Skip to Main Content

Physical Therapy

Levels of Evidence

This is an evidence base pyramid.  It is often used as a reference to illustrate evidence based literature.  Before starting a literature search, use this illustration as a reference on where to begin.  Beginning at the top will give you the highest quality of literature. However, the source of evidence you choose will also be determined by the question you want to answer and the study's methodological quality.  For instance, if you want to answer a foreground question about effectiveness of an intervention you will start with synthesized information (from the top of the pyramid) followed by recent experimental studies not yet synthesized by any systematic review and/or CPG.

evidence-based pyramid

 

Filtered and Synthesized Resources:

When searching for evidence-based information, it is suggested to begin by searching for the highest level of evidence possible.  This includes, systematic reviews  or without meta-analyses and critically appraised topics or articles. Filtered resources have gone through the process of a systematic search and appraisal of the evidence.  They have "filtered" and synthesized available evidence to best answer a clinical question or practice issue.

Unfiltered and Unsynthesized Resources

Information that has not been critically appraised is considered "unfiltered".  If you do not find an appropriate answer in the filtered resources or if there is insufficient evidence to answer your question, it is recommended to locate original research studies to answer your question. 

Unfiltered resource are individual articles that present different levels of research such as:

  • Randomized Control Trials (RCT)
  • Cohort Studies
  • Case Control Studies

With unfiltered resources, it is up to the reader to determine the quality of evidence.

Type of Question Type of Study
Therapy/Treatment Randomized control trial (RCT) > case control > case series
Diagnosis Prospective studies > comparison to a gold standard
Prognosis Cohort studies > case control > case series
Harm/Etiology RCT > cohort studies > case-control > case series
Prevention RCT > cohort studies > case-control > case series
Quality Improvement RCT