When searching for evidence-based information, one should select the highest level of evidence possible--systematic reviews or meta-analyses. Systematic reviews, meta-analyses, and critically-appraised topics/articles have all gone through an evaluation process: they have been "filtered".
Information that has not been critically appraised is considered "unfiltered".
As you move up the pyramid, however, fewer studies are available; it's important to recognize that high levels of evidence may not exist for your clinical question. If this is the case, you'll need to move down the pyramid if your quest for resources at the top of the pyramid is unsuccessful.
Greenhalgh, Trisha. How to Read a Paper: the Basics of Evidence Based Medicine. London: BMJ, 2000.
Glover, Jan; Izzo, David; Odato, Karen & Lei Wang. EBM Pyramid. Dartmouth University/Yale University. 2006.
Different types of publications have different characteristics.
Primary sources are original materials. It is authored by researchers, contains original research data, and is usually published in a peer-reviewed journal. Primary literature may also include conference papers, pre-prints, or preliminary reports.
Secondary literature consists of interpretations and evaluations that are derived from or refer to the primary source literature. Examples include review articles (specifically meta-analysis and systematic reviews) and reference works. Professionals within each discipline take the primary literature and synthesize, generalize, and integrate new research.
Tertiary literature consists of a distillation and collection of primary and secondary sources such as textbooks, encyclopedia articles, and guidebooks or handbooks. The purpose of tertiary literature is to provide an overview of key research findings and an introduction to principles and practices within the discipline.
Original research results in journals,
|Abstracting and indexing services, review articles, systematic reviews, meta-analysis, practice guidelines||Textbooks,encyclopedias, handbooks, newspapers|
Sources: NEJM, JAMA
|Sources: PubMed, CINAHL, Cochrane Library, Web of Science||Sources: Goodman & Gilman's, Williams Obstetrics|
Adapted from the Information Services Department of the Library of the Health Sciences-Chicago, University of Illinois at Chicago.
Different kinds of information sources are useful depending on your topic and the type of information you need.
Books are good for general background and in-depth coverage of a topic. They are often not as current as journal articles because they take a long time to research, write, and publish.
Books may be useful when:
Books may not be useful when:
Articles tend to be narrow in scope and are good for focused treatment of a topic. Scholarly journals contain high-quality articles usually written by experts and use data and statistics to back arguments. Popular magazines and newspapers (such as Newsweek, People, or the New York Times) are good for current treatment of a topic and are good resources for editorials and opinions. Note that popular magazines are not peer-reviewed.
Articles may be useful when:
Articles may not be useful when:
Web sites can be very good for finding quality information including primary sources, statistical information, educational sites on many levels, policy, opinion of all kinds, and much more. However, you have to take the responsibility to rigorously evaluate each site for quality; anyone can post a Web page, regardless of their expertise or intentions.
Adapted from the University of Connecticut Libraries