Background: There are an increasing number of information resources in the so called Web 2.0. Social network sites like YouTubeTM, Facebook, Twitter, MySpace, etc. and a plethora of blogs, wikis and other information applications allow internet users to post and find health-related information. Objectives: To assess the nature and quality of health information in Web 2.0 using the pilot example of the antidepression drug Sertraline (Zoloft) in the video clip platform YouTubeTM. Methods: We searched YouTubeTM for Zoloft (10 April 2009). The identified sample of video clips was assessed according to existing quality criteria for health information: Discern, Check-in, Good Practice Patient Information and AIMQ. Results: We identified 602 video clips. After deleting non-relevant clips or clips about mental disorders not explicitly informing about Zoloft, 24 clips remained, posted between April 2007 and March 2009. These clips could be categorised into patient experiences (n=7), commercials (n=7), reports (n=6), physician informing (n=2), TV news (n=1), TV reports (n=3) and hearings (n=1). We assessed the clips by the following criteria: ‘aims clear’, ‘relevant’, ‘additional information’, ‘date of production’, ‘balanced and unbiased’, ‘benefits and risks’ and ‘uncertainty’. The existing quality assessment criteria could hardly be applied for this medium and format of information as some clips fulfilled only two of nine items. Conclusions: Web 2.0 has a big potential for disseminating timely health information. But for users it is not possible to judge the quality as defined by current academic quality instruments. YouTubeTM clips can therefore not be recommended as reliable source of health information. Regarding the user driven character of such networks a top-down implementation of quality instruments seems futile. Implications for practice could be to develop Cochrane evidence products which fit into these formats.