Rigorous method to assess quality and generalizability of clinical practice guidelines.

Clinical practice guidelines (CPGs) are important tools for clinical decision-making in modern health care.1The introduction of CPGs into clinical practice has revolutionized the way clinicians care for patients, by allowing them to integrate principles of evidencebased medicine with patient-specific factors and clinical judgment.2 Through dissemination of specific recommendations, CPGs attempt to standardize care according to established best practices. They also provide a means for monitoring prescriber practices at both individual and institutional levels.3 However, certain aspects of CPGs may, directly or indirectly, have negative effects on care.4 For example, use of expert opinion, lack of a stringent review process, or direct financial sponsorship may compromise the validity of a published guideline. A recent Point Counterpoint debate in the CJHP raised many points both supporting and refuting the use of CPGs in clinical practice.5,6 With this discussion as background, our aim was to demonstrate a process of rigorous guideline appraisal by using a standardized method to assess the recently published Joint National Committee hypertension guidelines.7 The hypertension CPG was appraised using the Appraisal of Guidelines for Research and Evaluation II (AGREE II) instrument.8 Application of this tool involves ranking 23 items within 6 domains and completing additional categories for “Overall Assessment” and “Recommendation”. For the 23 items and the overall assessment, the appraisers were asked to assign a rank between 1 (strongly disagree) and 7 (strongly agree). For example, the first item under the domain “scope and purpose” states, “The overall objective(s) of the guideline is (are) specifically described”, and appraisers used the 7-point scale to rate how well the CPG fulfilled this criterion. Standardized domain scores were subsequently calculated according to the formula provided in AGREE II.8 Domain scores are reported as percentages, on a scale from 0 to 100%, with 100% being the highest score possible. For the final recommendation, appraisers were asked to state whether or not they would recommend using the guideline, or if they would recommend using the guideline with required modifications. The CPG was appraised independently by 6 investigators (the authors of this letter). Upon completion, all appraisals were forwarded to one investigator (K.J.W.), who reviewed and validated the data by ensuring that no item had scores differing by 6 points or more. If any item had such widely divergent scores, the guideline was redistributed to all investigators for review of the discrepant item. Once final rankings were received, standardized scores were calculated for each domain.8The group’s final recommendation was the recommendation selected most frequently by appraisers (Yes, Yes with modifications, or No). The highest-scoring domain was clarity of presentation (90.7%) and the lowest was applicability (34.7%) (Table 1). The overall assessment was 75%, and the group’s final recommendation was that the CPG should be used in clinical practice, with modifications (2 appraisers recommended using the CPG as is, 4 recommended using the CPG with modifications, and none recommended not using the CPG). Although clarity of presentation, the highest-ranked domain, is an important aspect of a CPG, it does not indicate the meth odologic strength of a guideline, and other domains, such as rigour of development and applicability, should be weighted more heavily. However, these results align with a previously published report that ranked highly the clarity of pharmacotherapy guidelines.9 For the hypertension guideline, the domains of applicability and stakeholder involvement were ranked lowest. The applicability domain is largely based on the provision of discussion and/or tools to facilitate implementation of the guideline in practice. Therefore, statements about facilitators of and barriers to guideline adoption, resource implications, monitoring and auditing criteria, and/or provision of advice or tools for implementation will be helpful in enhancing applicability. Although overall recommendations presented in a CPG may be generalizable to many settings, inclusion of these considerations will allow clinicians to apply the recommendations in practice. Additionally, guideline creators