Background: Consumer feedback may be able to improve the quality and relevance of patient information, but the best ways to do this are unclear. Feedback from members of consumer or self-help groups can provide valuable expert input from an expert patient point of view on many topics, but does not represent most of the likely users of the information. Objectives: To assess the feasibility and value of a pilot programme of routine independently conducted focus groups for professionally-written evidence-based patient information, including extended summaries based on Cochrane reviews. Methods: Focus groups, usually of 5 people ( testers ), were recruited by an external group of academic experts. The testers were drawn from enrolees of a patient university , although some were recruited through a university clinic or other means when a specific profile (such as adolescents) was requested by the editorial team. The testers received a small honorarium, were not informed of the identity of the publisher, and were as a rule only able to be a tester once. Each group reviewed 4 items of information, many of which were extended summaries of Cochrane reviews. Testers independently rated the information on a range of criteria such as readability and potential adverse effects (such as anxiety), including criteria related both to patients and their family/friends. The 5 people then participated in a lengthy focus group led by the external experts, systematically discussing each piece of information using guidelines. The rating questions and guidelines were developed with input from the editorial team. Focus groups were taped. Individual ratings, a summary of the focus group discussions and demographic data were provided to the editorial team. Results: Between 2008 and early 2010 summaries of 52 Cochrane reviews were tested in 30 focus groups including 136 different testers. The testers were probably representative of the general public in terms of self-help group membership (7%). However, women and people of higher educational levels were over-represented (65% were women; 36% had university-level education). Nearly 70% were 45 years or over. A period of 4 weeks was almost always sufficient for the external group to complete the process. Conclusions: The focus groups and external experts provided valuable insight into the way people unfamiliar with evidence summaries reacted to the information, and strongly influenced the development of our methods of communicating about evidence. Groups which were recruited with very specific profiles (be they patients or their family and friends) were particularly useful. However, even groups for whom the information was of no personal interest identified areas for improvement. Focus group methodology provided insight into differing opinions among potential users and could helpfully elaborate on these. The primary convenience sample (patient university enrolees) reduced costs and time, but it had drawbacks in terms of representativeness. For selected information, this was overcome by profiled recruitment. This methodology will be continued as part of our editorial routine, including using experts with a variety of recruitment options.