GRADE Series
GRADE guidelines: 15. Going from evidence to recommendation—determinants of a recommendation's direction and strength

https://doi.org/10.1016/j.jclinepi.2013.02.003Get rights and content

Abstract

In the GRADE approach, the strength of a recommendation reflects the extent to which we can be confident that the composite desirable effects of a management strategy outweigh the composite undesirable effects.

This article addresses GRADE's approach to determining the direction and strength of a recommendation. The GRADE describes the balance of desirable and undesirable outcomes of interest among alternative management strategies depending on four domains, namely estimates of effect for desirable and undesirable outcomes of interest, confidence in the estimates of effect, estimates of values and preferences, and resource use. Ultimately, guideline panels must use judgment in integrating these factors to make a strong or weak recommendation for or against an intervention.

Introduction

In prior articles in this series devoted to the GRADE approach to systematic reviews and practice guidelines, we have dealt with the process before developing recommendations, namely framing the question and choosing critical and important outcomes [1], rating the confidence in effect estimates for each outcome [2], [3], [4], [5], [6], [7], [8], dealing with resource use [9], rating the confidence in effect estimates across outcomes [10], and creating an evidence profile and a Summary of Findings table [11], [12], [13]. The immediately previous article described GRADE's approach to classifying the strength and direction of recommendations and discussed the implications of strong and weak recommendations, and the options for presentation and wording [14]. The present article presents GRADE's approach to moving from evidence to recommendations. As we did in the previous article, we will refer to guideline developers as “the panel.”

The pithy summary by Eisenberg [15] on the relationship between evidence and recommendations, “globalize the evidence, localize the decisions,” provides fundamental guidance for those working to produce evidence-based recommendations [15]. Summaries of evidence regarding alternative management strategies from the medical literature should ideally be very similar, no matter the site of the application of the recommendation.

Rating of confidence in estimates of effect (quality of evidence) may, however, differ for a variety of reasons. First, desirable and undesirable outcomes may be valued differently, leading to different thresholds of acceptability. This could lead to different judgments regarding imprecision, as we have highlighted in the article in this series dealing with imprecision [5].

Second, differences in values and preferences could lead to differences in the overall balance of desirable and undesirable outcomes and the rating of confidence in estimates: an outcome judged as critical by one panel (and thus included in the rating of overall confidence in estimates) may be judged important but not critical by another (and thus not included in the overall rating).

Finally, ratings of confidence may also differ as a result of uncertainties in the risk profile of untreated populations (baseline risk). We may be very confident of baseline risk in one setting but not at all confident in another. This could lead to rating down confidence in estimates for indirectness.

Continued rapid uptake of GRADE by organizations that produce systematic summaries of evidence will greatly facilitate the production of transparent evidence summaries. If organizations work together to produce summaries, there will be an enormous gain in efficiency [16]—even if, in the end, judgments about confidence in estimates will differ across settings, for reasons described in the preceding paragraphs.

We now turn to a systematic presentation of the determinants of direction and strength of recommendations.

Section snippets

Determinants of direction and strength of recommendations

GRADE has identified six determinants of the direction and strength of recommendations, namely the magnitude of estimates of effect of the interventions on important outcomes, confidence in those estimates, estimates of typical values and preferences, confidence in those estimates, variability of values and preferences, and resource use. In the presentation here, we will present these six determinants in four domains. We package magnitude of effect and typical values and preferences together

Baseline risk (control event rate) can influence the balance

Table 3 presents an example of how guideline panels can move from evidence to recommendations in an explicit and transparent way. The final column in Table 3 presents the issues (if one calls the four determinants domains, then one might call these issues subdomains) that guideline panels should consider under each domain. One of these subdomains, which may be critical in the decision, is baseline risk.

Because, we usually determine absolute risk differences through applying the relative risk

Conclusion

Patients, clinicians, and policy makers will all be better served by a more systematic and transparent system for judging the direction and strength of recommendations. Explicit presentation of how panels view the four domains to consider in the direction and strength of recommendations could play an important role in improving the transparency of panel decisions (Table 3).

References (44)

  • G.H. Guyatt et al.

    GRADE guidelines: 12. Preparing summary of findings tables: binary outcomes

    J Clin Epidemiol

    (2013)
  • G.H. Guyatt et al.

    GRADE guidelines: 13. Preparing summary of findings tables: continuous outcomes

    J Clin Epidemiol

    (2013)
  • G. Guyatt et al.

    GRADE guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables

    J Clin Epidemiol

    (2011)
  • J. Andrews et al.

    GRADE guidelines: 14. Going from evidence to recommendations: the significance and presentation of recommendations

    J Clin Epidemiol

    (2013)
  • H.J. Schunemann et al.

    A vision statement on guideline development for respiratory disease: the example of COPD

    Lancet

    (2009)
  • S. Bates et al.

    VTE, thrombophilia, antithrombotic therapy, and pregnancy: Antithrombotic therapy and prevention of thrombosis, 9th ed.: American College of Chest physicians Evidence-Based Clinical Practice Guidelines

    Chest

    (2012)
  • S.G. Goodman et al.

    Acute ST-segment elevation myocardial infarction: American College of Chest physicians evidence-based clinical practice guidelines (8th Edition)

    Chest

    (2008)
  • H.J. Schunemann et al.

    WHO Rapid Advice Guidelines for pharmacological management of sporadic human infection with avian influenza A (H5N1) virus

    Lancet Infect Dis

    (2007)
  • B. Levin et al.

    Screening and surveillance for the early detection of colorectal cancer and adenomatous polyps, 2008: a joint guideline from the American Cancer Society, the US Multi-Society Task Force on Colorectal Cancer, and the American College of Radiology

    Gastroenterology

    (2008)
  • F. van Lier et al.

    Effect of chronic beta-blocker use on stroke after noncardiac surgery

    Am J Cardiol

    (2009)
  • S. Bangalore et al.

    Perioperative beta blockers in patients having non-cardiac surgery: a meta-analysis

    Lancet

    (2008)
  • J.M. Eisenberg

    Globalize the evidence, localize the decision: evidence-based medicine and international diversity

    Health Aff (Millwood)

    (2002)
  • Cited by (948)

    View all citing articles on Scopus

    The GRADE system has been developed by the GRADE Working Group. The named authors drafted and revised this article. A complete list of contributors to this series can be found on the Journal of Clinical Epidemiology web site.

    View full text