For more information, see the Cochrane Handbook 4.5 Documenting and reporting the search process
Searches should be reported in enough detail so that they could be reproduced by someone else if necessary. It is often the responsibility of the CIS to check the Search Methods section in protocols and to keep track of the search process so that they can make sure the Search Methods sections in reviews meet this requirement. Other sections of the review should also be checked including; Abstract, Plain Language Summary, Appendices, Discussion, Results of the Search, to ensure that the search methods are reported consistently throughout the review.
CISs are often required to help complete the Study flow diagram (PRISMA diagram) as well. The flow diagram depicts the flow of information through the different phases of a systematic review. It maps out the number of records identified, included and excluded, and the reasons for exclusions. PRISMA stands for Preferred Reporting Items for Systematic Reviews and Meta-Analyses, and is a minimum set of items for reporting systematic reviews and meta-analyses.
Documenting the search process may be done through Review Manager (RevMan). The MECIR standards give guidance on what should be reported in the review, and this is also available in RevMan. A Search Report Form template is available to download and can be adapted as neccessary. The Cochrane Style Manual should also be checked to ensure that the search methods sections of reviews and protocols are written in line with Cochrane house style.
6.5.1 Protocol: Search Methods Section
This section describes what sources are to be searched. It is split into two parts: Electronic Searches and Searching Other Resources.
In the Electronic Searches section, planned searches of any bibliographic databases are recorded. This section should list:
- the source e.g. Medline
- the provider e.g. Ovid
- the date from which the source is intended to be searched (if available).
A full search strategy from at least one database should be added so that peer referees can comment on it. Full search strategies should be provided in the Appendix of the review, and referenced in the protocol.
Specialised Register searching should be listed first, for example
‘We will search Cochrane X Group Specialised Register’, and the preferred order for other sources is CENTRAL, MEDLINE, Embase, other databases.“
In the Searching Other Resources section, the authors should document whether or not they plan to carry out:
- handsearching on subject-specific journals
- searches of conference proceedings
- checking of reference lists
- searches of trials registries, as per MECIR guidance
- contacting experts in the field.
If the authors plan to search journals and conferences, they should give journal names and years to be searched, and specific conference details. If the authors intend to search for literature covering adverse effects or events they should indicate that in this section.
The CIS Support team have worked with CIS colleagues, the copy-edit support team and the MECIR standards and Cochrane Style Manual to create template text for search methods in a protocol. Use of the text is optional and it can be adapted to reflect a Group's local practices. The text is available on the Information Specialists Portal, under the section on Searching: reporting.
6.5.2 Review: Abstract and Plain Language Summary
See Chapter III.3.1 Abstract in the Cochrane Handbook for more details on this topic.
The abstract should be no more than 400 words long, and as the abstract will be published in MEDLINE and Science Citation Index it must be written as a standalone document.
Abstracts should always use the active voice, for example ‘we searched MEDLINE', rather than ‘MEDLINE was searched’.
In the search methods section, the abstract should list the sources searched in the same order as they are listed in the main text e.g. Specialised Register, CENTRAL, MEDLINE, Embase, ClinicalTrials.gov, World Health Organization International Clinical Trials Registry Platform (ICTRP), including the dates of search. Any other searches, e.g. of adverse effects information, should be briefly outlined.
It is possible to use the generic phrase ‘reference lists of articles’ for searching bibliographies.
When reporting contact with pharmaceutical companies, it is preferable to use ‘we contacted pharmaceutical companies’ rather than listing of all the pharmaceutical companies that were contacted.
The Plain Language Summary should also report the date of search, see the Cochrane Standards for the Reporting of Plain Language Summaries.
6.5.3 Review: Search Methods Section
As with the protocol, the search methods section is divided into Electronic Searches and Searching Other Resources.
The text in the Electronic Searches section should be similar to the protocol, but you should give the date each source was searched, and change the tense, i.e. from "We will search" to "We searched".
The full search strategy for each source must be provided. This should be included in an Appendix at the end of the review, and linked to from the Search Methods section. Each search strategy for each source must be given as it was run in the database.
The Searching Other Resources section should be the same as the protocol, but dates of searches should be added as appropriate, and the tense changed as in the Electronic Searches section.
The CIS Support team have worked with CIS colleagues, the copy-edit support team and the MECIR standards and Cochrane Style Manual to create template text for search methods in a review. Use of the text is optional and it can be adapted to reflect a Group's local practices. The text is available on the Information Specialists Portal, under the section on Searching: reporting.
6.5.4 Review: Results of the Search Section, Study Flow Diagrams, Other Sections
CISs may be required to check the ‘Results of the search’ section. Authors should give the total number of hits retrieved by the search in this section, and briefly describe the screening and selection process.
The Study flow diagram illustrates the results of the search and the process of screening and selecting studies for inclusion in the review. There is a template in RevMan which can be modified.
The Cochrane Handbook, Section 11.2.1 gives more information, and describes the importance of making a distinction between a study and reports of a study when creating the flowchart.
The template text for search methods referred to above includes an example of text for the Results of the search section, illustrating how the screening process may be described.
CISs should also check other sections of the review where the search methods may be reported (e.g. the Discussion section). Checking that the references are formatted correctly is often a requirement. CISs should use the Cochrane Style Manual as a guide.