A literature review focused on a research question that tries to identify, appraise, select and synthesize all high quality research evidence relevant to that question
A review can be defined as, any attempt to synthesise the results and conclusions of two or more publications on any given topic.
Reviews are useful because of the sheer volume of biomedical literature that is now available. However, they have their limitations, a study by Mulrow of 50 reviews in the four of the major medical journals found that 49 had no statement of the methods used and 47 had inappropriate summaries of the information included.
Current medical reviews do not routinely use scientific methods to identify, assess and synthesise information.
C.D. Mulrow, 1987
These ‘traditional’ or narrative reviews have a number of biases. There is normally the personal bias of the author/s, a bias in the selection of included material, and with no clear methodology they cannot be reproduced independently, so assumptions cannot be easily verified.
In contrast a systematic review strives to comprehensively identify and track down all the available literature on a topic with clear comprehensive methodology described. The best examples of these are the Cochrane reviews.
The main stages in the systematic review process are :
- Report all relevant primary studies
- Assess methodological quality
- Identify common definitions for outcomes
- Extract estimates of outcomes
- Meta-analysis where appropriate
- Narrative summary where data sparse or of too low quality
- Explore robustness of results
- Clear presentation of key aspects
- Appraisal of methodological limitations of primary studies and systematic review.
When a specific statistical strategy for combining the results of studies included in a systematic review is conducted this is termed a meta-analysis. The key differences between narrative and systematic reviews are shown below.
|Feature||Narrative review||Systematic Review|
|Sources & Search||Not usually specified||Comprehensive & explicit|
|Synthesis||Often qualitative summary||Quantitative summary|
|Inferences||Sometimes evidence-based||Usually evidence-based|
- Mulrow CD. The medical review article: state of the science. Ann Intern Med. 1987;106:485-8. (click here for the abstract)
- AMSTAR – Assessing the methodological quality of systematic reviews
- Cochrane Reviewers’ Handbook
- MOOSE – Reporting meta-analysis of observational studies in epidemiology
- PRISMA – (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) the PRISMA Statement is an update and expansion of the now-out dated QUOROM Statement.
- PROSPERO – prospective register of systematic reviews
- QUADAS – quality assessment tool for use in systematic reviews of diagnostic accuracy studies
- Undertaking Systematic Reviews of Research on Effectiveness – CRD University of York
Systematic reviews in health care from The British Medical Journal 2001
( In order to access these BMJ articles you will need to register with the BMJ )
- Assessing the quality of controlled clinical trials
- Investigating and dealing with publication and other biases in meta-analysis
- Systematic reviews of evaluations of diagnostic and screening tests
- Systematic reviews of evaluations of prognostic variables
Meta-Analysis Series from The British Medical Journal 1997
( In order to access these BMJ articles you will need to register with the BMJ
- Meta-analysis: Potentials and Promise
- Meta-analysis: Principles and Procedures
- Meta-analysis: Beyond the Grand Mean
- Bias in Location and Selection of Studies
- Spurious Precision: Meta-analysis of Observational Studies
- Meta-analysis: Unresolved Issues and Future Developments
Annals of Internal Medicine Articles 1997
( In order to access these articles you will need to subscribe to teh journal)
- Systematic Reviews: Synthesis of Best Evidence for Clinical Decisions
- Using Numerical Results from Systematic Reviews in Clinical Practice
- The Relation between Systematic Reviews and Practice Guidelines
- Quantitative Synthesis in Systematic Reviews