Good day to everyone. My name is Eric Estrada. I'm a gynecological oncologist from Hospital General San Juan de Dios in Guatemala. Today I was appointed to give a lecture on literature search. When we talk about literature search, there's something important that we need to highlight. There's the, when we say about literature, literature is the only way we can actually pretend to know what's happening around the world. For this talk, I have no conflict of interest. However, we know the how and where to start because you already have the who and the when. When we have something that might just impact in our, in our field, we have to have a genuinely and clearly formulated question that could answer or reinforce the gap in knowledge that we may have. This is very important when you start. However, I think it's, I believe that it's important to share this question with a mentor or any other colleagues that could give us an input. And when we have clearly what we want, we have to look out in the literature that will allow us to review and identify as many relevant studies as possible to answer this question or reinforce the knowledge that we may already have. So how do we identify the articles that are relevant? First of all, this question that already has been in our mind is what we call a PICO. PICO means what is our patient or population? What is the intervention that we are in interest? How we want to compare and the outcomes that we want to highlight? So the most important thing is when we actually want that articles is to identify the proper sources and database, then develop a research strategy, run the search and apply practical screens. When we talk about practical screens means that we have to exclude something that is not part of our strategy. And we can actually put it down in some of the sources and database that we're looking for. And then when we have all the database that we wanted, for example, all the amount of articles that we might extract it from this database, we can then export the reference and then we can further evaluate these articles. But the problem is having the question. And I believe it's very important in this slide. How the question influences our search. If we have a broader question, that means we may have highs and lows of articles when we retrieve it from the search results. However, if we have a narrow question, we have a low and very low volume of articles, but we may have an impact on the high and directly relevant articles that we may have. So this is very important. When we ask our question, the broader the question is more articles and low relevancy. And when the narrower question is done, we have a high and directly relevant articles. And this influence always influence our impact when we are actually looking for research. So later on, where we start searching, that's important because searching is the current based evidence in the medical literature and have become a central skill in our clinical practice. First of all, sometimes performing a randomized controlled trial can be very difficult. Can be very... Basically, it's something that actually has a lot of work to be done. However, in our cases, most of in Latin America, probably in areas of Asia and Africa, we can do, for example, retrospective data. So this is where we look at article databases. The standards of what we have, it's a PubMed, Ovid, Medline, Embase, Cochrane, Ray Library, Scopus, or the Cochrane Central Register for Controlled Trials. This is actually one of the highest searches that we have or we may have available. However, depending on the type of literature that we may want, there's also the gray literature. This means the articles that are not actually published in the databases, however, are part of important and relevant studies. For example, like epidemiology. This is why we have the gray literature. That means the literature that is actually published in association, the organization, government reports, for example, the WHO, or for example, in Latin America, we have the Pan American Health Organization. However, we may look at web research engines. For example, Google Scholar has a web search engine. That means only for proper literature. When we have those databases accessed, we can have what we call baby steps. First, select your database. Break your question into concepts. This is very important because we're going to talk about what is a MESH, what stands up for MESH, what's important of MESH, and identify the subject headings for each concept. When we have a subject heading, you can actually have the subheadings and actually that reinforces the concept that you may want to look for. The fourth thing is that you need to do is identify the text words for each concept. The tips are important here. When you have a tip, the tip means that if you have a target article, it can help you identify search terms. For example, if you have an article that's very relevant to your practice, this could be your target article. Sometimes target articles have what we call search terms. These terms are actually very important for your relevant questions and what you're looking out in your database. Sometimes it's very useful to use a worksheet to keep on track of their terms. That means that you do a proper worksheet. You may actually not repetitively looking for the same terms. If you have one term already being used, you don't have to be repetitive all over the way. A MESH is actually very important. It's a medical subject heading. This is actually the practically vocabulary thesaurus done by the National Library of Medicine used for indexing articles in Medline and PubMed database, which is actually Medline and PubMed database are the largest pool of articles are being published there. MESH, it's actually something that will help us in each article citation because it's actually being associated to these terms that describe the contents of the citation. This means that, for example, most of the database that we use are actually