• etc

Grey literature
70
Identifying unpublished studies

• Finding out about unpublished studies, and
  including them in a systematic review, may be
  important to minimizing bias.
• There is no easy way to obtain information about
  studies that have been completed but never
  published

71
Identifying unpublished studies

• Sources
• Colleagues
• Web sites
• Informal channels of communication
• Formal letters
• sending a list of relevant articles asking if
  they know of any additional studies (published or
  unpublished) that might be relevant
• send the same letter to other experts or others
  with an interest in the area

72
Identifying unpublished studies

• Asking researchers for information about
  completed but never published studies has not
  typically been fruitful

73
Identifying unpublished studies

• Identifying ongoing studies may also be important
  so that when a review is later updated
• No single, central register of ongoing randomized
  trials currently exists.
• There are hundreds of distinct, predominantly
  online registers that vary widely in content,
  quality, and accessibility
• Various efforts have been made by independent
  groups to begin to provide central access to
  ongoing trials

74

• TrialsCentral
• www.trialscentral.org
• Current Controlled Trials
• www.controlled-trials.com
• clinicaltrials.gov (by NLM)
• www.clinicaltrials.gov

75

• SIGLE
• Gateway.nlm.nih.gov
• www.theses.com
• www.dialog.com
• www.dissertationsandtheses.com

76
TrialsCentral

• provides free and confidential access to listings
  of clinical trials. Information about current
  clinical research helps support informed
  evidence-based decision making in healthcare
• www.trialscentral.org

77
(No Transcript)
78
Current Controlled Trials

• published by BioMed Central
• providing immediate free access to peer reviewed
  biomedical research.
• www.controlled-trials.com

79
(No Transcript)
80
clinicaltrials.gov

• By NLM
• Provides regularly updated information about
  federally and privately supported clinical
  research in human volunteers.
• ClinicalTrials.gov gives you information about a
  trial's purpose, who may participate, locations,
  and phone numbers for more details
• www.clinicaltrials.gov

81
(No Transcript)
82
System for Information on Grey Literature in
Europe(SIGLE)

• Bibliographic database
• Covering European grey literature
• Fields of natural sciences and technology,
  economics, social sciences, and humanities
• Produced by EAGLE ,a consortium of major European
  libraries, of reports
• 1976 to the present
• More than 800,000 citations
• Updated monthly with about 45,000 citations/yr

83
Gateway

• Provided by NLM
• Searches
• MEDLINE/PubMed
• LOCATORplus
• MEDLINEplus
• ClinicalTrials.gov
• DIRLINE,
• Meeting Abstracts
• HSRProj
• Gateway.nlm.nih.gov

84
(No Transcript)
85
(No Transcript)
86
Theses.com

• Comprehensive listing of theses with abstracts
  accepted for higher degrees by universities in
  great Britain and Ireland
• www.theses.com

87
(No Transcript)
88
dialog.com

• providing online-based information services to
  organizations seeking competitive advantages in
  such fields as business, science, engineering,
  finance and law.
• Retrieve data from more than 1.4 billion unique
  records of key information
• www.dialog.com

89
(No Transcript)
90
(No Transcript)
91
dissertationsandtheses.com

• Searchable database of research papers
• Lists more than 25,000 quality research papers,
  experimental studies, and even examples of entire
  theses
• www.dissertationsandtheses.com

92
(No Transcript)
93
5.Personal communication

• To know of studies you havent yet found
• Reviewers send a list of the studies they have
  found to the authors of those studies, asking if
  they are aware of any other relevant studies.
• Or write to the manufacturers of relevant drugs
  or devices and ask if they are aware of any other
  studies.

94
Evidence on adverse effects

• sources
• reports from trials or other studies included in
  the systematic review (including excluded
  reports)
• Electronic sources
• Current Problems (by the UK Medicines Control
  Agency)
• http//www.open.gov.uk/mca
• MedWatch (by FDA)
• Australian Adverse Drug Reactions Bulletin
• http//www.health.gov.au/
• Other regulatory authorities and the drug
  manufacturer
• Other types of studies (e.g. cohort and
  case-control studies,uncontrolled trials, case
  series and case reports) that must be interpreted
  with caution.

