Evidence-based medicine and levels of evidence

Evidence-based medicine is an approach to medical practice that aims to optimise decision-making by emphasising the use of evidence from well-designed and well-conducted studies. Although all of medicine is based on science and has some degree of empirical support, evidence-based medicine goes further still, ranking the evidence by its epistemological strength and requiring that only the strongest types of study be regarded as capable of generating strong recommendations. Weaker types of study can yield only weak recommendations.
Levels of evidence
Level I Systematic Reviews,
Meta-Analyses
Level II Randomised Controlled
Double Blind Studies
Level III Cohort Studies
Level IV Case Control Studies
Level V Case Series
Level VI Case Report
Level VII Ideas, Opinions, Editorials
Level VIII Animal Research
Level IX In vitro research
In recent years, therefore, an international classification of scientific studies has been drawn up, beginning with those of least strength and rising to the top of the pyramid, which represents systematic reviews and meta-analyses of multiple randomised controlled double blind studies. The latter are considered to carry the greatest scientific weight and are the ones that determine the recommendations in guidelines across every medical specialty.
In using the levels of evidence, however, scientists should be aware that they are a guide to the quality of studies. Assessing them in depth requires a critical appraisal of each individual study. Randomised clinical trials are also not always feasible, and level III or even level IV studies may likewise be of considerable value to the clinician. Furthermore, everyday medical practice shows that the answer to a clinical question should rest on a combined assessment of all the evidence of every type, and a single study cannot produce a definitive answer.
Source: Wright JG, Swiontkowski MF, Heckman JD. Introducing levels of evidence to the journal. J Bone Joint Surg Am. 2003 Jan;85-A(1):1-3.