Background: "Click to get sick", wrote Time Magazine. Pravda told us to "Beware of Internet Health advice". The publicity was enormous. More than 50 other international media outlets carried similar stories last autumn, making this one of the Cochrane Reviews that received most attention in 2004. They referred to Cochrane Review Issue 4 2004 by Elizabeth Murray and colleagues entitled "Interactive Health Applications for people with chronic disease".
The Norwegian Centre for Telemedicine discovered a few days later that this review contained an error, resulting in an incorrect direction of change in several of the clinical and behavioural outcomes. The review was withdrawn, but it should turn out to be a much more difficult job trying to correct the unfounded message given to the public.

Objectives: This presentation is about this process, and about what Cochrane did, and did not, contribute in trying to correct this.

Methods: A review of the data material in the meta-analysis resulted in the conclusion that the review contained serious errors. Web searches are used for determining the media coverage and impact. The presenter had a long discussion with the Cochrane Collaboration, both publicly through web feedback and by email.

Results: The erroneous picture given to the public about the effect of Internet on health, is not corrected.

Conclusions: Since the enormous publicity about these results is a result from the quality stamp of the Cochrane Collaboration. Therefore it is also Cochrane's responsibility to correct this when an unfounded image is given to the public. Cochrane failed to take this responsibility.

NOTE: An updated version of Murray's review will be available in Issue 2 2005. This presentation's focus and/or conclusion might be affected by these results and the following media attention.