Eyes on Guidelines: Evaluating the Quality of Diabetic Eye Disease Clinical Practice Guidelines

Purpose : Diabetic retinopathy is a leading cause of vision impairment globally. Clinical practice guidelines have been developed to assist practitioners in providing evidence-based care to improve vision outcomes. However, the methodological rigor of these guidelines remains largely unknown. This study aimed to systematically review the quality of existing diabetic eye care guidelines.

Methods : A systematic search of guidelines for comprehensive clinical eye care of adult patients with diabetes was conducted on MEDLINE, Scopus PubMed, EMBASE, ProQuest, Web of Science and websites of relevant guideline developers and professional societies. Four reviewers independently rated the quality of the included guidelines using the Appraisal of Guidelines, Research and Evaluation (AGREE II) instrument. Aggregate scores (%) for six domains and overall quality assessment were calculated. A ‘good quality’ guideline was defined as the one with ≥60% score for ‘rigour of development’ and in at least two other domains.

Results : Eighteen guidelines met inclusion criteria, of which 13 were evidence-based guidelines (involved systematic search and grading of evidence). In general, guidelines scored high in ‘clarity of presentation’ (Median (interquartile range (IQR)): 86.6% (76.7% to 94.4%), and ‘scope and purpose’ (73.6% (54.2% to 80.6%)) and low in ‘applicability’ (28.6% (18.0% to 37.8%)) and ‘stakeholder involvement’ (48.6% (29.2% to 71.5%)). The median scores for ‘rigour of development’ and ‘editorial independence’ were 60.2% (30.9% to 78.1%) and 67.7% (24.0% to 83.3%), respectively. The median overall score (out of 7) of all guidelines was 5.1 (IQR: 3.7 to 5.8). The evidence-based guidelines scored significantly higher in overall assessment and all domains except for ‘clarity of presentation’ compared to expert-consensus guidelines. Half (n=9) of the guidelines (all-evidence-based) were of ‘good quality’.

Conclusions : A wide variation in methodological quality exists among diabetic eyecare guidelines, with nine guidelines demonstrating ‘good quality’. Future iterations of guidelines could improve by appropriately engaging stakeholders, following a rigorous development process, including support for application in clinical practice and ensuring editorial transparency.