Those who are interested in the public mood, including politicians and economists, comment that the public are becoming ever more sceptical about many things, but health risk information should not be one of them. If health risk information is perceived by the public as ‘just another scary story’, or ‘more of the same we heard last month’, then the ability of risk messages to convey urgency and recommend action could be greatly diminished; the ‘cry wolf’ scenario becomes more real every time a threat appears in the media but fails to materialise. This thesis explores how avian influenza, (H₅N₁), as a health risk category, has been reported and represented in the New Zealand media. By analysing avian influenza-related items in four New Zealand newspapers over a six-year period, 2002-2008, and by comparing results with those found in a U.S. study by Dudo, Dahlstrom & Brossard (2007), this thesis explores the dominant themes and discourses the media drew upon when reporting the health threat of avian influenza. In addition, data from four focus groups sessions was analysed for the purpose of exploring public perceptions of health risk messages and the influence of the media on those perceptions. This thesis was situated within a constructionist epistemology, and employed a mixed-methods methodology with content, thematic and textual analyses. Risk communication theories and models, media conventions of agenda-setting and framing, and sociological concepts informed how the topic of health risk communication was operationalised. The analysis of the focus group data explored how the participants discussed the threat of H₅N₁; how they constructed concepts of personal and community risk, what role, if any, they attributed to the media in their construction and how they positioned themselves in regards to illness and contagion. The focus group analysis revealed that three dominant themes - risk, media and ‘othering’ – represented how the focus group participants talked about the risk of avian influenza. These and several sub-dominant themes shared similarities to those found in the newspaper analysis. Whilst initial discussions seemed to indicate a nonchalant attitude towards the risk of avian influenza, the many topics and themes that characterised the way the participants discussed the risk between them, showed that they had thought about the personal consequences of a possible health risk, and had formed strong opinions about many facets of that risk. Results from the newspaper analysis largely mirrored those of the above U.S. study, and showed that the New Zealand media favoured episodic over thematic framing; sensationalising the reporting of avian influenza, whilst providing little in the way of scientific and contextual information. Moreover, the analysis showed that, when reporting health risks, media templates are well established. The analysis of the focus group data revealed that the participants wanted media health risk messages to be clear, concrete and factual. However, this desire for messages that communicate certainty about risk, which is, by definition inherently uncertain, raises questions about the very nature of risk communication. Findings of this thesis suggest that future risk communication research should focus, not on how the media are reporting health risks, but how the public conceptualise risk, construct it in times of crisis and evaluate their ability to control it.