BackgroundThis study examined how the Risk Perception Attitude Framework (RPA) could be used to segment audiences in the COVID-19 prevention campaign in Vietnam. It investigated how the identified segments differed regarding staying at home and mask wearing behaviors and whether exposure to prevention messages via social media and mass media might predict the identified segments.MethodA cross-sectional survey (N = 360) was administered online in Ho Chi Minh City, Vietnam in the early stages of the COVID-19 outbreak in 2020. The survey captured participants' risk perceptions, efficacy beliefs, campaign message exposure, behavioral intention, and demographic variables. Two-step cluster analysis, analysis of variance, and multinominal regression were conducted.ResultsThree segments were identified for staying at home behavior (avoidance group, proactive group, and responsive group) and four segments were found for mask wearing behavior (avoidance group, proactive group, moderately responsive group, and highly responsive group). Results revealed that segment membership was associated with behavioral intention and exposure to COVID-19 prevention message via social media predicted the likelihood of participants belonging to the responsive group.ConclusionData showed that segments differed significantly regarding behavioral intention and that exposure to social media prevention messages predicted the responsive group. Findings extended the RPA to a novel behavioral context, while providing useful implications for practitioners.