95
Search Strategy
96
Developing a search strategy

• For many CRGs, the specialized register is still
  the best available source of studies for a given
  review
• Many Trials Search /Review Group Coordinators
  search their CRGs specialized register for
  reviewers on request
• Specialized registers can also be searched
  through CENTRAL, which contains a recent version
  of the registers for most CRGs

97
Developing a search strategy

• The ultimate goal in developing a specialized
  register for a CRG is that it can serve as an
  all-inclusive source of reports relevant to the
  CRGs scope and topic area such that a relatively
  simple search using some key words related to the
  intervention could be run against the specialized
  register to identify all relevant studies.

98
Developing a search strategy

• Most CRG specialized registers have not yet
  reached this point of comprehensiveness
• For many CRGs, the specialized register is still
  the best available source of studies for a given
  review
• Many Trials Search /Review Group Coordinators
  search their CRGs specialized register for
  reviewers on request
• Specialized registers can also be searched
  through CENTRAL, which contains a recent version
  of the registers for most CRGs

99
Developing a search strategy

• Balance between comprehensiveness and precision
  when developing a search strategy is necessary.
• Increasing the comprehensiveness of a search
  entails reducing its precision and retrieving
  more non-relevant articles

100
Developing a search strategy

• Developing a search strategy is an iterative
  process in which terms that are used are
  modified, based on what has already been
  retrieved.
• There comes a point where the rewards of further
  searching may not be worth the effort required to
  identify the additional references

101
Developing a search strategy

• The decision as to how much to invest in the
  search process depends on the question a review
  addresses, the extent to which the CRG's
  specialized register is developed, and the
  resources that are available

102
Developing a search strategy

• Search other electronic bibliographic databases
• Information specialists
• To avoid many errors
• Ensure that database-specific search term syntax
  will be appropriate.
• Advanced searching techniques can be employed
  where available.
• Aware of greater importance of high recall (i.e.
  sensitivity) as compared to precision in
  searching for studies for systematic reviews
• Ideally,reviewers should be present when the
  search is done
• Judgments about what to download often need to
  be made while the search is being done. (because
  of cost)
• Helpful in suggesting terms for the health
  condition and intervention

103

• Electronic search strategy
• terms to search for the health condition of
  interest
• terms to search for the intervention(s) evaluated
  
• terms to search for the types of study design to
  be included

104

• Approach
• Begin with multiple terms that describe the
  health condition of interest and join these
  together with the Boolean 'OR' operator
• Do likewise for a second set of terms related to
  the intervention(s) and for a third set of terms
  related to the appropriate study design
• Three sets of terms can then be joined together
  with the AND operator.
• CAUTION
• If an article does not contain at least one term
  from each of the three sets, it will not be
  identified.
• No language restrictions should be included in
  the search strategy

105
One problem

• In Medline
• The term randomized controlled trial from 1991
  in publication type
• The term controlled clinical trial from 1995 in
  publication type

106

• The Cochrane highly sensitive search strategy for
  MEDLINE was developed specifically with the needs
  of Cochrane reviews in mind. (first version in
  1994)
• It may still be worthwhile for reviewers to
  search MEDLINE using the Cochrane highly
  sensitive search strategy and to obtain and check
  the full reports of possibly relevant citations.
  CRGs typically use this plus subject matter terms

107
Cochrane high sensitive search strategy for RCTs
108
Documenting a search strategy

• How,what,when
• Help to avoid repeat search
• Help people using your review appraise how well
  they think youre minimized bias
• All details should be documented in the search
  strategy section of review

109
Documenting a search strategy

• Electronic databases
• described in sufficient detail
• included for each electronic bibliographic
  database each time it is searched.

