The successful elimination of vaccine-preventable diseases is contingent on high-vaccine coverage rates in targeted populations. The proliferation of vaccine misinformation on social media has led to vaccine hesitancy in the past two decades. A highly contextual phenomenon, areas with an increased prevalence of vaccine hesitancy and vaccine exemption have been shown to correlate with decreased immunization coverage and intermittent vaccine-preventable disease outbreaks worldwide. Although the Canadian government has recommended the use of social media to increase public confidence in vaccines, little documentation exists regarding the perceptions of advisors and decision-makers in policy and communications for immunization towards vaccine hesitancy on social media, and the use of social media to increase public confidence in vaccines in the context of Ontario, Canada’s largest province. This thesis employed 3 unique mixed-methods studies to explore the role of social media in addressing the problem of vaccine hesitancy facilitated through misinformation about childhood vaccines in Ontario. The first study is a social network analysis that incorporates sentiment analysis to demonstrate that pro-vaccine and anti-vaccine communities operate in siloes with little interaction with one another. Those interactions that do occur are most commonly facilitated by sentiment and geographic location, rather than profession or affiliation of the social media user. The second study is a mixed methods content analysis illustrating significant differences in user attributes (emotion, medium shared in tweets, direction of information-sharing, and use of Twitter functions) among pro-, neutral, and anti-vaccine Twitter users, suggesting different motivations underlying Twitter use. Qualitative inquiry of links and reasons for negative vaccine sentiment illustrate the proliferation of pseudo-experts occupying social media, as well as concerns about vaccine safety and mistrust towards the government. The third study complements the first two studies, and uses documents and in-depth interviews with 23 advisers and decisionmakers in policy and communications to illustrate that although vaccine hesitancy is of concern, the use of social media to increase public confidence in vaccines is met with resistance due to a myriad of barriers at all levels of immunization policy and program delivery in the Province of Ontario. Implications for policy and practice of this study include the recognition that a multi-pronged approach is needed to increase the public’s confidence in vaccines. Elements of this multi-pronged approach could include: i) commitments to investing in understanding social media’s use in informing immunization at all levels of governance and decision-making; ii) the active surveillance of public sentiment and the public’s concerns about vaccines on social media using network analysis and content analysis; and iii) the fostering of interdisciplinary collaboration to design interventions that facilitate connectivity between siloes. The implications for future research include the need for continued commitment to the design, implementation, and evaluation of public health interventions on social media in the Ontario context. This study points to the need to pay attention to the behavioral attributes and affordances of social media in order to develop policies, communicative strategies, and programmatic designs that comprehensively address public concerns towards vaccines and, in turn, promote increased confidence in them. / Thesis / Doctor of Philosophy (PhD) / Immunization efforts are integral to maintaining herd immunity. Over the past two decades, it has been observed that vaccine hesitancy brought forth by vaccine misinformation has led to reduced confidence in vaccines, contributing to declining vaccination rates that have subsequently led to outbreaks of vaccine-preventable diseases. Vaccine misinformation on social media has played a crucial role in exacerbating vaccine hesitancy. Limited research has explored the use of social media in the Canadian context in relation to how vaccine information is communicated, what is being discussed and with whom. The extent to which decision-makers working in the immunization policy arena in Canada consider the role of social media as a tool for addressing vaccine hesitancy in order to increase vaccine uptake is also unclear. Using a mixed methods approach, this study, carried out in Ontario, Canada, illustrates that communities supporting and opposing vaccines operate in silos that do not necessarily communicate with each other through social media. Although decision-makers acknowledge the role of social media in the salience of vaccine hesitancy, they consider social media to be a less feasible method to increase vaccine confidence. By exploring the networks and conversations about vaccination on social media, and by understanding decision-makers’ perceptions towards vaccine hesitancy and social media, this study identified gaps between the recommendations for addressing vaccine hesitancy, provincial decision-makers’ preference for addressing immunization, and concerns of the vaccine hesitant on social media. These findings can inform the design of public health messaging to increase the public’s confidence in vaccines in Ontario.

Ma thèse a permis de développer une nouvelle approche pour étudier, par imagerie in vivo, le comportement des cellules présentatrices d’antigènes (CPA) de la peau suite à la vaccination par voie intradermique chez le primate non-humain. Le ciblage des CPA a été réalisé par injection in vivo d’un anticorps monoclonal anti-HLA-DR fluorescent. L’effet sur les CPA d’un adjuvant (R-848, ligand du TLR7/8) et l’immuno-ciblage des cellules de Langerhans par une protéine de fusion vaccinale anti-VIH (anti-langérine-VIHGag) ont ainsi été évalués par imagerie in vivo, vidéomicroscopie confocale ex vivo et cytométrie en flux. Ce travail a contribué à améliorer nos connaissances immunologiques sur les effets locaux et précoces des vaccins et /ou adjuvants. / My pHD project allowed the development of in vivo imaging approaches to study the skin antigen presenting cell (APC) behavior post-intradermal vaccination in non-human primates. APC targeting was performed by in vivo injection of fluorescent anti-HLA-DR monoclonal antibody. The effect of an adjuvant (R-848, ligand of TLR7/8) on skin APC and the immunotargeting of Langerhans cells by anti-HIV vaccinal fusion protein (anti-langerin-HIVGag) were assessed by in vivo fluorescent imaging, ex vivo confocal videomicroscopy and flow cytometry. This work contributed to improve immunological knowledge on local and early events post-vaccination with or without adjuvant.