Systematic reviews aim to identify all the relevant studies on a topic. This means searches for this type of research are extensive and time consuming. Studies in these reviews are limited to primary source research studies, never reviews or secondary sources. You are looking for randomized control trials.
KCU offers a variety of databases that may be used to complete a systematic review or meta-analysis. Which ones a researcher chooses will depend on the precise topic and PICO question. For assistance in selecting the most appropriate databases for your review topic, please reach out to a KCU librarian.
Commonly used databases are:
For a full list of KCU databases, please click here
Developing search strategies starts with identifying key terms. This can be accomplished by using PICO to develop your research question. Your best bet with keywords will come from each part of your PICO. Your list should be as comprehensive as possible. You can always narrow down to the best terms later.
MeSH terms are used by the National Library of Medicine in indexing articles for MEDLINE. These are not quite the same as keywords, and will not be found in non-indexed articles on PubMed. If you're not sure how to use MeSH terms, reach out to any librarian in our libraries.
Two things need to be balanced to obtain an optimal search strategy: precision and recall. A successful strategy is precise enough that is pulls a small number of irrelevant studies, but its recall, or sensitivity, is enough to pull a high number of relevant studies. Looking at one or the other doesn't provide a full picture of the search strategy. In statistics and information retrieval, as well as machine learning, these things are calculated and formulated to fall within specified ranges. For developing search strategies, it tends to be much less defined: if you run your search and get a lot of irrelevant results, the strategy is not good and you need to try different terms.