Levels of evidence and grades of recommendation

Sources of evidence in PROSPECT

The evidence for prospect is derived from three separate sources, and this evidence is considered by the prospect Working Group to determine the prospect recommendations:
? Procedure-specific evidence derived from the systematic reviews of the literature
p Transferable evidence from comparable procedures identified by the members of the prospect Working Group
¢ Current practice – A commentary on each of the interventions from the members of the prospect Working Group
u Practical prospect recommendations are based on all the information

PROSPECT grades of recommendation

The recommendations of the PROSPECT Working Group are graded A–D, based on the level of evidence from the studies, which is in accordance with the Oxford Centre for Evidence-Based Medicine (CEBM website accessed Dec 2003, Sackett 2000) (see table below) (http://www.cebm.net)

In the context of PROSPECT, grades of recommendation are dependent on whether the evidence is from specific studies, transferable studies or clinical practice:

? Specific evidence – grade A
p Transferable evidence – grade B/C
¢ Clinical practice – grade D

CEBM grades of recommendation

Study criteria	Level of evidence	Criteria for grading of recommendation	Grade of recommendation
Systematic review (with homogeneity) of randomised, controlled trials	1a	Consistent level 1 studies	A
Individual, randomised, controlled trials with statistically significant results	1b
All or none, i.e. prior to availability of new therapy, all died, now with therapy some survive; or, prior to therapy some died, now with therapy none die	1c
Systematic review (with homogeneity) of cohort studies	2a	Consistent level 2 or 3 studies (or extrapolations* from level 1 studies)	B
Individual cohort study (including low quality randomised controlled trial, e.g. <80% follow up)	2b
Outcomes research	2c
Systematic review (with homogeneity) of case-controlled studies	3a
Individual case-controlled study	3b
Case-series, and poor quality cohort and case-controlled studies	4	Level 4 studies (or extrapolations* from level 2 or 3 studies)	C
Expert opinion without explicit critical appraisal, or based on physiology, bench research or first principles	5	Level 5 evidence (or troublingly inconsistent or inconclusive studies of any level)	D
*Extrapolations are where data are used in a situation that has potentially clinically important differences to the original study situation. In the case of PROSPECT, extrapolation largely refers to transferable evidence.