Monday, September 21, 2009

Checklist of Questions For Critical Appraisal

To be able to do a good critical review is very essential in this era where evidence-based medicine is getting more and more important. There are so many journals, so many articles that sometimes we can get lost in the jungle mess of the information that is available online and in print. To be able to select which article to read, to be able to critically appraise it, therefore, is important. I have here listed the searching questions that are important to ask ourselves whenever we read or review or appraise and article or even when we are asked to review a manuscript.

Overall Originality and Importance of the Paper

- Is the work original?
- Is there a clear overall message conveyed by the paper?
- To what extent do the authors' observations further our knowledge?
- How important are the observations? Does the message (or new knowledge) really matter?

- Is there a succinct introduction which is helpful to readers unfamiliar with the subject?

- What is the rationale of the study?

- Were the following clearly stated/addressed:
1. Research question?
2. Clinical relevance?
3. Hypothesis/theory stated (if appropriate)?
5. Key terms well defined?

- Does the research intend to add new information to research community?

Methods What kind of study design is it?
Was the design appropriate for the question being asked?
Sample population:
- Appropriate?
- Group similar? (age, gender, etc)
- Inclusion and exclusion criteria stated and appropriate?
- Co-morbidity of subjects addressed?
- Any other extraneous variables identified?
- Randomly assigned? Was this done appropriately?
- Were all the subjects at the start accounted for at the study’s end?

Treatment intervention?
- Evidence based?
- Good rationale?
- Clearly described?
- Intervention manipulated correctly?
- Easily reproduced in clinical setting?
- Consistency with treatment/intervention?
- Setting appropriate?
- Control for confounding variables?-

Evaluators and treatment administration
- blinded?
- Inter-rater and intra-rater reliability?

- Clear? Appropriate?
- How large was the treatment effect?
- How precise was the estimate of treatment effect? (Look at confidence intervals, standard errors)
- Clinical and statistical significance reported?

Tables and Figures
- Tables/figures easy to interpret? Appropriate? Are all the tables and figures essential?
- Is each table and figure clearly presented and appropriately labelled?
- Is there undue repetition of the results?

Discussion Inferences/conclusions make sense or coherent?
Answer research question, address original hypothesis?
Are data referred to which should have been mentioned in the results section?
Identify limitations of study?
Concise? Clear?
Can the results be applied to local population?
Were all clinically important outcomes considered?

Conclusion Key findings presented?
Conclusion for each objective
Implications discussed?
Future research? Realistic?

In summary, ask:
Why did they start? (Rationale)
What did they do? (Validity)
What did they find? (Results)
What does it mean? (Relevance)

No comments:


PLEASE NOTE: All contents in this blog are copyrighted materials, unless otherwise stated. Even if you encounter materials in this page without a copyright notice, it does not mean that it is not copyrighted (Click here to read TEN BIG myths on copyright explained). This is especially so as most nations are signatories of the Berne Convention on international copyright law (World Intellectual Property Organization). Nevertheless, I have licensed almost all the materials contained here under Creative Commons licenses strictly for educational, non-commercial purposes only. Kindly email me at should you want to use any of the materials for commercial purposes. Thank you.