For systematic and scoping reviews, the search strategy is also the research method. You'll need to be meticulous in creating and keeping track of all the details that go into it (more on that on the How Do I Report My Methods? page) so you can give a clear report in your paper. Ideally, you will provide enough information in your paper so that any reader could replicate your search and get the same results.
Since your review may take several months and cover multiple rounds of searching, keeping track of your search strategy documentation from the beginning is always a best practice. It will also make completing the PRISMA flow diagram and potentially make answering questions from peer reviewers much easier!
Below is a link to the PRISMA Search Guidelines, which tell you which type of information you should keep track of during your searches. Further down the page, you'll find some tools and documents that you can use to keep track of the information.
You can save yourself some time and frustration if you plan your search strategy before you start looking for articles. It's like having a shopping list before going to the grocery store. Of course, you may be fuzzy on some of these items to begin with, and that's OK. You can continue to refine your search strategy as you learn more about your topic.
However, you will need to stop refining and define your search strategy before you begin the title/abstract phase.
What's in a search strategy?
For more about all of these elements, take a look a our Expert Keyword Searching Guide.
Keywords are the main ideas that describe your topic. You'll usually find them prominently listed in your research question. That said, because the English language is vast and varied, there are many words that describe similar ideas. You'll want to include as many of those variations as makes sense for your topic.
Getting the right combination of search terms is crucial to building an effective literature review. We highly recommend you have a librarian involved with at least this step of the review process.
The big challenge here is being broad enough with your search to capture all of the relevant literature without being so broad that you are overwhelmed with too much to actually look through.
How do you achieve a broad search?
As you begin to gather your keywords and test out your searches, here are some things to keep in mind.
Your search should incorporate subject headings, in addition to keywords. Subject headings are a set of standardized terms, assigned by databases to describe concepts in an article. They can take some of the guesswork out of the job of coming up with keywords, because regardless of what term an author uses, the database should assign that standardized term for the concept.
An example would be, you want to do a search that talks about soft drinks; you know some regions call it pop, while others call them soda, others refer to them by their brand names. PubMed assigned the subject heading "Carbonated Beverages" to this concept. Now your search can also include "Carbonated Beverages" with the hope that it will pull in any articles that use terms you didn't include in your list of keywords.
Still Unsure about Subject Headings?
Take a look at this description in our Expert Keyword Searching guide.
Search hedges and filters are premade search strings of keywords and subject headings (database-dependent) related to specific topics. The idea is that these should be comprehensive and include all relevant terms, saving you time when developing your search string.
You may need to adapt these searches and translate the subject headings for the databases you are using in your searches, but you will always want to cite the search hedges you use.
Example Resources
This is a common question we get from our researchers, and technically, there's no absolute answer. But we like to provide the following guidance.
Spot Check Reference Lists
Using a handful of articles that you know you'll be using in your review, review the search results to make sure they appear in your search. Alternatively, you can use the references list from similar scoping or systematic reviews, and check to see if those results are appearing in your search.
What if they aren't appearing in your search?
This can happen for a few reasons. If there are only two or three articles that weren't appearing in the results, but a majority of them did appear, then it should be ok. It's possible that the articles aren't available in the databases you're searching in.
However, if you are unable to find a majority of the articles, try incorporating Google Scholar into your search.
If after adding in Google Scholar, you still cannot find most of the articles, it's likely at a point that we don't have the right types of databases for you to do this type of review. While this is frustrating, you may be able to pivot to a different type of review.
While working on a systematic or scoping review, you should be searching across multiple databases. Not all databases use the same format for searching; in this case we want to "translate" the search to work in each database.
As an example, you can use "Artificial Intelligence"[tiab] to search for the phrase "artificial intelligence" in the title, abstract, or keywords in PubMed, but that same search won't work in CINAHL. You would want to search "(TI artificial intelligence) OR (TX artificial intelligence) OR (AB artificial intelligence)"
Tips and Tools to Automate the Translation Process
Polyglot
A word of caution regarding Polyglot: while it automates the translation of search syntax from NLM PubMed/OVID MEDLINE formatting into other databases’ formats to reduce manual translation time and effort, it is not perfect. It will not translate subject headings or filter field tags - you must manually do that part.
How to Use Polyglot
Optimal Databases and Macros for MS Word
This review from Bramer et al. found that the most optimal database combination for systematic reviews is Embase, MEDLINE, Web of Science, and Google Scholar. Not all libraries have access to these databases, but using a free-to-use database like MEDLINE (PubMed) and Google Scholar along with another database like CINAHL or PsycINFO, depending on your topic, will help ensure you have a broad search. If you need database recommendations, please reach out to a librarian.
