Search

Any evidence-based recommendation needs careful assessment of its methodological background as well as of its content trustworthiness, especially given that following it will not necessarily produce the intended clinical outcomes. There are no established instruments to evaluate guidelines for their content, while useful tools assessing the quality of methods followed are well recognised and adopted. We suggest a 'safety bundle' considering methodological aspects and content trustworthiness of guidelines, by adopting the GRADE method in a backward fashion. Sharing the critical analysis of the guidelines with patients, including any eventual uncertainty about them, is of key importance in order to avoid the possible adverse effects derived from following the wrong guidelines. Such critical approach is also helpful and beneficial in producing better care pathways, health policy decisions and more relevant and ethical research.

Evidence-based guidelines are considered an essential tool in assisting physicians, policymakers and patients when choosing among alternative care options and are considered unbiased standards of care. Unfortunately, depending on how their reliability is measured, up to 50% of guidelines can be considered untrustworthy. This carries serious consequences for patients' safety, resource use and health economics burden. Although conflict of interests, panel composition and methodological flaws are traditionally thought to be the main reasons undermining their untrustworthiness, corruption and waste of biomedical research also contribute. We discuss these issues in the hope for a wider awareness of the limits of guidelines.