Keeping Up With... Systematic Literature Reviews

This edition of Keeping Up With… was written by Cynthia Thomes.

Cynthia Thomes is a Reference and Instruction Librarian at University of Maryland University College, email:  cynthia.thomes@umuc.edu.

Introduction to SLRs

Although systematic literature reviews (SLRs; also known simply as systematic reviews or as systematic overviews, evidence summaries, integrative reviews, or research syntheses) have been conducted for more than one hundred years [1], it is only within the past two to three decades that they have become so common that they are now considered to be “a research standard” [2].

Because so many researchers are now conducting SLRs, and because librarians may well be asked to assist with the SLR research process, librarians may wish to familiarize themselves with SLRs.

How do SLRs differ from traditional literature reviews?

With traditional literature reviews (also known as narrative reviews or critical reviews), the goal is to discover what’s already known about a topic and perhaps to identify areas where gaps in the research exist, which can lead to new studies in order to further the state of knowledge on the topic.

Traditional literature reviews rarely contain a comprehensive list of documents on a topic. Rather, a set of documents is presented that is meant to be representative of the documents discovered during the course of the author’s research. It is not typical for the author of a literature review to explicitly state how the documents included in his or her review were found or how they were chosen for inclusion in the review. The findings of literature reviews may be biased, then, due to incomplete literature searching and/or to selective inclusion of documents in order to support an author’s opinion.

SLRs, however, are frequently conducted in order to create or revise policy or to make a decision, and therefore an attempt is made to identify and assess all relevant literature on a topic, so that any action(s) taken will be based on the best available evidence.

It’s worth noting, incidentally, that SLRs may examine qualitative evidence as well as quantitative data. A review of qualitative evidence is known as a meta-synthesis, while a review of homogenous quantitative data is known as a meta-analysis. An examination of both qualitative and quantitative evidence is known as a mixed-methods systematic review.

Because SLRs are used as the basis for evidence-based practice, they require specific iterative steps that must be taken and documented. SLRs are therefore much more comprehensive and transparent than traditional literature reviews; as one pair of researchers put it, SLRs “attempt to bring the same level of rigor to reviewing research evidence as should be used in producing that research evidence in the first place” [3].

What are the steps involved in conducting an SLR?

Per the Cochrane Collaboration [4], one of the earliest and best-known advocates for the evidence-based practice movement, the general steps involved in conducting an SLR are:

  1. Identification of relevant studies from a number of different sources (including unpublished sources).
  2. Selection of studies for inclusion and evaluation of their strengths and limitations on the basis of clear, predefined criteria.
  3. Systematic collection of data.
  4. Appropriate synthesis of data.

As noted earlier, SLRs are meant to be as comprehensive as possible, so searches should be run in multiple subscription research databases, as well as in resources that contain grey literature, etc. Researchers should document which resources were searched, as well as what search strategies were used, what limiters were applied, how many results were retrieved, etc., so that other researchers can replicate the search process and confirm SLR findings.

Selection criteria should be determined before research is conducted, so as to minimize possible bias with regard to which items are included in or excluded from the SLR. Such criteria may include publication language, publication date, population studied, type of intervention, etc.  Justification should be provided for the inclusion/exclusion criteria selected.

Thorough appraisal of search results will help to determine the soundness of the results discussed in the documents. The Centre for Evidence-Based Medicine provides links to several tools that can be used to assess research quality. [5]

There is not currently an agreed-upon standard to be used for reporting the results of SLR research; in fact, many guidelines are tailored to specific types of systematic reviews (e.g., epidemiologic reviews, interventional reviews, diagnostic reviews, etc.). Standards that can be adapted to fit multiple types of reviews include PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) [6] and STARLITE (Standards for Reporting Literature searches [7].

How might librarians be involved with SLRs?

There are many opportunities for librarians to become involved with the SLR research process, and, indeed, it has been suggested that “librarians’ expertise in literature searching is essential [emphasis mine] for a comprehensive and replicable search of the existing literature” [8].

Librarians can recommend subscription research databases, as well as sources for grey literature.  And since a truly comprehensive SLR should include all relevant evidence, including unpublished data, librarians may also need to identify subject matter experts whom researchers could contact to learn about any in-progress research. Librarians can also help develop or provide feedback on researchers’ search strategies and can help determine appropriate inclusion/exclusion criteria.

Librarians may also provide training on bibliographic management systems (e.g., RefWorks), which researchers can use to store and organize the citations for their search results and which can be used to de-dupe results. Librarians may assist researchers in setting up publication alerts or in conducting cited reference searching to find other documents on a topic. And they can help researchers with “snowball searching,” in which a given document’s reference list is used to identify other relevant documents.

Staff who process document requests may find that there is an increase in the number of interlibrary loan requests submitted as researchers attempt to gather all relevant documents on a topic and discover sources not available through their own library’s collections.

Finally, librarians wishing to shape policy at their own institution may want to brush up on SLR research techniques, since what they find during the course of their research may well be useful to administrators who are looking to make data-driven decisions.

Notes

[1] McKibbon, K. Ann. “Systematic Reviews and Librarians.” Library Trends 55, no. 1 (2006): 202-215. http://muse.jhu.edu/journals/library_trends/v055/55.1mckibbon.html.

[2] Harris, Martha R. “The Librarian’s Roles in the Systematic Review Process: A Case Study.” Journal of the Medical Library Association 93, no. 1 (2005): 81-87. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC545126/.

[3] Hemingway, Pippa and Nic Brereton. What is a Systematic Review?, 2nd ed. (2009). http://www.medicine.ox.ac.uk/bandolier/painres/download/whatis/Syst-review.pdf.

[4] John Wiley & Sons, Inc. “About Cochrane Reviews.” Accessed November 30, 2015. http://www.cochranelibrary.com/about/about-cochrane-systematic-reviews.html.

[5] Centre for Evidence-Based Medicine. “Critical Appraisal Tools.” Accessed November 30, 2015. http://www.cebm.net/critical-appraisal/.

[6] Moher, David, Allessandro Liberati, Jennifer Tetzlaff, Douglas G. Altmann, and the PRISMA Group. “Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement.” PLoS Medicine 6, no. 7 (2009): 1-6. http://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1000097.

[7] Booth, Andrew. “‘Brimful of STARLITE’: Toward Standards for Reporting Literature Searches.” Journal of the Medical Library Association 94, no. 4 (2006): 421-429. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1629442/.

[8] Murphy, Susan A. and Catherine Boden. “Benchmarking Participation of Canadian University Health Sciences Librarians in Systematic Reviews.” Journal of the Medical Library Association 103, no. 2 (2015): 73-78. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4404858/.

Further Information