Grading of evidence for bladder pain syndrome: a comparative review of study quality assessment methods

Abstract

Clinical guidelines on bladder pain syndrome (BPS) report quality ratings for evidence based on study design. The Grading of Recommendations Assessment, Development and Evaluation (GRADE) system takes into account several domains in addition to limitations of study design for assigning quality ratings. We compared the quality of evidence described in current BPS literature. All existing systematic reviews and guidelines on BPS management were reviewed, and included evidence was rated according to GRADE on a four-point scale (1-4, from very low to high). These ratings were compared to the two reported quality assessments that assigned levels or strengths to evidence; both had a four-point scale: level of evidence 1-4 from meta-analysis of randomised studies to expert opinion; and strength of evidence 1-4 from very low to high. Of the 19 treatments for BPS with GRADE ratings, comparison with level of evidence ratings showed that, on average, the latter overestimated quality by 1.8 points [1.1 v 2.9; 95% confidence interval (CI) 1.2-2.3; p = <0.0001). Comparison of GRADE ratings with strength of evidence ratings showed that, on average, the latter overestimated quality by 1.7 points (1.1 v 2.8; 95% CI 1.3-2.1; p = <0.0001). GRADE, a refined method of assigning quality to evidence, provided a more conservative gauge, giving a realistic assessment of the value of recommendations for consideration in practice.