Letters to the editor Social Marketing and the FAST Campaign Dear Editor, We write in reference to the paper by Bray et al. (1) which provided useful information about the best way in which to present information about the signs of stroke in social marketing programs such as the FAST campaign, used in many English speaking coun- tries around the world. Indeed, the National Stroke Foundation (NSF) has used some of the outputs of this research to modify and improve the FAST cam- paign in Australia. There are, however, significant lim- itations to this study as an evaluation of the Australian FAST campaign. The NSF has run the FAST campaign since September 2006. The FAST test is a simple way of remembering the key signs of stroke – facial weakness, arm weakness, speech difficulties and time to act fast, ensures rapid recognition and activates a series of steps to ensure rapid assessment and diagnosis of stroke. Be- fore this, a campaign outlining the signs of stroke began in 2004. The change to FAST provided a way to develop a simple and easy to remember message. The Bray and colleagues survey was conducted in two Melbourne public hospitals from August 2006 until April 2008 with a nonrandom sample of 34% of all patients admitted for stroke. We note both the unusually high propor- tion of ‘in-hospital’ stroke patients excluded, 23% of all stroke admissions, and that patients for whom unidentified bystanders acted, 8% of all admissions, were also excluded from the study – an important potential source of selection bias. During the period of the study by Bray and colleagues, the NSF ran only two low-budget and time-limited mar- keting campaign of four-week duration each September in Melbourne to pro- mote FAST around national stroke week. It is highly unlikely that interview participants in a small, nonrepresenta- tive sample of hospital patients spread over almost two-years would recall a television advert they may have seen three or four times over a period of weeks, in some cases up to 11 months earlier. Evaluation of such marketing campaigns is best done using larger, community-based surveys randomly se- lecting people in the target market. As the authors correctly note, com- munity awareness of the FAST message will be highest around the time of our two mass media pushes and diminish over time. This is an important argu- ment for more consistent promotion of the FAST message. Since 2004, NSF awareness campaigns have been evalu- ated through an annual computer-as- sisted telephone interview survey of a representative sample adults. This sur- vey has been conducted six-weeks after national stroke week with a sample size calculated to validly demonstrate change in awareness over time. This survey has shown clear improvements in awareness of the signs of stroke over the length of the campaign (see Table 1). Measurement of the campaign reach shows that the FAST mass media cam- paign has been able to reach about 80% of our target audience of people aged 55 years and older three times during the advertising period of one-month dura- tion when it is funded. However, it is generally accepted that greater reach into the target market is required for sustained change some months after the campaign with recommendations that such campaigns are aired in short and frequent bursts. Many non-Governmental social mar- keting campaigns are run with pro- bono advertising and therefore have limited reach into the target market. A paper recently presented at the World Stroke Congress in Seoul demonstrated that increased funding and campaign reach leads to significantly greater im- provement in awareness of the signs of stroke, emphasizing that exposure to, or the ‘dose’ of, the campaign is as impor- tant as the content (2). The NSF welcomed the opportunity to work with the authors of this study and there are many valuable lessons to be learnt from their paper. However, the paper by Bray and colleagues is not a robust public health evaluation of the Australian FAST campaign. Such eva- luation requires significantly larger sample sizes of people randomly se- lected from the community in whom the intervention was performed, with consideration of the funding for, and reach of, the campaign, that is, the adequacy of the campaign dose. Bruce Bolam à , Erin Lalor, and Richard Lindley National Stroke Foundation, Melbourne, Vic., Australia Correspondence: Bruce Bolam à , National Stroke Foundation, Level 7, 461 Bourke, Street, Melbourne, Vic. 3000, Australia. E-mail: bbolam@strokefoundation.com.au DOI: 10.1111/j.1747-4949.2011.00630.x References 1 Bray JE, O’Connell B, Gilligan A, Livingston PM, Bladin C. Is FAST stroke smart? Do the content and language used in awareness campaigns describe the experience of stroke symptoms? Int J Stroke 2010; 10:440–6. 2 Trobbiani K, Freeman K, Arango M, Lalor E, Thrift A. Comparison of stroke warning sign campaigns in Australia, United Kingdom and Canada. World Stroke Conference, Seoul, 2010. Table 1 Knowledge of signs of stroke in Australia – changes over time 2004 (%) 2008 (%) Know at least one sign 69 78 Know at least two signs 43 59 Know at least three signs 19 35 Know four signs or more 8 16 Do not know/incorrect 30 22 & 2011 The Authors. International Journal of Stroke & 2011 World Stroke Organization Vol 6, August 2011, 370–373 370