Evidence-Based Medicine

Describe the features of evidence-based medicine, including levels of evidence (e.g. NHMRC), meta-analysis, and systematic review

What is Evidence-Based Medicine?

  • Evidence-based medicine (EBM) is "the conscientious, explicit, and judicious use and appraisal of current best evidence in making decisions about the care of individual patients."
  • The purpose of EBM is to provide a framework for acquiring knowledge and making optimal decisions around medical care. It means integrating individual clinical expertise with the best available external clinical evidence from systematic research."

There are five stages of EBM:

  1. Ask an answerable question
  2. Search
  3. Critically appraise the evidence
  4. Integrate the evidence with the patients unique circumstances and values
  5. Evaluate the result

Levels of Evidence

Levels of evidence grade studies on liklihood of bias and internal validity. The NHMRC defines 6 levels of evidence, graded from I-IV (with three level III subtypes).

In general:

  • Level I is evidence from a systematic review of RCTs
  • Level II is evidence from at least one good RCT
  • Level III-1 is evidence from a pseudo-RCT
  • Level III-2 is evidence from a comparative study with concurrent controls, such as a cohort or case-control study
  • Level III-3 is evidence from a comparative study without concurrent controls, such as a cohort study with historical controls
  • Level IV is evidence from a case-series

Note that expert opinion is not part of NHMRC levels of evidence, though it is included on the Oxford Centre for Evidence Based Medicine system, used by the NHS.

Level Intervention Diagnostic Accuracy Prognostic Aetiology Screening
I A systematic review of level II studies A systematic review of level II studies A systematic review of level II studies A systematic review of level II studies A systematic review of level II studies
II A randomised controlled trial A study of test accuracy with: an independent, blinded comparison with a valid reference standard, among consecutive persons with a defined clinical presentation A prospective cohort study A prospective cohort study A randomised controlled trial
III-1 A pseudorandomised controlled trial A study of test accuracy with: an independent, blinded comparison with a valid reference standard, among non-consecutive persons with a defined clinical presentation All or none All or none A pseudorandomised controlled trial
III-2 A comparative study with concurrent controls A comparison with reference standard that does not meet the criteria required for Level II and III-1 evidence Analysis of prognostic factors amongst persons in a single arm of a randomised controlled trial A retrospective cohort study A comparative study with concurrent controls
III-3 A comparative study without concurrent controls Diagnostic case-control study A retrospective cohort study A case-control study A comparative study without concurrent controls
IV Case series with either post-test or pre-test/post-test outcomes Study of diagnostic yield (no reference standard) Case series, or cohort study of persons at different stages of disease A cross-sectional study or case series Case series

Grades of evidence

Evidence is graded to "indicate the strength of the body of evidence underpinning a recommendation" (e.g. in a clinical guideline).The NHMRC grades recommendations from A to D as follows:

  • A: Body of evidence can be trusted to guide practice
  • B: Body of evidence can be trusted to guide practice, in most situations
  • C: Body of evidence provides some support, but care should be taken in its application
  • D: Body of evidence is weak and recommendation must be applied with caution

Study types: Systematic Reviews and Meta-analyses

Systematic Review

Process of evaluating all of the (quality) literature to answer a specific clinical question. Does not necessarily involve statistical analysis. If it involves quantitative analysis of multiple trials, it is known as a meta-analysis.


Mathematical technique of combining the results of different trials to derive a single pooled estimate of effect. Can be done by:

  • Pooling the results of each trial
  • Pooling all of the raw data and conducting a reanalysis
  • Meta-analyses usually use random-effects models, which assumes there will be a variety of similar treatment effects
  • Individual trials are summarised with an odds ratio, and weighted, usually by sample size

Stages of a [meta-analysis] and systematic review:

  1. Inclusion and exclusion criteria are predefined
  2. Search: including online databases, reference lists, citations, and experts
  3. Validation of potentially eligible trials (critique of interval validity, i.e. trial quality)
  4. [Heterogeneity Analysis]
  5. [Meta-analysis]
  6. Reliability of result determined
    i.e. Consistency accross studies, statistical significance, large effect size, biological plausibility.
  7. Sensitivity analysis
    Repeating the analysis with an alternative model, excluding borderline trials or outliers. If the result is unchanged, then the findings are robust.


For the pooling of results to be valid, the trials need to be similar. Differences between trials is called heterogeneity, and is important because:

  • Heterogeneity analysis affects the type of model that can be used (fixed or mixed effects)
  • Highly heterogenous data is not appropriate for meta-analysis.

Heterogeneity is divided into:

  • Statistical Heterogeneity
    The effects of the intervention are more different than would be expected to occur through chance alone.
  • Clinical Heterogeneity
    Due to trial design it would be inappropriate to pool the results.
    • E.g., conducting a meta-analysis on the effects of the same drug in a paediatric and adult population may be inappropriate, as these two trials had different inclusion criteria.
  • Methodological Hetreogeneity
    Where the methods used in different trials are too different to allow pooling of the data.

Forest Plots

Results of meta-analyses are presented in a blobbogram, or more boringly, a Forest Plot.


  • The x-axis plots the odds ratio, remembering that an OR of 1 indicates no difference
  • The y-axis lists the studies included, and the overall summary statistic
  • The dot (or square) indicates the point estimate (from its x-location) and the weight given to the study (by its size)
  • The horizontal line indicates the upper and lower bounds of the confidence interval
  • The diamond indicates the overall point estimate and (by its width) the confidence interval for the point estimate
  • The result of the heterogeneity test should also be displayed
    P < 0.1 indicates significant heterogeneity.

Funnel Plots

Funnel plots are a graphical tool to detect publication bias.

  • Due to statistical power, larger studies should be a closer representation of the true effect
  • Therefore, when evaluating an number of studies, one would expect that large studies cluster around the 'true effect', and smaller studies to scatter further
  • A graph is then plotted of OR on the x-axis, and standard error on the y-axis
    • Publication bias is suggested when results cluster on one side of the funnel plot
    • No evidence of publication bias would have studies clustered around the true effect

Strengths and weaknesses of meta-analyses

Strengths Weaknesses
Enhanced precision of estimates of effect Publication bias
Useful when large trials have not been done or are not feasible Duplicate publication
Generate clinically relevant measures (NNT, NNH) Heterogeneity
Inclusion of outdated studies

Because of these weaknesses:

  • Positive meta-analyses should be considered largely hypothesis-generating, and should be confirmed by (a large) RCT
  • Negative meta-analyses can probably be accepted


  1. Sacket DL, Richardson WS, Rosenberg W, Haynes RB. Evidence-based Medicine: How to practice and teach EBM. Churchill Livingstone, London 1997.
  2. Sackett David L, Rosenberg William M C, Gray J A Muir, Haynes R Brian, Richardson W Scott. Evidence based medicine: what it is and what it isn't BMJ 1996; 312:71.
  3. NHMRC. NHMRC additional levels of evidence and grades for recommendations for developers of guidelines. National Health & Medical Research Council. 2009.
  4. Myles PS, Gin T. Statistical methods for anaesthesia and intensive care. 1st ed. Oxford: Butterworth-Heinemann, 2001.
  5. Lalkhen AG, McCluskey A. Statistics V: Introduction to clinical trials and systematic reviews. CEACP 2008.
Last updated 2017-10-03

results matching ""

    No results matching ""