What's in the message? Applying extended parallel process model to wood smoke pollution messages

Abstract

Wood smoke has been linked to serious negative health impacts such as respiratory disorders and cognitive decline in the elderly. Traditional health risk communication models focus on educating and providing risk information about the perceived threat (e.g., Witte, 1992). However, such models downplay the complex nature of the public's understanding of risk information. Communicating effective messages is crucial to encourage positive, protective behaviour change. The extended parallel processing model offers a practical framework to examine how messages may be processed, and subsequently accepted or rejected. The present study investigated the effectiveness of a range of wood smoke messages comprising Threat, Efficacy, or both, in motivating message acceptance (or rejection). Participants (N = 225, Mage=35.32 years, SD=13.6, 53.8% male) were randomly assigned to one of the four experimental conditions: (1) Threat only - participants viewed a message depicting "threat"; (2) Efficacy only - participants viewed a message depicting "efficacy"; (3) Threat and Efficacy - participants viewed a message depicting both "threat" and "efficacy"; and (4) neither Threat nor Efficacy (control). After viewing a message, participants completed a range of measures assessing their perceived levels of response efficacy, self-efficacy, threat susceptibility, threat severity, danger control (message acceptance), and fear control (message rejection). A 2x 2 analysis of variance (ANOVA) revealed significant main effects of Threat only and Efficacy only for danger control responses. The ANOVA results for fear control responses found significant main effect of the Threat only and the Threat by Efficacy interaction. Post-hoc analyses revealed that participants in the Threat only condition reported significantly higher fear control responses. Follow-up multiple mediation analyses found that the Efficacy message component impacted danger control by increasing response efficacy, self-efficacy, and threat susceptibility. On the other hand, the Threat message component influenced fear control via threat susceptibility and threat severity, with the higher reported levels of the threat severity associated with lower fear control. The present findings may help inform and guide real-world education campaigns to reduce wood smoke pollution. Implications are discussed regarding the optimum use of Efficacy and Threat components in increasing the effectiveness of communication strategies.