| Summary: | <strong>Background</strong> The development of robust clinical guidelines requires standardised development methods informed by robust evidence synthesis. <strong>Objectives</strong> We evaluated the methodological quality of endometriosis guidelines, mapped their recommendations, and explored the relationships between recommendations and research evidence. <strong>Search Strategy</strong> We searched: [1] EMBASE; [2] Medline; and [3] Pubmed from inception to February 2016. <strong>Selection Criteria</strong> We included guidelines related to the diagnosis and management of endometriosis. <strong>Data Collection and Analysis</strong> The search strategy identified 879 titles and abstracts. We include two international and five national guidelines. Four independent authors assessed the methodological quality of included guidelines using the Appraisal of Guidelines for REsearch & Evaluation (AGREE-II) instrument and systematically extracted the guideline recommendations and supporting research evidence. <strong>Main Results</strong> One hundred and fifty-two different recommendations were made. Ten recommendations (7%) were comparable across guidelines. The European Society of Human Reproduction and Embryology was objectively evaluated as the highest quality guideline (methodological quality score: 88/100). There was substantial variation between the supporting evidence presented by individual guidelines for comparable recommendations. Forty-two recommendations (28%) were not supported by research evidence. No guideline followed the standardised guideline development methods (AGREE-II). <strong>Conclusion</strong> There is substantial variation in the recommendations and methodological quality of endometriosis guidelines. Future guidelines should be developed with reference to high quality methods, in consultation with key stakeholders, including women with endometriosis, ensuring their scope can truly inform clinical practice and eliminate unwarranted and unjustified variations in clinical practice.
|
|---|