Dr Oliver Finlay
Medical peer-reviewed journal articles play a crucial role in helping us to evaluate evidence-based research to inform healthcare decisions. Healthcare professionals must read and rank evidence effectively to make informed clinical decisions and similarly, as we discuss scientific findings related to brain health, mindfulness, and flow states, we must consider how valid and reliable the research is.
In this article, I will explore four sources on how to read and rank evidence in medical peer-reviewed journal articles, including "How to Read a Paper" by Trisha Greenhalgh, the "Oxford 2011 Levels of Evidence" by the Oxford Centre for Evidence-Based Medicine, "Users' Guides to the Medical Literature: IX" by Guyatt et al. and “A new system for grading recommendations in evidence-based guidelines” by Harbour and Miller.
Greenhalgh (2000) provides a six-step framework for reading and evaluating medical papers. She recommends:
1) starting by identifying the type of article being reviewed,
2) assessing the validity of the study design to critically appraising the methodology of the study, including any potential biases,
3) examining the size and precision of the study, including the sample size, effect size, and confidence intervals,
4) evaluating the relevance of the study to the clinical question being addressed, considering whether the study's population, intervention, comparison, and outcomes match the patient's characteristics and clinical question,
5) analysing the overall evidence and considering any potential limitations of the study,
6) applying the results of the study to the patient's needs, including considering their preferences, values, and circumstances.
The "Oxford 2011 Levels of Evidence" provide a standardised approach to ranking evidence (Oxford Centre for Evidence-Based Medicine [OCEBM], 2011). The OCEBM proposes a hierarchy of evidence ranging from level 1 (highest quality) to level 5 (lowest quality). The levels of evidence are based on the study design, methodology, and potential for bias.
LEVEL 1 evidence consists of systematic reviews of randomized controlled trials (RCTs) or individual RCTs with narrow confidence intervals.
LEVEL 2 evidence includes individual cohort studies or low-quality RCTs.
LEVEL 3 evidence includes case-control studies or non-systematic reviews.
LEVEL 4 evidence includes case series or expert opinion.
LEVEL 5 evidence includes anecdotes and editorials.
The OCEBM (2011) recommends using the appropriate level of evidence for the clinical question being addressed. For example, high-quality RCTs are best suited for questions about therapeutic interventions' efficacy and safety, while case series or expert opinion may be appropriate for rare or unique clinical scenarios.
“Users' Guides to the Medical Literature: IX," (Guyatt et al., 1995) provides a framework for grading health care recommendations based on the quality of evidence and the balance of benefits and harms. Guyatt et al. (1995) propose that recommendations should be based on a careful and transparent assessment of the quality of the evidence, including the study design, methodology, and potential for bias.
Guyatt et al. (1995) suggest that the strength of a recommendation should be based on the balance of benefits and harms, including the severity of the disease or condition, the patient's values and preferences, the potential impact of the intervention, and the cost-effectiveness of the intervention. The author provide a framework for grading evidence based on randomised controlled trials (RCTs), observational studies, and expert opinions. They recommend using a four-point scale (strongly recommend, recommend, suggest, and not recommend) to grade recommendations based on the evidence.
"A New System for Grading Recommendations in Evidence-Based Guidelines" (Harbour and Miller, 2001) proposes an evidence-based approach to grading recommendations in clinical guidelines. The authors propose a three-tiered approach to grading recommendations based on the strength of the evidence and the balance of benefits and harms. The three levels of recommendation are strong, conditional, and weak.
The quality of the evidence is assessed based on the study design, risk of bias, precision, and consistency of the results. The balance of benefits and harms is assessed based on the magnitude of the effect, the number needed to treat or harm, and the severity and frequency of adverse effects. The strength of the recommendation is based on the importance of the problem, the values and preferences of the patient, the cost and resource implications, and the feasibility of implementation.
For a strong recommendation, the authors suggest that the benefits of the intervention clearly outweigh the harms, and there is a high degree of certainty about the quality of the evidence supporting the recommendation. A strong recommendation is based on high-quality evidence from well-designed studies with a low risk of bias, consistent results, and a precise estimate of the effect. The authors suggest that strong recommendations should be implemented uniformly, and alternative approaches should not be considered.
For a conditional recommendation, the authors suggest that the balance of benefits and harms is less clear, and the quality of the evidence is not as strong. A conditional recommendation is based on lower quality evidence, with greater uncertainty about the balance of benefits and harms. The authors suggest that conditional recommendations should be individualized based on patient preferences and values, and healthcare professionals should consider alternative approaches.
For a weak recommendation, the authors suggest that the benefits and harms of the intervention are unclear, and there is a high degree of uncertainty about the quality of the evidence. A weak recommendation is based on very low-quality evidence from studies with a high risk of bias or inconsistent results. The authors suggest that weak recommendations should be used cautiously and in the context of a shared decision-making process with the patient.
Overall, Harbour and Miller's approach emphasizes the importance of assessing the quality of evidence, the balance of benefits and harms, and the strength of the recommendation when grading recommendations in evidence-based guidelines. The authors suggest that their approach provides a more nuanced and transparent approach to grading recommendations and can help healthcare professionals make informed clinical decisions.
In conclusion, reading and ranking evidence in medical peer-reviewed journal articles requires a critical appraisal of the study design, methodology, potential biases, and relevance to the clinical question being addressed.
REFERENCES
Greenhalgh T. (2000). How to read a paper. Second edition. London: BMJ Books
Guyatt GH, Sackett DL, Sinclair JC, et al. (1995). Users' Guides to the Medical Literature: IX. A Method for Grading Health Care Recommendations. JAMA. 274(22):1800–1804
Harbour R, Miller J. (2001). A new system for grading recommendations in evidence-based guidelines. BMJ. 323(7308):334-6. doi: 10.1136/bmj.323.7308.334.
OCEBM Levels of Evidence Working Group (2011). The Oxford 2011 Levels of Evidence. Oxford Centre for Evidence-Based Medicine.
Comments