- Published
- Updated
Detailed, well-organised reporting does not establish that a study is free from bias. Equally, an unreported detail does not prove that the method was incorrect. Appraisal should distinguish identified methodological risks from uncertainty caused by missing information.
Scope of the decision
Examine selection, comparison, measurement and follow-up according to the relevant study design. One total score can allow good features to obscure an important weakness. Select appraisal criteria suitable for the question; a simple quality label should not automatically establish support for a device claim.
Worked review
For example, a clearly reported observational study shows that device choice was related to disease severity. The team does not accept it as strong comparative evidence merely because reporting is thorough. It considers confounding and the limits of adjustments with appropriate methodological expertise.
Practical checks
- Record identified methodological risks separately from unreported information.
- Keep important bias domains visible rather than hiding them in a total score.
- Explain how each relevant risk could affect the clinical conclusion.
Acceptance and limits
Acceptance requires traceable connections between strengths, uncertainty and the final interpretation. A problematic study may contribute to a limited question; it need not be classified as either entirely useless or wholly reliable. Justify the overall conclusion using available data and appropriate scientific judgement.
Frequently asked questions
- Does peer review mean an absence of bias?
- No. It does not remove the need for separate methodological appraisal.
- Does missing information automatically mean poor quality?
- It can increase uncertainty, but is different from a known error and needs an appropriate explanation.