Table II.
| Evidence level (quality of evidence) | |
|---|---|
| High | One or more high-quality, well-conducted randomised controlled trials (RCTs) that provide consistent and directly usable results. This means that further research is very unlikely to affect the estimated effect. |
| Moderate | RCTs but with important limitations (i.e. biased assessment of treatment effect, high patient loss during follow-up, lack of blinding, unexplained heterogeneity), indirect evidence from similar (but not identical) study populations and studies with very small numbers of patients or observed events (endpoints). In addition, there is evidence from well-designed non-randomised controlled studies, well-designed analytical cohort studies or from case-control studies, and from multiple case series with or without intervention. This means that further research is likely to have an important impact on the estimated effect and could alter it. |
| Low | Observational studies, typically of low quality due to error risk. This means that further research will almost certainly have a significant impact on the estimated effect and will most likely alter it. |
| Very low | The evidence is contradictory, of poor quality or lacking in results, so that the benefits and risks ratio cannot be determined. This means that any estimated effect is very uncertain as evidence or is unavailable or does not allow to formulate any conclusions. |