Skip Navigation
Search Menu
Armstrong Institute for Patient Safety and Quality

In This Section      
Print This Page

JHMI Resources for Value-Oriented Care Research and Reviews

Welch Medical Library Informationists

View a list of our Welch Medical Library informationists with assigned departments and contact information.

Design and Analysis Support

Institutional Review Board

Definitions for Some Typical Types of Literature Reviews

Blair Anton, Welch Medical Library
banton2@jhmi.edu

  1. Narrative reviews: These reviews are also known as clinical reviews or expert summaries on an important clinical issue or problem. They may be written by a single author with one or two co-authors but rarely have more than three authors. Such reviews are often invited by a journal editor and written by a known expert in a discipline. Sometimes they are updates focused on recent changes in the scientific evidence base (e.g., past five years). These reviews serve as critical appraisals of the best available evidence on a specific clinical question, but they are not original research. They represent an amalgam of expert opinion from experience and expert opinion about the current scientific literature, but there is no way for readers to formally disentangle the authors’ opinions from more rigorous sources of evidence. Sometimes, there is a brief description of databases, search terms and limits used to help identify where evidence was sought. (From Annals of Emergency Medicine: http://www.annemergmed.com/content/categories and http://www.annemergmed.com/pb/assets/raw/Health%20Advance/journals/ymem/MOOSE.pdf)
  2. Systematic reviews and meta-analyses: These reviews are published as original research articles. They are the most rigorous form of review, due to their documented, replicable methodology, including the search for studies from multiple sources (electronic, print, grey literature and hand-search); unbiased review (title, abstract and full text levels) of the validity/quality of each study’s findings on a specific, clearly stated question with pre-defined eligibility criteria; data abstraction by multiple raters with means to adjudicate disagreements; systematic presentation; and synthesis of the characteristics and findings of the included studies. When systematic reviews are accompanied by a formal mathematical synthesis of the aggregate results across studies (e.g., the weighted mean effect size of a treatment), they are called meta-analyses. Systematic reviews and meta-analyses usually require a team of at least three people plus a search expert (e.g., medical informationist [formerly librarian]), with at least a year-long commitment. The Cochrane Systematic Review approach is considered the gold standard against which other systematic reviews are gauged. (Cochrane Handbook: http://handbook.cochrane.org/)
  3. Hybrid reviews (two levels)
    1. Semi-systematic review: Some journals have sought to bolster their narrative reviews by making evidence review criteria more explicit. We call these semi-systematic reviews. These review articles include a section (not necessarily methods) that provides readers the search strategy and selection criteria for included evidence. From Lancet Neurology: “This section should include a description of all the evidence that the authors considered before undertaking this study. Authors should state: the sources (databases, journal or book reference lists, etc.) searched; the criteria used to include or exclude studies (including the exact start and end dates of the search), which should not be limited to English language publications; the search terms used; the quality (risk of bias) of that evidence; and the pooled estimate derived from meta-analysis of the evidence, if appropriate.” (http://www.thelancet.com/laneur/information-for-authors/article-types-manuscript-requirements)

      Sometimes this format is used when the evidence base in an area is important but not robust enough to merit a full systematic review. From Annals of Emergency Medicine, Best Available Evidence format: “Articles that critically appraise the literature for specific emergency medicine clinical questions that lack sufficient evidence to support a formal meta-analysis or systematic review.”
    2. Contextualized systematic review with narrative component: Some journals have sought to bolster the clinical relevance of their systematic reviews by allowing authors to “fill in the gaps” where the evidence from systematic review is lacking (or where additional clinically relevant information could not realistically be included in a systematic review due to search complexity or time involved). These review articles include a robust methods section that provides the details of the search (search terms and strategy, databases and limits) and a brief description of the number of articles, their type and quality to answer a focused question. The review adheres to and documents PRISMA statement methods (http://www.prisma-statement.org/). But, unlike a formal systematic review, they do not restrict themselves to narrow answers on specific questions. Instead, they seek to review best practices, offering systematic review-level evidence where appropriate or available, and offering narrative review-level evidence otherwise. These are systematic reviews “plus.” These review types generally request that authors explicitly identify statements derived from the systematic review component, segregating them formally from the narrative/expert component of the review. (From CMAJhttp://www.cmaj.ca/site/authors/preparing.xhtml#_Research)