Oxford Centre for Evidence-Based Medicine 2011 Levels of Evidence

Evidence-based medicine (EBM) is the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients. The practice of EBM requires integrating individual clinical expertise with the best available external clinical evidence from systematic research. The Oxford Centre for Evidence-Based Medicine (CEBM) developed a widely recognized hierarchy of evidence to help clinicians and researchers evaluate the strength of research findings.

The original CEBM Levels of Evidence were first published in 1998 and have been updated several times. The 2011 version provides the most current framework for assessing evidence quality across different types of clinical questions. This system categorizes evidence from Level 1 (highest quality) to Level 5 (lowest quality), with specific criteria for different categories of clinical questions.

The Purpose of Evidence Levels

Levels of evidence serve multiple purposes in healthcare:

• Standardizing the assessment of research quality across different studies
• Guiding clinical decision-making by highlighting stronger evidence
• Helping researchers design more rigorous studies
• Enabling policymakers to develop evidence-based guidelines
• Facilitating the translation of research into practice
• Assisting in critical appraisal of published research

The 2011 Oxford CEBM Levels of Evidence

The Oxford CEBM 2011 Levels of Evidence consider study design, bias, precision, and directness when categorizing research. Importantly, the levels differ for questions about therapy, prevention, etiology/harm, diagnosis, prognosis, and economic analysis.

Level	Therapy/Prevention	Diagnosis	Prognosis	Economic Analysis
1a	Systematic review of RCTs with homogeneity	Systematic review of Level 1 diagnostic studies	Systematic review of inception cohort studies	Systematic review of Level 1 economic studies
1b	Individual RCT with narrow confidence interval	Validating cohort study with good reference standards	Individual inception cohort study with >80% follow-up	Analysis based on clinically sensible costs or alternatives
1c	All or none case series	Absolute SpPins and SnNouts	All or none case series	Absolute better-value/worse-value analysis
2a	Systematic review of cohort studies	Systematic review of Level 2 diagnostic studies	Systematic review of retrospective cohort studies	Systematic review of Level 2 economic studies
2b	Cohort study or low-quality RCT	Exploratory cohort study with good reference standards	Retrospective cohort study	Analysis based on clinically sensible costs or alternatives
2c	Outcomes research	Ecological studies	Outcomes research	Audit or outcomes research
3a	Systematic review of case-control studies	Systematic review of Level 3 diagnostic studies	Systematic review of case-control studies	Systematic review of Level 3 economic studies
3b	Case-control study	Non-consecutive study, or without consistently applied reference standards	Case-control study	Analysis based on limited alternatives or costs, poor quality estimates
4	Case series or poor-quality cohort/case-control studies	Case-control studies, poor or non-independent reference standard	Case series or poor-quality prognostic cohort studies	Analysis with no sensitivity analysis
5	Expert opinion without explicit critical appraisal	Expert opinion without explicit critical appraisal	Expert opinion without explicit critical appraisal	Expert opinion without explicit critical appraisal

Understanding Each Level

Level 1 Evidence

For therapy/prevention, Level 1a evidence consists of systematic reviews of randomized controlled trials (RCTs) with homogeneity (similar results across studies). Level 1b comprises individual RCTs with narrow confidence intervals. Level 1c refers to "all or none" clinical situations, where all patients died before the intervention became available, but some now survive with it, or where some patients died before the intervention, but none die with it.

For diagnosis, Level 1 evidence includes systematic reviews of Level 1 diagnostic studies, individual validating cohort studies with consistently good reference standards, and diagnostic tests meeting SpPin (Specificity-Positive results rule in disease) or SnNout (Sensitivity-Negative results rule out disease) criteria.

Level 2 Evidence

For therapy/prevention, this includes systematic reviews of cohort studies, individual cohort studies (including RCTs with less than 80% follow-up), and outcomes research. These observational studies can provide valuable evidence when RCTs are not feasible or ethical.

For diagnosis, Level 2 includes systematic reviews of Level 2 diagnostic studies and exploratory cohort studies with good reference standards. These studies typically involve patients with suspected disease who all undergo both the diagnostic test and the reference standard.

Level 3 Evidence

For therapy/prevention, this includes systematic reviews of case-control studies and individual case-control studies. These studies are particularly useful for examining rare outcomes or long-term effects.

For diagnosis, Level 3 includes systematic reviews of Level 3 diagnostic studies and non-consecutive case studies without consistently applied reference standards. These study designs are more vulnerable to bias but may still offer important insights.

Level 4 Evidence

For therapy/prevention, this includes case series and poor-quality cohort or case-control studies. Case series track patients with a known exposure or disease but lack a comparison group, making it difficult to establish causality.

For diagnosis, Level 4 includes case-control studies with poor or non-independent reference standards, where the interpretation of the diagnostic test may be influenced by knowledge of the disease status.

Level 5 Evidence

For all clinical question types, Level 5 evidence consists of expert opinion without explicit critical appraisal, or based on physiology, bench research, or first principles. While expert opinion is valuable in guiding practice when higher-level evidence is unavailable, it is inherently subjective and vulnerable to bias.

Practical Application of the Levels

The Oxford CEBM Levels of Evidence provide a framework for evaluating research, but they should be applied thoughtfully:

• Question type matters: Different clinical questions require different study designs. A well-conducted cohort study might provide stronger evidence for prognosis than a poorly conducted RCT.
• Levels are guides, not rules: Sometimes lower-level evidence may be more relevant to a specific clinical question than higher-level evidence.
• Evolution of evidence: What was considered Level 1 evidence yesterday might be challenged by new Level 1 research today. Evidence is dynamic and constantly evolving.
• Context is crucial: Applicability to specific patients and settings must always be considered, even with the highest level evidence.
• Quality within levels: Not all studies within a level are equal critical appraisal remains essential.

Limitations of Evidence Hierarchies

While evidence hierarchies like the Oxford CEBM Levels are valuable tools, they have important limitations:

• They may undervalue certain study designs, such as qualitative research that addresses patient preferences and experiences
• Hierarchies focus on study design and don't adequately account for study quality within design types
• Real-time evidence and case reports can be crucial for emerging conditions or rare adverse events
• Not all clinical questions can be answered by the top levels of evidence
• The system may not adequately evaluate complex interventions or whole systems approaches
• Evidence levels don't directly account for effect size or clinical significance
• The hierarchy can create the false impression that high-level evidence equates to high-quality evidence

Conclusion

The Oxford Centre for Evidence-Based Medicine 2011 Levels of Evidence provide a valuable framework for evaluating the quality of research evidence across different types of clinical questions. By understanding these levels, clinicians can more critically appraise research findings and make better-informed decisions about patient care.

However, these levels should be used alongside clinical expertise and patient values to achieve true evidence-based practice. Evidence hierarchies are tools to aid decision-making, not rigid rules that should override clinical judgment. The most effective clinicians combine the best available evidence with their professional expertise and an understanding of patient preferences and circumstances to deliver optimal care.

As research methodologies evolve and healthcare becomes even more patient-centered, our approaches to evaluating evidence may continue to change, but the fundamental principles of EBMintegrating evidence, expertise, and patient valuesremain essential to providing high-quality care.