During the coronavirus crisis librarians continue to offer SVHM staff the following services:
Perfect your search in Ovid Medline first
Some databases, such as Ovid MEDLINE(R) ALL, include many thousands of non-indexed records and pre-print publications. These new records flood in from electronic publications and there is a time lag before they can be fully indexed or described. The records received from publishers are basic and don't (yet) include subject headings. Of course these non-indexed records include the latest research, so we don't want to lose them.
There are the only two limits we recommend:
Don't 'limit to humans'
If you select 'limit to humans' you risk losing all non-indexed and pre-print publications in a database which have not been indexed as 'human'. This could be thousands of records.
Instead, remove the animal studies
A safer alternative is to search for the things we don't want (in this case animal studies) and remove them from the results. Adjust the line numbers below as appropriate.
Ovid Medline
6. [last line of search - your line number will be different]
7. exp animals/ not humans/
8. 6 not 7
Ovid Embase and Ovid Emcare
6. [last line of search - your line number will be different]
7. exp animal/ not human/
8. 6 not 7
APA PsycINFO (Ovid)
6. [last line of search - your line number will be different]
7. (animal not human).po.
8. 6 not 7
CINAHL (EBSCOhost)
S6. [last line of search - your line number will be different]
S7. (MH "Animals+") NOT (MH "Human")
S8. S6 NOT S7
There are times when you will need to record and share your search methodology, especially in the context of study or writing for publication. In addition to saving your search strategy in the database itself so you can run it again later, we strongly suggest keeping records in a Word document as you go.
Records to keep for a systematic review or similar publication
It takes very little effort to record this information at the time, but it is impossible to recreate it later.
This is how a record might look for our sample search:
Ovid MEDLINE(R) ALL 1946 to August 28, 2020
Last searched 1 September, 2020 (1200 results)
1. exp *Dementia/
2. (dementia or alzheimer*).ti,ab.
3. 1 or 2
4. *Aggression/
5. *Violence/
6. (aggress* or violen*).ti,ab.
7. 4 or 5 or 6
8. 3 and 7
9. limit 8 to (english language and yr="2010 -Current")
This search strategy is easily replicated by another person. Note the inclusion of:
If you are searching multiple databases for a review you will need to keep these records for every database you search. Strategies will differ from one database to another due to different subject headings and syntax/commands.
The following diagram shows the type of content, and a rough idea of the overlap, of key databases. It demonstrates the need for multiple databases to be included in a comprehensive literature search, but it does not attempt to be exact or exhaustive. Databases change constantly and there are many layers of duplication.
Choosing which database to search
Keep in mind that the specialty area of a database is very broad, and it indicates the types of journals indexed rather than the specific content. For example a nursing and allied health database will still include articles on general medicine, psychology and radiology which are published in nursing and allied health journals. Likewise, there are many nursing related topics in Medline and PsycINFO. Medline and Embase cover an extremely wide range of topics.
For most health related systematic reviews we would include:
If you just need to dip in and find a handful of articles in one database (say for a presentation), then we suggest Ovid Medline.
Don't confuse the name of the database with the years of coverage
Database content constantly changes and records are added retrospectively so years of coverage are fluid.
It is important to understand that the years indicated in a database name – such as Ovid MEDLINE(R) ALL <1946 to present> often refer to the date the catalogue started (1946 in this case) and not to the date of the earliest publications indexed in the database, which in Medline go back to the late 1700s.
For many bibliographic databases it is not possible to accurately define the years of coverage. That is why we keep these records for systematic reviews:
Helen Wilding, Senior Research Librarian, St Vincent's Hospital Melbourne, 2020
Other Ovid databases - Embase, Emcare, APA PsycINFO, Ovid Nursing
CINAHL (EBSCOhost)
Cochrane Library
Informit health databases
Carl de Gruchy Library, St Vincent's Hospital Melbourne
+613 9231 2541 Helen.Wilding@svha.org.au
Literature Searching, Systematic Reviews, Mental Health liaison
Thursdays, Fridays & alternate Wednesdays
Helen's profile | Researchgate | Orcid