Exploring Political Mistrust in Pandemic Risk Communication: Mixed-Method Study Using Social Media Data Analysis

"The findings highlight the critical role of trust in effective public health management and underscore the need for strategic communication to foster this trust, thereby informing public health communication and policy."

Risk communication is an integrated part of any emergency response that consists of a real-time exchange of information, advice, and opinions between experts, community leaders, officials, and the people who are at risk. In Finland, the Finnish Institute for Health and Welfare (THL) developed tools that monitored the trust of the public toward authorities during COVID-19 through surveys and qualitative social media analysis. The THL then conducted a qualitative study based on Facebook and Twitter narratives to identify and describe factors linked with pandemic-related risk perception. The purpose of the study was to identify concepts that can be embedded in a digital platform to serve as keywords for risk communicators to conduct social listening during future epidemics and pandemics. Using a big data set spanning 3 years, the present study analysed communications between THL and its audience on Facebook and Twitter. The paper aims to investigate variations in trust levels over time and across social media platforms, as well as to further explore subcategories of political mistrust. It seeks to understand the dynamics of political trust and to compare qualitative research findings with those obtained through computational methods.

Data were gathered from a large-scale data set consisting of 13,629 Twitter and Facebook posts from 2020 to 2023 related to COVID-19. For analysis, a fine-tuned FinBERT model with an 80% accuracy rate was used for predicting political mistrust. The BERTopic model was also used for superior topic modeling performance.

Given that most people who engage with THL on social media do so because they are concerned about COVID-19, a large proportion of the posts are either trustful or neutral toward the THL. However, there is also a significant amount of mistrust, which fluctuates considerably throughout the pandemic. Mistrust peaked in early 2020 when the pandemic was spreading rapidly and causing global alarm. However, mistrust declined around April 2020, as people became more accustomed to the situation. Twitter users expressed more mistrust than Facebook users throughout the pandemic, in part due to the presence of malicious bots that targeted THL's social media accounts with misinformation and antivaccination propaganda.

The preliminary analysis identifies 43 mistrust-related topics categorised into 9 major themes. The most salient topics include COVID-19 mortality, coping strategies, polymerase chain reaction testing, and vaccine efficacy. One theme is "insufficient information on COVID-19 health effects and complications". This theme encapsulates the public's ongoing quest for accurate and detailed information on a wide array of subjects. Per the researchers, understanding these aspects is instrumental in shaping the public's perception of the risk associated with the virus. For instance, a lack of comprehensive knowledge about the range of symptoms may hinder early detection and accurate self-assessment of personal risk. "The accessibility and availability of comprehensive, accurate information regarding the health impacts of COVID-19 is paramount in mitigating misconceptions and facilitating a balanced understanding of risk, thereby directly influencing public health outcomes."

The wide-ranging social effects of the pandemic and the complex behavioural responses from the public were another critical area of focus in the social media data studied here. For example, the study revealed a degree of public dissatisfaction with government and public health communications, especially pertaining to the clarity and consistency of messaging about COVID-19. In fact, "policy choices such as school openings, mask mandates, and vaccination campaigns have been met with varying degrees of acceptance, compliance, and resistance worldwide. These variations in public responses underscore the need for public health policies that are not only scientifically grounded but also socially attuned, taking into account the public's attitudes, beliefs, and cultural contexts."

A lack of trust was found to potentially elevate the spread of misinformation and disinformation, thus negatively influencing adherence to public health measures. This finding underscores the need to ensure public comprehension of key preventive measures and treatments. Because providing facts alone does not stop the spread of misinformation and disinformation, strategies such as debunking and rebunking are often recommended. Discontent stemming from the information provided by governmental authorities highlights the importance of effective risk communication - not only during the acute phase of a crisis but also in the recovery phase - to preserve public trust and curb the dissemination of misinformation and disinformation.

Discourse related to mistrust in authority is associated with perceptions of disease severity, willingness to adopt health measures, and information-seeking behaviour. The paper highlights how the distinct user engagement mechanisms and platform features of Facebook and Twitter contributed to varying patterns of mistrust and susceptibility to misinformation during the pandemic. "Specifically, Twitter is more vulnerable to manipulation by trolls and malicious bots in the Finnish case. The dynamic changes in public attention on various pandemic-related topics revealed through this study indicate that trust can be fostered by tailoring messages to align with public trends and concerns, and by devising platform-specific communication strategies that account for the potential risks and limitations of different platforms."

In terms of research methodologies, the researchers note that, given that COVID-19 has resulted in an "infodemic" where people mainly seek and access information through the internet, computational methods are needed to handle the large volume of user engagement and provide guidance to public health agencies. The results of this study demonstrate that the new developments in natural language processing (NLP) methods can not only capture the nuances identified by qualitative studies but also several other topics that may be of interest to public health agencies.

In conclusion, this study highlights the effectiveness of computational methods like NLP in managing large-scale engagement and misinformation and "underscores the critical role of trust in health authorities for effective risk communication and public compliance. The findings also emphasize the necessity for transparent communication from authorities, concluding that a holistic approach to public health communication is integral for managing health crises effectively."