Background When guidelines are called evidence-based (EBGs) the implication is that there was a transparent and systematic assessment of evidence. However, evidence is often insufficient to draw conclusions. In these cases, what do guideline committees do?

Objectives To estimate the proportion of recommendations in a set of EBGs for which there was insufficient evidence and to determine the implications for the guidelines.

Methods We reviewed EBGs developed by the Cystic Fibrosis Foundation and classified each recommendation as developed using evidence (i.e., graded) or consensus.

Results We identified 143 recommendations from 7 EBGs. More than half of the recommendations were consensus-based (61%) while 39% were evidence-based. Of those classified as evidence-based, 41% were non-recommendations; these were statements where the committee felt unable to make a recommendation for or against a particular action (insufficient grade). The consensus-based recommendations included 11% adopted from other organisations; only a quarter of the consensus-based statements were developed using an apparent formal method.

Discussion Committee members often feel compelled to provide recommendations for the community even in the face of insufficient evidence. This limits the transparency of the EBG. Further research is needed to determine the potential implications for implementation.

Implications for Guideline Developers/Users Guidance is needed for guideline developers on best practices for if or when to consider consensus-based recommendations and how to develop such recommendations. The guidelines should clearly outline choices, such as when a consensus-based recommendation was made versus not making a recommendation, and the methods used.