Appraising information: going deeper
To critically appraise is to essentially dissect published research, usually a journal article, for its validity eg is it biased, and relevance. To aid with this it is helpful to use a checklist or a scale.
There are a number of resources to help you critically evaluate a paper or article or research.
One of the best:
- Trisha Greenhalgh has written the very useful How to read a paper. Contents include: Assessing methodological quality; Statistics for the non-statistician; Papers that report trials of drugs, complex interventions, diagnostic or screening tests, summarise other papers (systematic reviews and meta-analyses), tell you what to do (guidelines), tell you what things cost (economic analyses) and more. Note there are earlier editions including an ebook of the 5th ed 2019.
Other useful resources include:
- How do you know a paper is legit? from the University of Washington.
- Try the Understanding health research tool. It will 'guide you through a series of questions to help you to review and interpret a published health research paper'.
- Look at Retraction watch.
7 videos on critically appraising various types of studies from the Cochrane Common Mental Disorders group
'This playlist includes seven modules that address critical appraisal concepts and methods for six different research designs. In each video, we walk through the CASP checklist alongside an open access research article, to demonstrate how to approach critical appraisal in practice.'
|
Appraisal tools:
- If you are critically appraising evidence based research remember you can use the GATE model.
- There are also lots of help sheets at the:
- Critical Appraisal Skills Programme (UK) website.
- Critical Appraisal Tools and Worksheets (CEBM)
- Critical appraisal tools from the Joanna Briggs Institute
- DISCERN Instrument
- Duke University Medical Center Library has a number of worksheets for a variety of study types.
- Johns Hopkins Nursing Evidence Based Practice: Model & Guidelines includes tools to appraise a variety of studies
- Appendix H Quality appraisal checklist – qualitative studies in Methods for the development of NICE public health guidance (third edition) Published date: September 2012
- The EBM Toolkit in BMJ Best Practice has a selection of critical appraisal checklists and statistics calculators
- There are a bunch of 'named' tools:
- QUADAS - quality assessment tool for diagnostic accuracy studies
- QAREL - for diagnostic reliability
- COSMIN - outcome measurement instruments
- ROBINS-I - risk of bias in non randomised studies
- Newcastle-Ottawa Scale for non-randomised studies in meta-analyses
- ROB2 - Cochrane tool for randomised trials - more information
- Jadad scale for clinical trials
- AMSTAR 2 - 'critical appraisal tool for systematic reviews that include randomised or non-randomised studies of healthcare interventions, or both.' See also AMSTAR website.
- SANRA-a scale for the quality assessment of narrative review articles.
- EQUATOR network lists a variety of named tools
- The NHMRC Guidelines for Guidelines has a section on Assessing risk of bias including named tools
- The OCEBM Levels of Evidence (Oxford Centre for Evidence Based Medicine) has a table of hierarchies of evidence and additional levels of evidence information. It also has EBM tools including information on study designs.
- The GRADE (Grading of Recommendations, Assessment, Development and Evaluation) Working Group on evaluating guidelines has useful information including videos in its website.
- McMaster University has additional information, including videos, and help for GRADE.
- See Cochrane Methods GRADEing to evaluate the quality of systematic reviews and
- The Cochrane GRADE approach to evaluating the quality of evidence: pathway
- Another appraisal instrument for guidelines is Appraisal of Guidelines Research and Evaluation (AGREE II)
- The NHMRC Guidelines for Guidelines has a section on Assessing evidence of certainty
Reporting guidelines and frameworks
Using a guideline or framework can help researchers with the quality of research and keep the process and methods transparent.
The EQUATOR network website has a list of guidelines for main study types eg randomised controlled trials, observational studies, qualitative research, prediction models and more. See also ENTREQ for qualitative research.
See also:
- Cultural Identity Interventions Systematic Review Proforma. For development of the proforma see MacLean S, Ritte R, Thorpe A. Assessing compliance with Aboriginal and Torres Strait Islander health research guidelines within systematic reviews. Australian Indigenous Health Bulletin 2015; 15. http://healthbulletin.org.au/articles/assessing-compliance-with-aboriginal-and-torres-strait-islander-health-research-guidelines-within-systematic-reviews/
- ARRIVE - Animal Research: Reporting of in vivo experiments
Statistics
- The Medical Center Library at Duke University has a variety of resources to help you calculate results
- Free statistical calculators from MedCalc
- Know your chances: Understanding health statistics
- The National Collaborating Centre for Methods and Tools (Canada) has videos on absolute and relative risk, confidence intervals, forest plots, odds ratios and clinical significance
- NNT and risk assessments for therapies
Definitions
- The Whatis series has brief information on eg NNTs, confidence intervals. Bandolier has similar information including an EBM glossary.
Beware of pitfalls
in the way research results are communicated. For an example see
- Ben Goldacre's Ted talk on What doctors don't know about the drugs they prescribe.
- Kupferschmidt K. Tide of lies. Science. 2018;361(6403):636-641. doi: 10.1126/science.361.6403.636
- Bolland MJ, Avenell A, Gamble GD, Grey A. Systematic review and statistical analysis of the integrity of 33 randomized controlled trials. Neurology.87(23):2391-402
Read also the follow up editors' note and author response. - Criado-Perez C. Invisible women : exposing data bias in a world designed for men. London : Chatto & Windus. 2019
- Carlisle JB. Data fabrication and other reasons for non-random sampling in 5087 randomised, controlled trials in anaesthetic and general medical journals. Anaesthesia. 2017;72(8):944-52.
- Dr Andrew Klein: Catching the fraudsters in medical research.
Source: RNZ National Sunday Morning 7.18 August 25th 2019.
Edit page