Background: The Cochrane home page gives two statements describing The Cochrane Collaboration: ‘The reliable source of evidence in health care’ and ‘Improving healthcare decision making globally, through systematic reviews of the effects of healthcare interventions, published in The Cochrane Library’. Although this represents a passive approach to the implementation of Cochrane reviews, there seems to be surprise that Cochrane reviews have not been extensively adopted. We examine both statements from the perspective of end users, in this case, guideline developers. Methods: Examining the evidence: We examined records in the Cochrane Methodology Register that considered some quality aspect of Cochrane reviews and summarised the findings in a narrative review. These ranged from ‘Cochrane reviews were better than paper based journals in terms of assessment of methodological quality of primary studies’ to ‘the Cochrane review conclusion was neither robust nor valid’. Results: Improving healthcare decision making: For use in guideline development, Cochrane reviews should be reliable and relevant. Decision making should be related to the appropriate healthcare system. Cochrane reviews may have to be modified or re-done: they are not always applicable to the UK or to the guideline’s protocols; they may need to be updated; and their quality is variable. If it is true (or perceived) that not all Cochrane reviews are of high quality, to what extent can the end user rely on a particular review? We discuss the need for more robust quality assurance across Cochrane reviews. For NICE guidelines, interventions should be those licensed for use/in common use in the UK. Update searches are required to be within six weeks of guideline submission. Therefore, the end user may need to adapt Cochrane reviews, e.g. excluding irrelevant studies or adding new ones. Implementation of Cochrane reviews would be facilitated if individual meta-analyses and tables could be downloaded and then adapted by the end user. Conclusions: Cochrane reviews may be more readily implemented in guideline development, if the end user is confident the review has undergone rigorous quality assurance, and if there is the facility to download and modify the review to suit the user’s own needs.