Audiology and Deaf Education

Please contact Cathy Sarli if you need assistance with literature searching.

 What is a Quality Literature Search?

A quality literature search is a systematic survey of the literature, (published and unpublished), on a research topic or clinical question. It goes beyond a cursory search of the literature to selecting appropriate databases, creating a series of search queries using relevant keywords including controlled vocabulary keywords, reviewing each result, filtering out non-relevant results based on specific criteria, reading the full text content of the selected results and performing a critical appraisal of the literature to understand the context of a research topic or to answer a clinical question.

Components of a quality literature search include:

• Formulating a structured question using PICO analysis if the topic is based on a clinical question
• Applying specified criteria for inclusion or exclusion before executing a search to reduce selection bias
• Use of more than one database or resource including the Web
• Use of appropriate databases
• Locating references that provide a clear chain of evidence in demonstrating the accumulation of knowledge on a particular topic
• Identification of appropriate keywords including controlled vocabulary keywords
• Develop queries using controlled vocabularies and keywords based on natural language for each database or resource used
• Testing of various search queries on multiple databases
• Use of applicable search limits
• Use of applicable search tags
• Refining search queries as needed
• Identification of what is known
• Identification of gaps or flaws
• Use of primary sources including peer-reviewed materials
• Use of secondary and unpublished materials
• Reading the full text content of materials
• Following up on relevant cited references to track subsequent research related to a specific publication
• Performing critical appraisal of the literature for quality and relevance
• Checking references for errata or retractions
• Identification of possible adverse events or potential harm for human subjects
• Awareness of the validity of the methods used for a particular study
• Documentation of resources consulted, search strategies, and search queries as used
• Documentation of findings for proper attribution and prevention of plagiarism

After completing a quality literature search, PACS students should have a thorough understanding of the topic, be able to demonstrate knowledge of the topic, provide original concepts to supplement or expand on the topic, and for clinical questions, be able to locate the best evidence for answering the question.

Where to Start?

For PACS students, there are typically two reasons for searching the literature; the first being to locate the best evidence for answering a clinical question, and secondly, to locate resources for a research-based question for capstone or independent study projects and other projects.

Literature Reviews for Clinical Questions

Searching of the literature for resources to answer clinical questions may benefit from using a PICO analysis. Use of a PICO analysis can help structure clinical-based questions into a framework for formulating search queries in order to retrieve optimal search results.  The use of a PICO analysis stems from the concept of evidence-based medicine which is defined as “the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients” (see PICO Chart box). The first step in using a PICO analysis to identify the components of a clinical question into concepts based on each PICO element as noted on the PICO Chart.

The next step is to identify all known keywords such as controlled vocabularies or thesauri keywords including natural language keywords that relate to each PICO concept (see Keywords box). One recommended strategy for starting a search of the literature for clinical questions is to formulate individual search queries based on a single concept and combine the queries. Many databases (see Databases tab) allow for flexibility of combining multiple single keyword queries into a single search string. The idea is to experiment and “get your hands dirty” with various queries on different databases. Try out a search query using keywords from each PICO concept or limit to one keyword from a PICO concept. As each new query is tested out, document each query so that the same queries can be tested out in other databases. The same query will not produce the same results in different databases. 

Literature Reviews for Research Questions

Searching the literature for research questions requires extensive review of the literature which is recommended for exploring topics for capstone and independent study projects. Once a capstone or independent study project topic or other research question is identified, the first step is to identify all known keywords that relate to the research question as done for clinical-based questions and develop queries using more than one database. Searching for a topic related to a capstone or independent study research project is a reiterative process and requires using multiple keywords to formulate queries and use of several databases.

Searching the literature is a skill that is best learned through trial and error. Practice with various databases to learn more about the nuances of each database. Experiment with different keywords such as author keywords, controlled vocabularies and natural language keywords to discover how various databases interpret the query. Review the keywords as noted for each work.

Do not get discouraged if results are not relevant to the query. Points to consider are:

• Can the query be paraphrased or re-worded?
• Can the number of keywords be reduced?
• Are the Boolean operators being used correctly?
• Was correct spelling used for all the queries?  One misspelling can produce negative results.
• Were keywords from a controlled vocabulary used? Databases that index using controlled vocabulary keywords usually offer a tool to search the keywords in the controlled vocabulary.
• Are there any other keywords based on natural language to consider for the query? 
• Was more than one database used? One database will not contain all references to the literature.
• Was gray literature searched?
• Were the features and functions of each database utilized for the search?