Practitioner Review: On the trustworthiness of clinical practice guidelines – a systematic review of the quality of methods used to develop guidelines in child and youth mental health

Background Numerous practice guidelines ( PG s) relevant to child and youth mental health ( CYMH ) are available, but their quality is uncertain. We used systematic review methodology to identify the methods employed to develop PG s in CYMH and assess whether they align with international quality standards. Methods We used prespecified inclusion criteria to search for CYMH PG s (2009–2014) in journals of professional associations or websites of organizations who produce or house PG s. Eligible PG s and organization websites were screened to identify PG development methods. Two reviewers assessed the alignment of the PG development methods with PG quality criteria using the Appraisal of Guidelines for Research and Evaluation ( AGREE II ) domains and Institute of Medicine ( IOM ) standards. Results Five sets of eligible development methods were identified in 70 eligible PG s. Three sets adhered to all (National Institute for Health and Care Excellence; Scottish Intercollegiate Guidelines Network) or most (U.S. Preventive Services Task Force) AGREE II domains and IOM standards, and were used to develop 31.4% of PG s. The two remaining sets of development methods had important weaknesses (e.g. lack of mandatory rigorous systematic reviews, multidisciplinary development groups, or transparent conflict of interest methods) and were associated with 21.4% of PG s. No development methods could be identified in 40.0% of PG s; ineligible development methods were referenced in 7.1% of PG s. Conclusions Up to 69% of available CYMH PG s may have been developed using methods that do not align with AGREE II quality criteria or IOM standards. The quality of available CYMH PG s needs to be assessed, and strategies designed to guide practitioners to high quality PG s and facilitate adherence by PG developers to international quality standards are needed.