Part of this article references Macros that can be added to Microsoft Word to help automate the translation of searches. These are more relevant for those who are tech-savvy though. You can also use the link below to find instructions on how to download and use the macros.
Using Excel to Create a Search Strategy
University of Illinois Chicago has a guide about creating a comprehensive search using Excel to manage the search as you go. At the bottom of the page, you can download an Excel document (Systematic Review Documentation & steps) to help track your search. Alternatively, if you're just interested in a tool to help create a search with subject headings, you may want to try the "Spreadsheet for formatting and compiling terms."
Once you're satisfied with your search, you'll want to export it, and ideally, import it into a citation management (like Zotero) and/or a systematic review management tool (like Covidence).
Our Zotero guide has instructions on how to export all of the search results from EBSCO-based (like CINAHL), PubMed, and Ovid-based (like PsycINFO) databases.
One of the ways you can try to make sure you didn't miss anything is to search reference lists and citing articles.
Typically, you only do this for the articles that make it through the screening and full-text review process, in other words, only those articles that will be included in your review. You might also consider setting aside the systematic and scoping reviews your search found and mining their reference lists as well.
You can certainly conduct the search manually, meaning comb through all of the reference lists and consult Google Scholar to find the articles that have cited yours, but there's an easier way. You can use an online application (see the links below)!
Once you have the list of articles from the reference lists and the articles that cited your articles, import them into whatever application (e.g. Covidence) you are using to manage your review. Many will be duplicates of articles you've already seen and will be immediately eliminated. Then you start the screening process all over again with the remaining articles. It should move rather quickly, because most will be easily eliminated as irrelevant.
Online Applications for Citation Mining
If you watched the video in the Introduction to Grey Literature box, you will have learned that finding grey literature is often time consuming and, frankly, a pain in the neck. Google searches produce WAY too many results to reasonably look through in any systematic way. If your library does have subscriptions to any grey literature databases (e.g. Conference Proceedings Citation Index or SafetyLit), they often have limited value (consult with your librarian to learn more about those sources).
The thing you want to remember as you approach your search for grey literature is that you want to make it as systematic and replicable as possible. The searches, after all, are your research methods.
Here is what I recommend as an approach to grey literature.
The next section will help you think about possible sources for grey literature.
Searching for grey literature is a little more of a project than searching library databases. On the following tabs, we'll provide examples of various types of grey literature and where you can find it.
Many new ideas come to conferences long before they're published in journals. You will want to check the archives of any major conference in your discipline or relevant to your research question.
Communication Sciences & Disorders
Nursing
Occupational Therapy
PhysicalTherapy
Physician Assistant
Online trial registries may include results of completed but unpublished clinical trials.
The United States government loves to collect data. These are just a few of the places they make it available to us. If your topic doesn't fall under the domain of one of these areas, another agency undoubtedly covers it.
Please be aware that content on U.S. government websites is in transition due to President Trump's recent executive orders. We are attempting to download what we can before it disappears or is modified. If you encounter broken links or incomplete information on the linked websites, please let us know.
There is an abundance of health-related advocacy organizations, many of which collect data and relevant information for their constituencies. They may also publish white papers and practice guidelines that could be informative to your work. If you do not know the advocacy organizations related to your area, these lists might help you find them.
Dissertations and theses are great sources for new developments and trends in the literature. They also tend to have amazing reference lists.
If you'd like to take a more international perspective on your topic, be sure you check data and white papers coming out of these international organizations (and any others that might be related to your topic).
The New York Times (NYT) is a daily newspaper based in New York City that covers domestic, national, and international news, and publishes opinion pieces, investigative reports, and reviews.
Preprint servers are a great place to find articles that will likely be published, before they are actually published (hence, the title, pre- "print"). These articles typically have not yet gone through formal peer review, editing, or publishing process.
Preprint servers are great for "topics that are under-studied, novel, or rapidly-changing" (Jewell, 2025).
Google has been a long-standing search engine for many to begin their searches. Google Scholar, which finds "scholarly" results across the internet. And some institutions have created their own specialized Google Search Engines to help you find relevant materials.
While it's not recommended to only search Google Scholar, ensuring that you are using it to help find grey literature for your systematic review is important. Haddaway et al. (2015) studied how useful Google Scholar was when it came to finding grey literature, and it worked surprisingly well for the most part; while it's not as effective as some library databases like Web of Science, if your library doesn't have a subscription to the database, it's a decent alternative.
Below are a few examples of Google-based Search Engines
Another option is to use Open Access databases to find grey literature. These are helpful since they provide free access to the referenced articles, but you should always be sure to evaluate article quality.
