 When searching for articles on a specific topic, it's important to select the most appropriate terms to use. Doing this often requires preliminary legwork in order to develop the best search strategy possible. In a comprehensive literature search where the goal is to identify all relevant articles, it's worth the time and energy spent finding all possible terms that might be used to represent the concept in your research question. But even if you're only looking for a small set of articles, it's still valuable to take some time to identify key search terms so as not to miss important articles on the topic. A common mistake that can cause researchers to miss important articles is that they're not searching with the terms that are being used in the literature. For example, searching for the term smartphones in PubMed, even though it is a term we commonly use in everyday life, returns far fewer results than searching with the term cell phones. This lecture will cover strategies for identifying terms to include in your search, focusing on identifying search terms by scanning the results from a basic search, analyzing a few relevant articles, and using the mesh database. Let's get started by using the following scenario. Let's say that you've been placed on a team that will be developing a new app to help those with high blood pressure eat a low sodium diet. As a first step, your team decides to conduct a quick review of the literature to see if any mobile apps like this already exist, and if they do, how and where were they used and were they successful. First, let's identify the main concepts we want to explore. From our question, we'll use the concepts low sodium diet, blood pressure, and mobile app. Sometimes the best way to find out what search terms to use is to dive into PubMed and see what you find. To do this, I'll start with a basic search. That is, I'll type low sodium diet and blood pressure and mobile app in the search box and click on the search button. There are many more sophisticated search techniques that we could employ here, but right now we're simply trying to get enough results to see how the topic is being discussed in the literature. As we scan through the results, a few alternative terms for our concepts begin to rise to the surface. For example, from our mobile app concept, I'm seeing mobile technology, mobile healthcare, and mHealth. For low sodium diet, I'm seeing sodium intake, salt intake, and dietary sodium. Keep track of these terms as you go so that you can easily incorporate them into your search later on. Personally, I like to toggle between PubMed and a word doc, writing down terms as I find them. As you go through the first page or two of your results, it's likely that you'll identify a few relevant articles on your topic. Taking a close look at these items can help you identify even more search terms. For our example, let's click on the article titled Development of Dash Mobile. In the detailed record, we can comb the abstract for additional search terms. Sodium intake is mentioned again confirming that this was a good term to add to our list dash diet, which is a diet that encourages reduced sodium intake, also sounds like something we should look into. As we scroll down the record, PubMed provides a list of similar articles and a list of articles that have cited this one. Use these to gather more relevant articles and to collect additional search terms. Last, if a citation has been indexed with mesh, you'll see a link to jump to the mesh terms using the right-hand side navigation. Remember, mesh or medical subject headings are the controlled vocabulary applied by specialized indexers to every article indexed in Medline. They cover biomedical topics including basic biological sciences terminology such as genes, proteins, and biological processes. You can tell a lot about the article that you found by looking at the mesh. For example, notice that the mesh terms hypertension slash therapy and mobile applications both have asterisks at the end of them. This is the symbol PubMed uses to show that these terms are considered a major concept of the paper. Indexers will assign anywhere from two to five terms as major concepts in one paper. The reason hypertension has a slash and then the phrase therapy is because therapy is a subheading. In addition to applying the most appropriate mesh to a citation, indexers also have a list of subheadings that can be applied to further emphasize the specific focus of the paper. For this term, it means that this article is not just looking at hypertension but focusing on the treatment of hypertension. The mesh list can also provide you with suggestions for new terms to include in your search. In the case of our example, hypertension and mobile applications look like important terms to add to our growing list. Using the mesh database is another great strategy for identifying relevant terms to include in your search. You can access the mesh database from the PubMed homepage. Click on the mesh database link underneath the explore heading. Or, if you're looking at the list of mesh terms associated with an article, you can click on any of them. When a pop-up box opens, click on the link that says search in mesh. While the mesh database looks a lot like PubMed, you're actually in a different database. You'll know that you're in this database because the box next to the main search bar will say mesh instead of PubMed as it usually does. To mine for search terms in the mesh record, the first place to go is the section titled entry terms. People often think that entry terms are more mesh terms but they're not. Actually, they're synonyms for the mesh term. They're called entry terms because when you enter one of these terms into the PubMed basic search box, mesh will automatically apply the assigned mesh term to the search. As synonyms, they represent the common terms people often use when writing their titles and abstracts and make for good alternative terms to include in your search. Thinking about our search, the search terms we might want to grab from the list shown on this page are portable electronic apps, portable electronic applications, portable software applications. We can grab the plural and singular version of these phrases. Notice I'm not including the entry terms that are inverted such as application comma mobile. Even though they are on the list, it's very rare that an author will use that term in this way and searching with inverted terms will yield few results. You might even notice that some of the mesh terms such as diet comma sodium restricted are written this way. Strangely enough, inverted phrases like this one are actually quite common in the mesh database. As a final point about using the mesh database to generate search terms, if you find yourself in a situation where you're struggling to find enough articles, look at the mesh hierarchy for your concepts and see if perhaps there are broader terms that you can use instead. This hierarchy is listed just below the entry terms. It shows the relationship between the mesh terms. If there were any narrower terms, they would be listed below our mesh term. The broader terms are listed above. For example, if you weren't finding enough articles about low sodium diet, perhaps you could expand your search by using the term diet therapy or nutritional therapy. If we were conducting a comprehensive literature search on our topic, we would search the mesh database for all three of our concepts and expand our list of search terms until we were confident that we were capturing all relevant articles on the topic. At this point, however, I think we have generated a pretty good list of search terms for our purpose of reviewing the literature. In summary, looking through the initial basic search results, closely examining relevant articles and exploring the mesh database was enough to help us generate a few additional search terms for our research question. If you find yourself really struggling to identify additional search terms for your topic, online tools such as dictionary.com, thesaurus.com, Wikipedia, and drug databases may be able to provide ideas of common terminology that you can use. And remember, you can always ask a Welch Medical Library informationist for additional suggestions. Once you have your list compiled, you can start developing a slightly more sophisticated search. As you continue to search, always keep an eye out for more relevant terms that you might have missed in this preliminary round. A good search strategy is one that continually evolves and gets refined as you go.