While OSU Libraries does not offer an evidence synthesis search service at this time, talking with a librarian for your college or program about your planned comprehensive search can help you be more efficient when:
1) choosing what databases/search tools to use for your comprehensive searches,
2) developing your finalized search strategy, and
3) deciding on how to manage your search results.
When scheduling a consult, it will help your librarian to know 1) what your finalized research question is, 2) what your draft search stratregy is, and 3) 2-5 articles that address your research question, likely found during your exploratory searching (see the "Seminal articles" module on the "Do Exploratory Searching" page). Be prepared to provide them with this information before you meet.
A comprehensive search is a systematic effort to find all available evidence to answer your specific question(s).
The validity and usefulness of a synthesis hinges, in part, on a high-quality comprehensive search. Like all the other stages of a systematic review and/or meta-analysis, the process itself should be replicable.
First identify the type of material that can answer your question - this may already be part of your eligibility criteria. For most research questions, you will likely need at least peer-reviewed empirical research.
In some cases, it may make sense to only include peer-reviewed research, or even a specific type of research like randomized controlled trials. Peer-reviewed research should be located systematically so that the search is replicable and comprehensiveness can be reasonably justified. Therefore, a comprehensive search for peer-reviewed literature takes place primarily in academic journal databases.
In other cases, grey literature may be required to properly answer a question. Grey literature is a broad term that varies across disciplines. Some common examples of grey literature include unpublished research, conference proceedings, government publications, social media content, blogs, newspapers, datasets, etc. Grey literature can rudimentarily be defined as anything that is not peer-reviewed, empirical research.
Because of this variation, finding grey lit in a systematic, transparent, and replicable manner can be challenging. Where you search will vary based on what kind of grey lit you're looking for - how you search will vary based on the options available within the interface or database. However, it is important to document your search terms and process to be as systematic, transparent, and replicable as possible. Use the Gray Literature guide to help you being this search process.
Once you've identified what kind of material you're looking for, you can identify where to search. This will include academic journal databases at a minimum. If your strategy includes grey Literature, use the resources on the Gray Literature guide.
The design of your search strategy will depend on what you're looking for and where you're looking.
PRISMA-S is a reporting guideline for the search strategy. It should be used in conjunction with a systematic review and/or meta-analysis reporting guideline (e.g., PRISMA) and can also be used with other methods that require the reporting of your search. Using this guideline will help you ensure "each component of a search is completely reported and...reproducible".
Credit: "Comprehensive Search" (Virginia Tech University Libraries) adapted by OSU LIbraries & Press under CC BY-NC-SA 4.0.
In a systematic review and/or meta-analysis, the search should collect all potentially relevant material that is available. In other words, the search must be comprehensive. Other evidence synthesis methods may allow for more flexibility in the comprehensiveness of the collected results, however the search strategy design should always be intentional
Though exact search strategies will vary according to where you search, you can start by designing a base search strategy by following these steps (Bramer et al, 2018):
Remember that the concepts you work with and develop search terms for should relate to the various parts of the research question you developed using the appropriate question framework.
Search strategies may be crafted in different ways and no one way is correct (though different evidence synthesis methods may require different strategies):
The search reporting guidelines (e.g., PRISMA-S) you use may ultimately determine how you craft, run, and document your search strategy.
Credit: "Designing a Comprehensive Search" (Viirginia Tech University Libraries) adapted by OSU Libraries and Press under CC BY-NC-SA 4.0.
References:
While it might be tempting to create database searches that make use of built-in limits (e.g., publication type, methodology, population types, etc.), using these limits may miss relevant studies when database entries representing those studies have not been fully or equally processed to include limit information. Furthermore, Lefebvre et al. (2024) offer the following guidance:
Searches should capture as many studies as possible that meet the eligibility criteria, ensuring that relevant time periods and sources are covered and not restricted by language or publication status (see MECIR Box 4.3.a). Review authors should justify the use of any restrictions in the search strategy on publication date and publication format (see MECIR Box 4.4.d).2
Pre-developed and (usually) validated search strategies (often called filters or hedges, especially in the biomedical disciplines) are designed to retrieve specific types of article records (e.g., specific publication type or methodology, specific research population, etc.) These strategies may be used in conjunction with the search strategies designed to address other concepts present in the research question(s). Most validated strategies/ are database-specific, and there may not be validated strategies for all disciplinary areas. Here are some well-regarded sources for validated strategies:
References:
While many strategies for constructing a database search are similar among different databases or search tools (e.g., the use of controlled vocabulary and/or natural language terms or the use of boolean operators to combine search terms and concepts), each one is unique and will require a customized search strategy. For example, the controlled vocabulary for the PubMed database will be different than the controlled vocabulary for the PsycInfo database; some search tools have little/no controlled vocabulary (e.g. Web of Science database, Google Scholar). Each database or search tool has its own syntax for truncation or wildcard characters or proximity searching (and some databases don't have all of these features). Some databases or search tools also have character limits that prevent copy/paste of lengthy search strategies. To address this, you will need to be familiar with the features and syntax of the different databases you plan to use for your searching (the Help files are your friend). You cannot simply copy/paste the search strategy from one database to another database or search tools. You must customize (aka translate) your searches when moving from one database/search tool to another.
There are some resources and tools that help simplify the customization process. These resources are not perfect and the resulting searches must be examined carefully and adapted as necessary. Not all databases are addressed or available in these resources.
OSU Libraries provides support and some training for the following tools:
References can then usually be exported from Zotero/EndNote to additional software for the remaining screening processes.
Systematic review software also has capabilities for saving search results. Some examples:
