Vaccine Hesitancy towards COVID-19 Vaccination: Investigating the Role of Information Sources through a Mediation Analysis

Abstract

Mass vaccination campaigns have been implemented worldwide to counteract the SARS-CoV-2/COVID-19 pandemic, however their effectiveness could be challenged by vaccine hesitancy. The tremendous rise in the use of social media have made them acquire a leading role as an information source, thus representing a crucial factor at play that could contribute to increase or mitigate vaccine hesitancy, as information sources play a pivotal role in shaping public opinion and perceptions. The aims of the study were to investigate if information sources could affect the attitude towards COVID-19 vaccination and if they could act as a mediator in the relationship between individual characteristics and vaccine hesitancy. A cross-sectional online survey was conducted by a professional panellist on a representative sample of 1011 citizens from the Emilia-Romagna region in Italy in January 2021. A mediation analysis using structural equation modelling was performed. Our results show how social media directly or indirectly increases vaccine hesitancy towards COVID-19 vaccination, while the opposite effect was observed for institutional websites. Given the global widespread use of social media, their use should be enhanced to disseminate scientifically sound information to a greater audience to counteract vaccine hesitancy, while at the same time continuing to promote and update institutional websites that have proven to be effective in reducing vaccine hesitancy.

Keywords: COVID-19; SARS-CoV-2; information sources; mediation analysis; pandemic; social media; structural equation modelling; survey; vaccine acceptance; vaccine hesitancy.

Figure 1

Final mediation model. Red lines denote positive effect and blue lines denote negative effect. Numbers on red and blue arrows represent standardized regression coefficients. Green lines denote the perceived risk of infection latent factor and the relationship with the corresponding observed variables. Numbers on green arrows represent standardized factor loadings.