Hello,
I would just add my two cents worth, perhaps obvious but not previously mentioned: in undertaking a systematic review, the range and selection of databases used depends on the subject. For example, if your topic is purely clinical, you will not find many additional citations beyond the core clinical databases (MEDLINE/PubMed, EMBASE, CINAHL), noting that you will almost always find a few additional in interdisciplinary databases such as Scopus and ISI Web of Science and in trials registries. If your topic will be covered in the grey (e.g., with a global health aspect) or complementary and alternative literature, databases covering those segments of the literature (IGOSearch, NGOSearch, AMED) should be included. ISI Web of Science, Scopus and Google Scholar are particularly useful in searching interdisciplinary topics because of the range of disciplines they include in their coverage (although what Google Scholar includes is not public knowledge). Attempts are being made to systematize the selection of databases and determine a point at which to stop adding databases, for example:
Booth A. How much searching is enough? Comprehensive versus optimal retrieval for technology assessments. Int J Technol Assess Health Care. 2010 Oct;26(4):431-5. Epub 2010 Oct 6. Review. PubMed PMID: 20923586.
Kastner M, Straus SE, McKibbon KA, Goldsmith CH. The capture-mark-recapture technique can be used as a stopping rule when searching in systematic reviews. J Clin Epidemiol. 2009 Feb;62(2):149-57. Epub 2008 Aug 22. PubMed PMID: 18722088.
In the search for Ahmed’s systematic review (thank you, Ahmed, for sharing it), the range of databases used and the variations of the strategies between databases
are impressive. The involvement of a professional librarian is apparent, and is commendable (from my biased viewpoint; should there be a professional librarian as co-author on all systematic reviews?). In general, searching ISI Web of Science and Scopus
requires use of exhaustive lists of free text synonyms for title and abstract searching, although, in this case, full advantage has been made of the limited database features (that is, Scopus and ISI Web of Science use little controlled vocabulary to index
and structure the database).
Jim
++++++++++++++++++++++++
Jim Henderson
Librarian, McGill University (Retired)
3546 Marlowe Avenue
Montreal QC H4A 3L7
Home: 514-807-5863
Cell: 514-668-2481
http://cvhl.ca
http://cvhl.ca/fr
http://wikisites.mcgill.ca/GlobalHealthGuide
From: Evidence based health (EBH) [mailto:[log in to unmask]]
On Behalf Of Ahmed Abou-Setta, M.D.
Sent: January-12-12 2:25 PM
To: [log in to unmask]
Subject: Re: Systematic reviews - searching Scopus
Following up on my previous E-mail, the link to our full report that I noted to be is as follows:
http://www.effectivehealthcare.ahrq.gov/hippain.cfm In the appendix B, we have the full search
strategy displayed for the 25 databases searched. The full search was prepared, and adapted for each database, by our research librarian and co-author (Lisa Lisa Tjosvold). Page B-9 contains the search for Scopus and ISI Web of Science. As you can see they
are a lot different than the Medline or Embase search strategies.
Hope this helps.
Ahmed
From: Ahmed M. Abou-Setta, M.D. [mailto:[log in to unmask]]
Sent: January 11, 2012 11:13 PM
To: [log in to unmask]
Subject: RE: Systematic reviews - searching Scopus
Hi Jane,
I’ll try not to repeat what Amy and Farhad said so I will give you some practical, front-line tips. Searching for evidence is not a perfect science
and each database uses a different algorithm for searching for articles. Studies have been done comparing Medline being searched via PubMed and Ovid and there are slight differences in the results produced even though they both are really searching the same
database.
Practical point #1: If you are doing an SR, then I highly recommend you search Medline, EMBASE, CENTRAL (call it the Axis of Medical Knowledge
if you may). You can (and should) supplement these with databases that cover non-English languages (e.g. LILACS), regional databases (e.g. KoreaMed), and content-specific databases (e.g. CINAHL). You should spread your net as far as practically possible.
Practical point #2: Scopus and ISI Web of Knowledge are relatively new players in the area (to the best of my knowledge). They do offer more ‘grey
literature’ searching (e.g. conference abstracts) and the wonderful ability to do forward searching (very beneficial, especially when updating a search). These unfortunately are not a replacement for other literature databases.
Practical point #3: Over the years, I have used or seen from one to over 25 databases being searched for citations. Most common is the big three,
and I highly discourage only one and think that high numbers are great if you have the resources but not particularly necessary.
Practical point #4: Quality over quantity: Search strategies need to be tailored to each database, if you want to get reliable results for an SR.
Scopus strategies are the most complex I have ever seen. If you would like to see how different search strategies look for different databases asking the same PICO I can send you a link to recent review we conducted that contained search strategies for 25
databases.
Hope this helps.
Ahmed
From: Evidence based health (EBH) [mailto:[log in to unmask]]
On Behalf Of Jane Reid
Sent: January-11-12 10:07 PM
To: [log in to unmask]
Subject: Systematic reviews - searching Scopus
Hi!
My questions are related to searching for articles for systematic reviews and which databases to search.
According to the Cochrane handbook, “CENTRAL and MEDLINE
should be searched, as a minimum, together with EMBASE if it is available to either the CRG or the review author.” I also search other databases which are more specific: for example, CINAHL if nursing related,
PsycINFO if the question has a psychological focus, AMED if alternative therapy related.
I have noticed that more authors of systematic reviews are searching Scopus. Scopus does have wide coverage (including 100% MEDLINE coverage)
and I can see the usefulness of Scopus for ‘forward’ searching for articles which have cited relevant articles. However, I’m not convinced of the benefit of searching Scopus in addition to the databases mentioned above.
In order to gain a sense of how databases are being searched, my specific questions are:
1.
Do many SR authors search Scopus in addition to CENTRAL, MEDLINE, Embase, CINAHL/PsycINFO/AMED? What is the advantage of doing this?
2.
Does anyone search Scopus alone?
I would appreciate answers/thoughts/suggestions.
With many thanks and kind regards,
Jane