The quality of clinical practice guidelines for chronic respiratory diseases and the reliability of the AGREE II: an observational study

Abstract

Objectives To survey the quality of evidence-based clinical practice guidelines for chronic respiratory diseases relevant to physiotherapy practice using the Appraisal of Guidelines for Research and Evaluation version II instrument (AGREE II) and to evaluate the inter-rater reliability of AGREE II. Design Observational survey. Procedures Guidelines indexed in the Physiotherapy Evidence Database (PEDro) on chronic respiratory diseases were evaluated by four assessors using AGREE II. Main outcome measures The six domains and two global items of AGREE II. Results Thirty-three guidelines were evaluated (58% were published in the last 5 years and 36% were for chronic obstructive pulmonary disease). The domains with the highest scores were scope and purpose (79%, SD 10%) and clarity of presentation (79%, SD 10%). The domain with the lowest score was applicability (37%, SD 23%). Mean overall quality was five out of seven (SD 1). Intraclass correlation coefficients ranged from 0.66 to 0.93 for the six domains and first global item, suggesting good to excellent reliability. The second global item had very poor reliability (Kappa 0.097). Conclusion The quality of evidence-based clinical practice guidelines for chronic respiratory diseases relevant to physiotherapy could be improved, particularly in consideration with applicability. The number of assessors for AGREE II could be reduced because of the good inter-rater reliability.