An Integrated Approach to Evaluating the Patterns and Impacts of Vaccine Confidence on Meningitis Dynamics in Nigeria and the UK

Abstract

Anti-vaccine opinions have long charged vaccine refusal in countries across the world, due to myths, low levels of infection and the religious and cultural convictions of individuals. This research explores the patterns and ramifications of vaccine sentiment on disease dynamics, specifically for meningitis, in Nigeria and the United Kingdom (UK). To address this research question, surveys and tweets are first employed to probe and quantify trends in vaccine confidence and vaccine-related conversations. Tweets are sampled from British and Nigerian Twitter users, while surveys are conducted in two states in northern and southern Nigeria (Zamfara and Oyo respectively). A mathematical model of meningococcal meningitis transmission is also constructed, which accounts for the impacts of vaccine confidence on overall coverage. Findings from surveys report elevated levels of awareness for more recent disease threats like meningitis, compared to presently low-burden infections like yellow fever. Furthermore, high levels of vaccine confidence and support are reported in both Zamfara and Oyo. Scores for various categories of vaccine confidence are found to be largely consistent across different sub-populations in both states. However, composite vaccine confidence is significantly different among those surveyed in Oyo and Zamfara. Nevertheless, this difference is marginal, revealing a convergence of opinion. Furthermore, results in both states suggest that the sub-score representing convenience in vaccine access (both physical and financial) is a crucial determinant of vaccine confidence overall. Analyses of historic and recent Twitter data suggest that tweeting communities are very responsive, often in a widely polarizing manner, to major vaccine-related news stories from within and outside their borders. Significant, but not substantive, differences in mean polarity between historic British and Nigerian tweets is reported, but not for recent British and Nigerian tweets. Sentiment scores across all samples are low and positive and predominantly reflect neutral content, particularly in Nigeria. Tweets are mainly sourced from urban areas in both nations, but major hubs are also traced to several agricultural regions, in the case of Nigeria. Finally, results from the meningitis transmission model suggest that increasing vaccine confidence can limit both carrier and infected populations in Nigeria. When the model is applied to the context of the UK, the impact of increased vaccine confidence on suppressing the burden of infection is only observed after a time lag. Additionally, higher levels of vaccine confidence have a more muted effect on infected and carrier populations in the UK than is predicted in Nigeria. Ultimately, this project aims to paint a more realistic portrait of communities’ interactions with disease, as opinion-driven threats to vaccine uptake may preclude the achievement of vaccination targets or result in dwindling coverage, both of which can lead to unexpected epidemics.