PubMed – Transcript

[website: Web of Science database page https://library.carleton.ca/find/databases/pubmed]

PubMed is a massive database with over 35 million citations focused on biomedical literature and life sciences.

It is curated by the National Center for biotechnology information under the National Institutes of Health in the United States.

Here we are at the landing page for PubMed.

We can see a link to PubMed and PubMed Central in the middle of the screen under the word ‘Connect’.

PubMed Central contains freely available full text resources, but we're going to stick with PubMed as not only will it include content from PubMed Central, but it will also connect with our subscription databases at Carleton, providing you with full-text access to many more results.

So let’s click on PubMed. If you simply did a web search for PubMed to begin your session, you would not be shown links back to Carleton resources.

I'm going to start off by talking about what are called MeSH terms.

MeSH stands for medical subject headings, what are known as controlled vocabulary.

Controlled vocabulary is what a database uses to logically index its contents.

Searching with MeSH terms is typically more precise than a simple keyword search. A keyword search will simply look for your keywords in an article, but doesn’t consider whether or not that article is relevant to you. MeSH terms are applied to an article because they are specifically relevant to it.

Scrolling down, I'm going to click on the MeSH Database toward the bottom right, under the heading ‘Explore’.

This will bring us to where we can search for relevant MeSH terms.

Our broader research interest is to learn about the impact of vacuum packaging on the quality indicators of fresh pork over time, excluding results discussing the process of modified air packaging.

As an example, let’s begin with the term pork, which I’ll type into the search box and click search to the right of the search box.

A choice of ten items appear, but ‘Pork Meat’ at the top of the list seems most relevant, so I will click on ‘Pork Meat’.

We are presented with a brief definition followed by subheadings we may wish to focus in on. We are able to tick more than one subheading, as appropriate.

Further down we can see ‘Entry Terms’ for a list of terms related to ‘Pork Meat’.

Below ‘Entry Terms’ is a hyperlinked hierarchical structure of how ‘Pork Meat’ relates to the broader category of ‘Diet, Food, and Nutrition’ toward the top of the first structure labelled “Phenomena and Processes Category”, along with each of the intermediary categories.

The second structure stems from “Technology and Food and Beverages Category”, which is another branch of all MeSH categories.

And so this screen, which we got to by simply searching for the word pork, is providing us with a lot of depth and context for how we might effectively discover more research surrounding this topic.

For now, I’m going to scroll back up to the top of the screen and simply add ‘Pork Meat’ to the ‘PubMed Search Builder’ at the top right by clicking on ‘Add to search builder’.

Next I’ll click the button below, labelled ‘Search PubMed’.

The results screen that comes up looks just like any other database, displaying our 736 results as shown at the top of the list, with filters on the left for refining our results.

In particular, under the filter titled ‘Article Type’ on the left-hand side, we can see a number of specific formats, which look a little more interesting that we saw with Web of Science.

If we scroll down to the bottom of the filter list and I click on ‘Additional Filters’, we can see a much more extensive list of attributes we can add to our available filters, including all sorts of other ‘Article types’.

For now, I’ll click ‘Cancel’ at the bottom of this list to return to the results screen.

Scrolling back up to the top, I want to bring us to the ‘Advanced’ search screen by clicking ‘Advanced’ under the search box. Here we can develop a more sophisticated search strategy, including using traditional search statements combined using Boolean Operators.

Lower down on this screen we can see a history of our searches today under the heading ‘History and Search Details’.

This is a record of all the searches we have done so far, including the number of results.

The only search listed so far is the MeSH term search for ‘Pork Meat’, but as we conduct more and more searches, a record of our search history will be preserved. This allows us to review and revisit our searches, as well as combine multiple searches together.

Under the ‘Actions’ column of our history, we can click the three dots which will give us the ability to ‘Add query’, which I’ll click.

This adds this search to the ‘Query box’ above. The Query box collects all of our search criteria in one spot.

Again, my broader interested has to do with vacuum packaging and quality indicators.

It may be possible to seek out relevant MeSH terms for other concepts we might be interested in, but instead, I want to demonstrate mixing MeSH terms with keywords, as another strategy you might try.

By default, our keyword search will apply to ‘All Fields’, as you can see next to the text field at the top.

I’m going to click the arrows, and among the many choices presented, land on limiting our keyword search to only the ‘Title/Abstract’ field. The idea is, if our keywords appear somewhere in the title or abstract of a result, it will likely be of more interest to us.

So let’s follow the same process we did with Web of Science, and type in our first concept as:

vacuum OR “vacuum pack*”

I’m going to click AND, which will add these keywords to our Query box below that already includes our MeSH term for ‘Pork Meat’. Next, let’s type in:

quality OR fresh* OR shelf-life

Again, clicking AND

For our final concept, excluding mention of ‘modified air packaging’, let’s type in:

map OR “modified air packaging”

But then click the drop-down arrow next to AND and select the option for ‘Add with NOT’, which I’ll go ahead and click.

Now looking toward our ‘Query box’ below, we can see our search statement spelled out in full. Let’s click ‘Search’ next to our ‘Query box’, which gives us only 13 results.

This may seem like a low number, but because we are using MeSH terms, or controlled vocabulary, theoretically our results should be more precise.

Similar to other databases, we can further refine our results by limiting to publication date and ‘Article Type’ on the left-hand side, but for now I’m simply going to click on the title of the third result listed.

We find that this article was published in 2019 in Meat Science. Scrolling down we see our list of authors, and we can view their affiliations by clicking ‘+ expand’ beneath the list of their names.

We can also see the red ‘Get it!’ button at the top right to link out to the full text.

Further down the screen we find the abstract, keywords applied to this paper, which are followed by ‘Similar articles’, meaning, if we liked this one, we may also be interested in these others. And beneath that we see ‘Cited by’ 3 other articles, with hyperlinks to each, for us to see how this paper has influenced further research.

If we continue scrolling down we see a number of associated MeSH terms for this article, including three references to ‘Pork Meat’.

Returning to the top of the screen, let’s say I’m interested in exporting this article to EndNote. I’m going to click on ‘Send to’ and choose ‘Citation manager’ at the bottom of the list. Next, I’ll click ‘Create file’, which will download a file which I will then have to import into EndNote.

You will have noticed I didn’t create an account for PubMed like I did with Web of Science, but you can certainly do so with PubMed as well, which opens up additional features like setting up email alerts for newly published material, saving searches, and more.

And so to close, this has been an introduction to PubMed, including using MeSH terms, or controlled vocabulary, combined with keywords, to give our search even greater precision.

Content last reviewed: