Epidemiology

Embase is a database that includes the MEDLINE database, which is the core of PubMed, but also contains more international journals & indexes pharmaceuticals more precisely than PubMed.  It's a good database to use in conjunction with PubMed, especially in projects where you need to search more comprehensively.  If you search only one database, you'll miss at least 40% of the available literature.

Remember that when you use a different database, you may need to use other search terms.  Articles in Embase are indexed by Emtree terms that can differ from MeSH terms.

We recommend that you begin to build a search in Embase that is as similar as possible to the one you ran in PubMed, using the Advanced page. You can just type in the search that you did in PubMed, but there's a better way, that takes only a bit more time & gives you better results

As you begin to type  one search term at a time, a list of words appears below the search box. This shows you the list of possible indexing terms that are used in Embase.  Here you can see that dietary intake is an index term in Embase. When the index (or Emtree) term appears, click on it to select the term. When a term/phrase is an index term, Embase adds single quotes around it.

Here you can see that both phrases that we used in PubMed are index terms in Embase.

Note also that at Embase mapping options, you should uncheck the boxes (by default, all options are checked) & then select options 1 & 3.  This helps to make your search more similar to the search in PubMed.

In Embase, it's useful to search one concept at a time. You can also choose to search for the indexing terms for each concept, then the keywords, then combine the searches together on the Results page.

In this example, we've searched for the 2nd concept, using the indexing term "dairy product" &, separately (with all mapping boxes unchecked), keywords for specific dairy products. To combine the 2 parts of the "dairy products" concept, select the 2 searches (here #2 & #3), change the Using option from AND to OR, & click the COMBINE button to put all of the terms together with OR.

Then combine the 2 concepts (#1 & #4) to create the complete search.

The re are a large number of results & they're not as focused on the topic as you would want. How can you revise the search?

• Add age filter
• Add "United States" to the search

Both of these techniques help, but you still have more than 1500 results. Next, look up an on-target article from your PubMed search in Embase & see how it's indexed. One very helpful feature of Embase is that you can see the indexing terms right on the search results page by clicking the arrow to the left of Index Terms.

You can see that this article only uses the phrase "dietary intake".  Check some other relevant articles in your search results & you'll see that most use "dietary intake" but not "food intake".

Try searching with only "dietary intake" in the search, & you'll find that you have a larger number of relevant articles in your results.

Filters will help focus the search further.  In Embase, there are filters at the top of the page & another set (some of which are the same) on the left. Those filters at the top of the page are on each Advanced search page & are standard. Those on the left of the search results are specific to your set of results. So, if you created a search & wanted to limit it to Adolescents, but there were no results for adolescents in your search results, you would still see that age filter at the top of the page, but you wouldn't see that option in the age groups on the left.

At the top, I'll use the Quick Limits tab to add Humans and English language.  Apply the filters by clicking the Search button above.

From the left, I'll select Adolescents from the Age category & choose the years 2020, 2021, 2022, & 2023 from Publication Date.  Click the Apply button at the top or bottom of the filter list to apply these filters.

A useful tool, cited reference searching, lets you look into the future, at articles that have referenced an article in your list of results.  In Embase, you'll see a Cited by:[#] notation after many articles, which tells you that this article has been cited by that number of articles in a particular database.

Click the link to be taken to the list, provided by the Scopus database.