Welcome to the systematic approaches to searching the health science literature. A solid literature search is critical to writing a scientific article. If you are writing any type of review, a systematic literature search is essential to the validity of your conclusion. In this series of tutorials, I'm going to cover the fundamental concepts and general procedure of searching the health science literature in a systematic manor. The keywords here are "systematic" and "review". A lot of you have probably participated in writing, or at least heard of, a systematic review, and there is a lot of confusion as to what exactly it is. The fact is, "systematic review" is a label given to a specific type of review that follows one of the few established standards, such as the Cochrane standard, which is detailed in the "Cochrane Handbook for Systematic Reviews of Interventions", the CRD standard, from the Center for Reviews and Dissemination at the University of York, which you can explore more in their guidance handbook. The IOM standard, from the Institute of Medicine--details of which can be found in this IOM report. And finally, the PRISMA standard, which you can find out more in the PRISMA statement. Now, at this point, I'm not going to delve into details of these standards. All you need to know, is that if you want to conduct a systematic review in the truest sense of the term, you should consider following one of these standards. As a matter of fact, quality control is so important in systematic reviews, that tools such as this AMSTAR tool, as described in this paper here, were developed to measure the methodological quality of systematic reviews. So why do systematic reviews have to follow such rigorous standards? Well, that's because the validity of a systematic review's conclusion directly depends on its methodological validity. As a researcher, you may be familiar with the concept of "study design". For example, you may often use the randomized controlled trial model in order to draw a scientifically valid conclusion in your study. In this case, randomized controlled trial is your study design. A systematic review is a scientific study itself, of all the available evidence on a certain topic. As a matter of fact, it is probably one of the few, if not the only types of health science study, that do not need an approval from your institutional review board. Because it is a scientific study itself, a valid study design is critical in order for it to arrive at a valid conclusion. Following the established standards when conducting the systematic review ensures that your review's study design is a valid one. Searching the literature systematically is an essential part of each of these standards, and is an important step in ensuring a valid study design for the systematic review. This series of videos will be focused on the search process. The search is only part of the systematic review process, and my next video will provide more details on the entire process, and how important a role search plays in the process. Today, I'm going to only make a rough comparison between a search done for a systematic review, and your ordinary literature search. Typically, without a rigorous study design, literature searching for a review tends to become a scatter shot process known as "cherry picking". You do a little bit of a search here and there, gather some results, and find a few articles that you can read which are usually published in well-known journals, and typically tend to support your preconceived argument. You then synthesize them for the review. As you can see, a lot of bias can be introduced in a review done in this way. Most importantly, you may not have seen all evidence available. For example, there may be valid evidence in unpublished literature or there may be evidence in articles not written in a language you can read. A systematic review, on the other hand, places a lot of emphasis on removing this kind of bias introduced by cherry picking. For example, a systematic review requires the most exhaustive literature search possible, not only in published literature, but also in gray literature. As a lot of scientifically valid studies, especially those with negative results, tend not to be published. For example, if the topic is related to the effect of a particular type of drug, you're then required to search the drug registries, especially for trials with negative or lackluster results conducted by drug companies that never saw the light of day. A systematic review also requires searches in literature from around the world, not only in the English language or from a specific geographical region, because in some cases, the most valuable evidence may not have been published in English, or in the academic journal literature. A systematic review may also require searches in a discipline other than the health sciences. For example, if the topic has an education aspect, like this one here: "Does the use of internet videos as an education tool lower HIV infection rate among adolescents", you're then required to search the education literature for relevant evidence regarding the educational effects of internet videos on adolescents, because a lot of the evidence may not have been reported in the health science literature. A systematic review also requires using every single possible search term for a given concept in a topic, which, not only includes controlled vocabulary (such as the medical subject heading MeSH), but also regular text words and all of their synonyms. In my previous tutorials, I always recommended using controlled vocabularies such as MeSH or the CINAHL Headings in a literature search. However, for a systematic review, you're also required to do text word searches in order to catch any possible indexing errors in the database. To learn more about the differences between the subject heading searches, and text word searches, please refer to my previous videos on MeSH and CINAHL Headings. A systematic review also requires that only methodologically sound studies can be included in the review. Articles that are right on the topic, but have flawed study designs, must be excluded from the review to ensure the validity of the review's conclusion. A systematic review may also require a hand search for the relevant articles in key publications and verification of the entire search process by contacting experts in the field regarding the topic. The search results gathered during the systematic search like this may include evidence that supports the hypothesis of the review, or that does not support the hypothesis. This brings us much closer to the complete picture of the review topic than a "scattershot", cherry picking-style search. We have been mostly talking about systematic review as a specific type of review that requires a systematic approach in terms of literature search. Sure, systematic reviews should be systematic, but that doesn't mean if you're not doing a systematic review per se, your search can be unsystematic. The term "systematic review" gives a false implication that there are also reviews that can be unsystematic. In fact, all reviews should be systematic. They should only differ in their degrees of rigor, exhaustivity, and comprehensiveness. How systematic your search should be depends on what type of review you're writing. This series of tutorials will be mainly focused on systematic searches required by a systematic review. If you are not writing a systematic review, you may not need to take every single step as detailed in the following videos. But no matter what type of review you decide to write, having a systematic approach to your literature search will only enhance your project. The goal of this series of tutorials is to ensure that your search is comprehensive, methodical, transparent, and reproducible - so that your conclusions are as unbiased and closer to truth as possible. Thanks for watching, and I'll see you next time.