Can Scopus replace Embase for Systematic Reviews?
It is a hard pass; however, there are ways to bend the rules.
An intro to Embase and Scopus
For those who are not familiar with these two databases, you should know their publisher is Elsevier. Embase and MEDLINE are the most important medical literature indexing and abstracting database. However, Embase covers lots of journals in European languages and recently many other languages and compensates for shortcomings of US-biased and English-biased MEDLINE. On the contrary to MeSH used by MEDLINE, Embase uses Emtree at its controlled vocabulary. MEDLINE is accessible for free via PubMed, Embase has no free access. Like MEDLINE, it is possible to access Embase via several interfaces, but Embase via Ovid SP and Embase.com are the most common ones.
Scopus is relatively new and launched in 2004 — the same year I entered the University — covering all sciences. In terms of the number of records, it became and maybe still is the largest bibliographic database. For indexing literature from too many sciences, it is probably a dilemma to use a controlled vocabulary. Initially, the first impression was that Scopus and Google Scholar were merged to compete with Citation Indexes produced by the Institute for Scientific Information (ISI).
Reasons why Scopus cannot replace Embase for Systematic Reviews
- Scopus is a multidisciplinary source, and it is not easy to limit the search to only a proportion of its coverage. Although it provides a limitation based on sciences, it is based on the journal subject categories and is not accurate.
- Scopus does not provide Emtree indexing (controlled vocabularies), which is one of the requirements to have a better retrieval experience. Emtree terms are also based on full-text indexing and synonym mapping, which has no meaning in Scopus. Having controlled vocabularies in Embase search strategies is required by PRISMA-Search and PRESS (peer-review of electronic search strategies) guidelines. Emtree has become so good at indexing synonyms of drugs and other concepts that some information scientists, including Wichor Bramer, draft their first search strategy in Embase and then translate it to the other sources. This is partly because Emtree used -semi-automated indexing for a few years and MeSH only recently announced such development. Automation has caused some troubles, so the information scientist Julie Glanville et al. produced a guide on Pruning Emtree.
- Scopus does not support long or very long search strategies, which is a requirement for many systematic reviews. The information professionals end up cutting corners to fit something that Scopus accepts rather than a proper search strategy.
- Coverage overlap between Embase and Scopus is unclear. Elsevier, the mother of both Embase and Scopus, never states clearly if Scopus covers all Embase records. Understandably, this serves the marketing strategy through confusion. Imagine if the answer was yes, then a library with limited sources would subscribe to Scopus and not Embase.
- Scopus does not provide the helpful search features that the Embase.com interface does. Pharmacovigilance (PV Wizard) search (including default drug names and adverse effects search filters), PICO search, Medica Device (including default adverse effects search filter), and Drug/Devices searches. The embedded adverse effects search filters are based on the work done by the information scientist Su Golder et al.
- Translation of search strategies and searching MEDLINE and Embase via Ovid SP interface is much easier and requires less effort. Scopus requires quite a different syntax for search.
- Embase has started indexing preprints. Scopus and MEDLINE will wake up one day!
Is there any way to search Scopus instead of Embase? or Ways to access Embase!
No. Not only that there is no way, but there is also no evidence that one can replace the other in a systematic review context.
Now, if you are from an institute or a country that cannot afford to get a subscription to Embase, you would sing P!NK’s ‘what about us?’ Yes, I know. I have been there, and I found a way to access Embase via my colleagues who had access, via TaskExchange.Cochrane.org, via requesting Embase directly, via global initiatives that provide access to low and middle-income countries via requests in library mailing lists and so on.
If I believe that you have tried all of the above and none have worked, I would say it is acceptable as a limitation to your systematic review to search Scopus instead of Embase when there is no way to access Embase.
When writing the protocol for your systematic review, consider the accessibility to the main sources and predict the challenges you may face and how you would deal with them. Try all the ways to access Embase before you shift to Scopus.
Conflict of interests and acknowledgements
Between 2007–2010, I have received free UMLS DVDs from the US National Library of Medicine, the producer of MEDLINE.
Between 2004 and 2018, I have helped journals and publishers to index their papers in bibliographic databases, including MEDLINE, Embase, Scopus, and Science Citation Index. This role has been partly paid and partly voluntarily.
In December 2019, Elsevier Embase paid for my flights between Nottingham and Amsterdam to speak in Embase Strategy Meeting, in which I requested Embase to index more sources, including preprints.
Between 2018 and 2020, I have delivered free training sessions via Embase channel on BrightTalk.
Elsevier Embase has provided me with complimentary access to Embase.com between 2018 and 2022.
Please Follow and Sign Up for this blog for similar readings :D