Objective This project aims to: (1) describe the frequency of news coverage in mass media related to the National Institute for Health and Clinical Excellence (NICE) draft or final guidance; (2) analyse the types of evidence and sources of information that was quoted in the news; (3) compare whether the patterns of coverage differ between media.

Design A survey of news articles related to decisions made by NICE's Health Technology Appraisal committees was conducted. Relevant news articles were retrieved from websites of major UK news media. Inclusion criteria were: (1) news articles related to specific NICE decision(s); (2) articles were written by a reporter/writer/editor of the news media. Articles that mentioned NICE for other reasons, columns and readers' letters were excluded.

Setting Major UK news media, including national newspapers and news channels.

Samples News articles published during 2007–2008 in 13 different sources of media including tabloids (Daily Express, Daily Mail, News of the World, The Mirror, The Sun), broadsheets (Independent, Guardian, The Telegraph, The Times) and news channels (BBC, ITN, Sky News and Reuters).

Main outcome measure The following data were collected by one author and checked by another: nature of guidance (disease area; positive or negative recommendation), use of generic or brand name and source/type of evidence that was quoted. Descriptive statistics were compiled and comparisons between types/sources of news media were made using χ² test.

Results 329 articles were included. BBC, Daily Mail and The Telegraph published more than 50 articles related to health technology appraisal whereas ITN and News of the World published less than 10 articles during the 2-year period assessed. Two-thirds (220/329) of the articles were related to negative recommendations. There was significant difference in the proportion of articles relating to negative recommendations between individual sources of media (p=0.001) but not between types of media (p=0.286). Cancer (33%), neurology–mainly Alzheimer's disease (22%), ophthalmology (13%) and rheumatology (10%) were most frequently covered areas. 58% (192/329) of the articles quoted only brand names without mentioning generic names of the drugs. Approximately 50% of articles included statements of effectiveness without referring to the source of evidence and another 40% did not describe clinical effectiveness. 24% of articles did not mention drug costs or cost-effectiveness.

Conclusion NICE decisions on new drugs, particularly negative recommendations, attracted significant media attention but the coverage and contents varied substantially between individual sources.