Chest
Clinical Recommendations Using Levels of Evidence for Antithrombotic Agents
Section snippets
PRIMARY STUDIES AND META-ANALYSIS
Rigorously designed double-blind randomized trials with high rates of successful follow-up offer the most valid evidence of treatment efficacy and thus are appropriate for application to individual patients (Table 1).6,7 Individual randomized controlled trials (RCTs), however, are often limited by inadequate sample size that leaves them open to missing important differences between treatment groups. A small study, even if it shows a difference between treatment and control groups, is likely to
LEVELS OF EVIDENCE
Table 3 summarizes the classification of levels of evidence that we describe in the following sections.
Level I: Randomized Trials or Meta-Analyses in Which the Lower Limit of the CI for the Treatment Effect Exceeds the Minimal Clinically Important Benefit
Level I evidence comes from studies of the strongest design, randomized trials. Within individual randomized trials or meta-analyses of a group of RCTs, we may have those that provide a precise estimate of the treatment effect, and therefore a clear course of action for most patients. Alternatively, the sample size of the individual study or meta-analysis may be relatively small and the estimate of the treatment effect imprecise. We represent the precision of the treatment effect with the CI: a
Level II: Randomized Trials or Meta-Analyses in Which the CI for the Treatment Effect Overlaps the Minimal Clinically Important Benefit
If the lower limit of the CI for a relative risk reduction falls below the important benefit that would trigger its clinical use (but the point estimate of its effect was at or above the minimal clinically significant benefit), the effect of treatment might be trivial. We classify such evidence as level II if it comes from an individual RCT. For example, the European Myocardial Infarction Project Group determined in a randomized double-blind placebo-controlled trial of 5,500 patients seen
The Grading of Recommendations About Therapy
We have described levels of evidence derived from both primary studies and meta-analyses of individual trials. Conference participants have been encouraged to classify their ultimate recommendations on the use of antithrombotic therapy into three grades, shown in Table 3, depending on the level of evidence used to generate them: grade A recommendation—supported by level I evidence; grade B recommendation—supported by level II evidence; and grade C recommendation—supported by level III, IV, or V
LIMITATIONS OF THE CURRENT LEVELS OF EVIDENCE
For more than 15 years, clinicians have been seeking an optimal method for summarizing the levels of evidence and grades of recommendations. The approach we describe in this article represents an intermediate step in this process. Limitations of the current system include a focus on establishing a treatment effect, but not establishing whether a treatment is ineffective. Moreover, this system does not capture the complexity that results from combining the strength of various study designs and
THE FUTURE
For the current conference, participants have used the formulation of levels of evidence in Table 3 to interpret their synthesis of the currently available data on antithrombotic therapy. The current approach to establishing levels of evidence and grades of recommendations, however, has a number of limitations. It does not clearly separate methodologic quality from the magnitude and precision of the treatment effect, nor does it easily accommodate recommendations not to treat. Most importantly,
REFERENCES (17)
Rules of evidence and clinical recommendations on the use of antithrombotic agents
Chest
(1989)- et al.
Rules of evidence and clinical recommendations on the use of antithrombotic agents: antithrombotic consensus conference report
Chest
(1992) - et al.
Methodological guidelines for systematic reviews of randomized controlled trials in healthcare from the Potsdam Consultation on Meta-Analysis
Clin Epidemiol
(1995) - et al.
A comparison of results of meta-analyses of randomized controlled trials and recommendations of clinical experts: treatments for myocardial infarction
JAMA
(1992) - Evidence-Based Medicine Working Group. Evidence-based medicine: a new approach to teaching the practice of medicine....
- et al.
Users' guides to the medical literature: I. How to use an article about therapy
JAMA
(1993) - et al.
Users' guides to the medical literature: II. How to use an article about therapy or prevention
JAMA
(1993)
Cited by (248)
The efficacy of psychodynamic psychotherapy: An up-to-date review
2019, Contemporary Psychodynamic Psychotherapy: Evolving Clinical PracticeAntimicrobial potential of pyroligneous extracts – a systematic review and technological prospecting
2018, Brazilian Journal of MicrobiologyEvidence of interventions for the risk of dry eye in critically ill patients: An integrative review
2016, Applied Nursing ResearchPerceived disagreement (mostly) not confirmed by evidence. A reply to Watine et al.: Clinical practice guidelines: Myths and misconceptions
2014, Journal of Clinical EpidemiologyBirth certificate of intersexual children in Brazil: Integrative Review (2011-2021)
2023, Revista Brasileira de Estudos Politicos