having MESH subheadings. It actually will help you once you've mastered the MESH subheadings or heading, you can actually have all the proper citations that you want for your research articles. Then you have, for example, M-Base. M-Base uses some things very similar to Medline and PubMed, but they call it M-Tree. For example, Cochrane also uses a MESH and the tip is here to select the subject heading that is closest to the match of your concept. Once you have this, you have practically most of your search to be done. We're going to explore, for example, in the next slide, how can you do it in a proper way and how it means that the broader the spectrum of your search, the higher amount of articles that you may like, you may obtain. However, if you shorten actually your heading, that means you only have the relevant studies that will actually have the most important impact in your research. MESH is something that usually is very helpful. This is actually a screen from the PubMed website. When you access the PubMed website, you have on the bottom of your screen the MESH database. Once you click there on the MESH database, it actually sets you up to the MESH advanced search builder. This means that the MESH terms are arranged hierarchically by subject category. For example, the principal heading is actually what is, for example, the basis of your search, what is actually you're looking for. For example, if we look at the PICO question that you may have already done, the interest probably would be, for example, in this case, cervical cancer. The intervention would be radical hysterectomy. The comparison could be, for example, lymphadenectomy or sentinel lymph node detection. The outcome could be what you're looking for, overall survival, progression for survival, or even complications. Once you have all this, you can actually look for all of the terms together or the terms one by one. For example, if you like, for example, cervical cancer, you can add and radical hysterectomy and lymphadenectomy or not, for example, sentinel lymph node. The outcome, you could only look for overall survival, progression for survival, or complications and not, for example, if you're not interested in overall survival, you can exclude overall survival from all the fields that you may like or you were wanting to learn. Once you actually master this thing, you can actually get the most out of it. For example, you can apply also screens and limits to your search. Applying, for example, screens or limits could set you up, for example, not looking for articles that are not actually published in English or, for example, French language. If you're not actually looking for articles past, for example, 2000 or present, you can actually set the year from a publication to the current status. And this actually would retrieve only articles published between 2000, for example, and the current date. Also, you can have the date of publication, the participants or subject, for example, you're not actually looking for locally advanced cervical cancer, just actually, for example, stage 1B, 1B2, or other subjects that you are actually looking. And if you're not looking also for research design, for example, if you're not looking for clinical trial, you're, for example, looking just for systematic reviews and meta-analysis, you can actually limit the amount of searches that you're actually looking for. Once you have this, you can narrow your search. For example, I did mine, in this case, I did a mesh and I said, I'm looking for uterine cervical cancer. And once you search uterine cervical cancer, the National Center of Biotechnology or the National Library of Medicine actually sets you for some of the popular mesh words that you are actually looking for. For example, you can actually click on surgery because you're actually looking for surgery, but you can actually put and, or, or not. For example, you can actually click not, for example, on all the other things that you're not interested. But if you are actually looking for surgery, you can only click for surgeries and add up to the search builder. Once you add for the search builder, you can actually, for example, if you have a narrow, want to be very broader, for example, if you just click on uterine cervical cancer, the search could retrieve, for example, 83,916 results. But if you, for example, narrow the search for uterine cervical cancer and, for example, stent and lymph node, for example, you can retrieve only 383 results. The others are not actually important to your question, but it's just only relevant to the one you're actually looking for. And this would be your search. Once you have your search done, it actually retrieves only the articles that you may, you're actually wanting, but you can actually broaden them, broaden them into, or diminish them. For example, if you don't want, for example, the article from 1966, you, for example, want only the articles from 2001, for example, to up to date, you have known until 2023, and you can actually narrow the search even more. So with this, it's very important that when you have a relevant question and you want to answer it with research, you have to mastermind the MeSH database, since it's very useful to find the search terms for your search, queries, and even in other databases. And when the MeSH databases can help you find all the related terms, like we saw before, and the synonyms to use as search terms in your query or in a specific concept. And the MeSH entry terms, instead of a keyword searching, it can let you focus on your search and find more relevant citations than if you could have already done by keyword searching. That means it would take a lot of time, tremendous effort, and probably you will be missing a lot of the relevant articles that you are actually wanting to put in your research conclusions. So with this, I'd like to thank you for hearing me today. Thank you.

literature searches

knowledge

PICO framework

sources and databases

gray literature