Importance of evidence grading for guideline implementation: the example of asthma

Abstract

The goal of evidence-based clinical guidelines is to improve the value of health care by recommending treatments with favorable benefit/harm ratios. Achieving this goal requires use of evidence-grading systems that explicitly address strength of evidence in terms of external validity (generalizability), internal validity, and patient-oriented outcomes. To be clinically useful, guidelines should also incorporate patient preferences, particularly when evidence is weak. The National Heart, Lung and Blood Institute recently published Expert Panel Report 3: Guidelines for the Diagnosis and Management of Asthma (EPR-3). This special report addresses the extent to which current guidelines adhere to the principles enunciated above by using EPR-3 as the prime example. EPR-3 used an unconventional evidence-grading system that emphasized precision and consistency (statistical significance, large sample sizes, and/or consistency of results) at the expense of patient-oriented outcomes and generalizability (applicability to the general population). EPR-3 did not report information on numbers needed to treat or numbers needed to harm, which are useful in eliciting patient preferences via shared decision making. Asthma guidelines (and others) are limited by lack of a generalizable research base, 3 awed evidence grading, and lack of attention to patient preferences. An evidence-grading system based on applicable populations, patient-oriented outcomes, and shared decision making might improve physician and patient guideline adherence and improve asthma outcomes.

Figure 1.

Strength of Recommendation Taxonomy (SORT): evidence-grading system for individual studies. Reproduced by permission from the American Board of Family Medicine.