Pro-smoking information scanning using social media and increased smoking among young adults

The amount of pro-smoking information appearing on social media has increased sharply in the past few years (Freeman & Chapman, 2007, 2010). This proliferation has expanded the potential for widespread exposure to information about smoking. Such potential highlights the need to understand how individuals acquire and use this information to make decisions about smoking initiation and cessation behaviors. Being in a critical age group (aged 18-25) for lifelong smoking behavior (Gilpin, White, & Pierce, 2005), young adults use social media ubiquitously. This study introduces information scanning (Niederdeppe, Hornik, Kelly, Frosch, Romantan, Stevens, Barg, Weiner, & Schwartz, 2007; Hornik & Niederdeppe, 2008) and the Integrative Model of Behavioral Prediction (Fishbein and Cappella, 2006; Fishbein, Hennessy, Yzer and Douglas, 2003; Fishbein and Yzer, 2003; Yzer, 2012) as useful constructs for understanding young adult smoking in the context of social media. Information scanning, understood in this research as routine patterns of exposure to mediated and interpersonal sources, has been found to be useful in predicting cancer-related behaviors (e.g., Kelly, Hornik, & Niederdeppe, 2009; Shim, Kelly, & Hornik, 2006) but has never used to understand smoking behavior. This study builds on research that has found that only a small number of variables need to be considered to predict, change, or strengthen a particular behavior in certain population (Fishbein & Ajzen, 1975, 2010). To understand the extent to which a young adult’s pro-smoking information scanning using social media affects the likelihood of being susceptible to smoking, being an experimental smoker, and being an established smoker. Specifically, this thesis hypothesizes (1) that pro-smoking information scanning using social media will influence smoking behavior, (2) that pro-smoking information scanning will interact with attitudes toward smoking, social norms regarding smoking, and smoking self-efficacy, interpersonal information scanning, and participation level on social media to impact smoking behavior, and (3) information scanning will contribute to the predictive validity of the Integrative Model of Behavioral Prediction to predict intentions to smoke. To test these hypotheses, a cross-sectional survey of 247 young adults (aged 18-25) was conducted. Results of this survey indicated that pro-smoking information scanning through social media significantly impacted attitudes toward smoking, social norms regarding smoking, and smoking self-efficacy. Pro-smoking information scanning using social media is independently related to smoking behavior after controlling for factors such as gender, ethnicity, academic achievement, interpersonal information scanning, attitudes toward smoking, social norms regarding smoking, and smoking self-efficacy. Only attitudes toward smoking and interpersonal information scanning mediate the relationship between pro-smoking information scanning through social media and experimental and established smoking. Additionally, inclusion of information scanning variables increased the predictive ability of the Integrative Model of Behavioral Prediction. This study should be a wakeup call for more comprehensive and concerted efforts on the interaction between tobacco control and social media use. It concludes with a discussion of the theoretical and practical implications of these findings, especially the theory-based antismoking interventions using social media.