Level of evidence

from Wikipedia, the free encyclopedia

Evidence class is a term used in evidence-based medicine (EbM) that describes and categorizes the formal and content quality of a clinical study . It describes a hierarchy of evidence.

Simplified classification

This classification is the most important one for non-doctors, and it has an important function , especially in the emergency services . The individual classes are:

  • Class I: according to the current study situation, is considered to be a reliable recommendation, always acceptable, safe and effective
  • Class IIa: is considered to be acceptable and sensible according to the current study situation (very well proven advantage)
  • Class IIb: according to the current study situation, is considered to be optionally applicable (well-proven advantage, therapy alternative for experienced people)
  • Class X: is considered indeterminable according to the current study situation; no current scientific evidence or subject to further investigation, or proven benefit or potential harm
  • Class III: according to current studies, it is considered not indicated, not useful, possibly harmful

External Evidence Hierarchy

EBM is not limited to randomized controlled trials (abbreviation: RCT s by English R andomized C ontrolled T rial ) and meta-analyzes limited. Nevertheless, these are to be regarded as the gold standard in the great majority of those questions that are about evaluating the benefits and risks of new and old therapies - especially in the area of ​​pharmacotherapy.

According to the recommendations of the AHRQ ( Agency for Healthcare Research and Quality ), a distinction is made between evidence classes I to IV. Class Ia studies have the highest evidence, class IV studies the lowest. The higher the evidence class, the better the scientific justification for a therapy recommendation derived from it.

  • Level Ia: At least one meta-analysis based on methodologically high-quality randomized, controlled studies
  • Level Ib: at least one sufficiently large, methodologically high-quality RCT
  • Level IIa: at least one high-quality study without randomization
  • Level IIb: at least one high-quality study of a different type, quasi-experimental study
  • Level III: more than a methodologically high-quality non - experimental study such as comparative studies, correlation studies or case-control studies
  • Level IV: opinions and beliefs of respected authorities (from clinical experience); Expert commissions; descriptive studies
  • Level V: case series or one or more expert opinions

Individual evidence

  1. ↑ Archive link ( Memento from September 27, 2013 in the Internet Archive ) From evidence to recommendation (classification systems) The German Cochrane Center
  2. ↑ Archive link ( Memento from November 8, 2014 in the Internet Archive ) p. 2.
  3. ↑ [1] On the methodology used to draw up the evidence-based diabetes guidelines of the DDG. Accessed October 12, 2010.