110

• Title of database
• Name of the host
• Date search was run
• Years covered by the search
• Complete search strategy used
• Summary of the search strategy(health condition
  ,intervention and type of study)
• The absence of any language restrictions

111

• Journal Handsearching
• journal years searched should be listed in the
  Search strategy
• Ideally, the full titles should be used for the
  journals
• Example British Journal of Surgery January 1948
  December 1998

112

• Conference Proceedings
• Details of the conference proceedings searched
  for the review should be provided
• Proceedings with a title in addition to the
  conference name
• Proceedings without a separate title
• Proceedings in a language other than English
• Proceedings also published as part of a journal

113

• Unpublished studies
• Brief summary including databases searched
• database details
• Efforts to contact investigators for information
  about unpublished studies.

114
Selecting studies

• The process by which studies will be selected for
  inclusion in a review should be described in the
  review protocol.
• The first stage of checking the results of an
  electronic search involves assessing titles and
  abstracts
• It is important to err on the side of
  over-inclusion because once a study has been
  excluded from the selection process it is
  unlikely to be reconsidered

115

• given the information available, it can be
  determined that an article definitely does not
  meet inclusion criteria, it can be rejected. If
  the title or abstract leave room for doubt that
  the article cannot definitely be rejected, the
  full text of the article should be obtained.
  Reading the full text may lead the reviewers to
  exclude the study because it does not meet
  inclusion criteria.

116

• Assessing for inclusion criteria
• More than one reviewer assess the relevance of
  each report
• Decisions concerning relevance will be made by
  content area experts, non-experts, or both
• people assessing the relevance of studies know
  the names of the authors, institutions, journal
  of publication and results when they apply the
  inclusion criteria
• How disagreements will be handled between 2
  reviewrs
• Consensus
• Third person
• categorize as one that is awaiting assessment

117

• 2 reviewers is recommended
• One reviewer is knowledgeable in the area under
  review
• Second reviewer who is not an expert in the area
• Some reviewers may decide that assessments of
  relevance should be made by people who are blind
  or masked to the journal from which the article
  comes, the authors, the institution, and the
  magnitude and direction of the results by editing
  copies of the articles (takes much time)

118

• Excluded studies should be listed
• the reviewer can show that consideration has been
  given to these studies.

119
Keeping track of identified studies

• You may find several reports of the same study
• You may find the same report of a study in
  several databases.

120
Keeping track of identified studies

• So we need
• Some way of keeping track of the references weve
  looked at.
• Some way of grouping together all the reports of
  a single study.
• Also we might like to keep a record of where you
  found each study ,so that you can report how
  useful different sources were.

121
Keeping track of identified studies

• Some people use reference management software to
  do all this
• ProCite
• Reference Manager
• EndNote
• IdeaList
• Other people prefer printing out citations and
  writing on them.

122
Keeping track of identified studies

• We need some system for keeping track of which
  references you think are relevant
• Which ones you have ordered from the library
• which ones youve received the paper for
• etc

123

• ProCite
• the most widely used package and the one for
  which support to editorial bases is most widely
  available.
• the preferred database for submitting controlled
  trials and specialized registers to CENTRAL
• eases the work of identifying duplicate
  references
• facilitates storage of information about the
  methods and process of a search
• Reference Manager
• EndNote

124

• General database packages
• Access
• FoxPro
• MeerKat
• developed by the UK Cochrane Centre
• allows for a specialized register to be organized
  around studies, instead of the publications or
  reports generated from these studies
• each report can be associated with the
  corresponding study
• Facilitating the work of the Review Group
  Coordinator/Trials Search Coordinator
• allows complex database searches
• Ease the task of managing references within a
  CRG.

125
Time saving for searching

• Suggestions
• Look at the terms used to index and describe a
  few studies you already know are relevant to your
  review, and use these terms in your search
  strategy
• Add new terms to your search strategy and then
  pilot them on part of the database
• Use date limits for your search if appropriate.

126
Thank you!