Revealing Public Opinion Towards COVID-19 Vaccines With Twitter Data in the United States: Spatiotemporal Perspective

Examining Public Sentiments and Attitudes Toward COVID-19 Vaccination: Infoveillance Study Using Twitter Posts

Background

A global rollout of vaccinations is currently underway to mitigate and protect people from the COVID-19 pandemic. Several individuals have been using social media platforms such as Twitter as an outlet to express their feelings, concerns, and opinions about COVID-19 vaccines and vaccination programs. This study examined COVID-19 vaccine–related tweets from January 1, 2020, to April 30, 2021, to uncover the topics, themes, and variations in sentiments of public Twitter users.

Objective

The aim of this study was to examine key themes and topics from COVID-19 vaccine–related English tweets posted by individuals, and to explore the trends and variations in public opinions and sentiments.

Methods

We gathered and assessed a corpus of 2.94 million COVID-19 vaccine–related tweets made by 1.2 million individuals. We used CoreX topic modeling to explore the themes and topics underlying the tweets, and used VADER sentiment analysis to compute sentiment scores and examine weekly trends. We also performed qualitative content analysis of the top three topics pertaining to COVID-19 vaccination.

Results

Topic modeling yielded 16 topics that were grouped into 6 broader themes underlying the COVID-19 vaccination tweets. The most tweeted topic about COVID-19 vaccination was related to vaccination policy, specifically whether vaccines needed to be mandated or optional (13.94%), followed by vaccine hesitancy (12.63%) and postvaccination symptoms and effects (10.44%) Average compound sentiment scores were negative throughout the 16 weeks for the topics postvaccination symptoms and side effects and hoax/conspiracy. However, consistent positive sentiment scores were observed for the topics vaccination disclosure, vaccine efficacy, clinical trials and approvals, affordability, regulation, distribution and shortage, travel, appointment and scheduling, vaccination sites, advocacy, opinion leaders and endorsement, and gratitude toward health care workers. Reversal in sentiment scores in a few weeks was observed for the topics vaccination eligibility and hesitancy.

Conclusions

Identification of dominant themes, topics, sentiments, and changing trends about COVID-19 vaccination can aid governments and health care agencies to frame appropriate vaccination programs, policies, and rollouts.

Spatial social network research: a bibliometric analysis

A restless and dynamic intellectual landscape has taken hold in the field of spatial social network studies, given the increasingly attention towards fine-scale human dynamics in this urbanizing and mobile world. The measuring parameters of such dramatic growth of the literature include scientific outputs, domain categories, major journals, countries, institutions, and frequently used keywords. The research in the field has been characterized by fast development of relevant scholarly articles and growing collaboration among and across institutions. The Journal of Economic Geography, Annals of the Association of American Geographers, and Urban Studies ranked first, second, and third, respectively, according to average citations. The United States, United Kingdom, and China were the countries that yielded the most published studies in the field. The number of international collaborative studies published in non-native English-speaking countries (such as France, Italy, and the Netherlands) were higher than native English-speaking countries. Wuhan University, the University of Oxford, and Harvard University were the universities that published the most in the field. "Twitter", "big data", "networks", "spatial analysis", and "social capital" have been the major keywords over the past 20 years. At the same time, the keywords such as "social media", "Twitter", "big data", "geography", "China", "human mobility", "machine learning", "GIS", "location-based social networks", "clustering", "data mining", and "location-based services" have attracted increasing attention in that same time frame, indicating the future research trends.

Integrating Google Trends Search Engine Query Data Into Adult Emergency Department Volume Forecasting: Infodemiology Study

Background

The search for health information from web-based resources raises opportunities to inform the service operations of health care systems. Google Trends search query data have been used to study public health topics, such as seasonal influenza, suicide, and prescription drug abuse; however, there is a paucity of literature using Google Trends data to improve emergency department patient-volume forecasting.

Objective

We assessed the ability of Google Trends search query data to improve the performance of adult emergency department daily volume prediction models.

Methods

Google Trends search query data related to chief complaints and health care facilities were collected from Chicago, Illinois (July 2015 to June 2017). We calculated correlations between Google Trends search query data and emergency department daily patient volumes from a tertiary care adult hospital in Chicago. A baseline multiple linear regression model of emergency department daily volume with traditional predictors was augmented with Google Trends search query data; model performance was measured using mean absolute error and mean absolute percentage error.

Results

There were substantial correlations between emergency department daily volume and Google Trends "hospital" (r=0.54), combined terms (r=0.50), and "Northwestern Memorial Hospital" (r=0.34) search query data. The final Google Trends data-augmented model included the predictors Combined 3-day moving average and Hospital 3-day moving average and performed better (mean absolute percentage error 6.42%) than the final baseline model (mean absolute percentage error 6.67%)-an improvement of 3.1%.

Conclusions

The incorporation of Google Trends search query data into an adult tertiary care hospital emergency department daily volume prediction model modestly improved model performance. Further development of advanced models with comprehensive search query terms and complementary data sources may improve prediction performance and could be an avenue for further research.

ANTi-Vax: a novel Twitter dataset for COVID-19 vaccine misinformation detection

OBJECTIVES

COVID-19 (SARS-CoV-2) pandemic has infected hundreds of millions and inflicted millions of deaths around the globe. Fortunately, the introduction of COVID-19 vaccines provided a glimmer of hope and a pathway to recovery. However, owing to misinformation being spread on social media and other platforms, there has been a rise in vaccine hesitancy which can lead to a negative impact on vaccine uptake in the population. The goal of this research is to introduce a novel machine learning-based COVID-19 vaccine misinformation detection framework.

STUDY DESIGN

We collected and annotated COVID-19 vaccine tweets and trained machine learning algorithms to classify vaccine misinformation.

METHODS

More than 15,000 tweets were annotated as misinformation or general vaccine tweets using reliable sources and validated by medical experts. The classification models explored were XGBoost, LSTM, and BERT transformer model.

RESULTS

The best classification performance was obtained using BERT, resulting in 0.98 F1-score on the test set. The precision and recall scores were 0.97 and 0.98, respectively.

CONCLUSION

Machine learning-based models are effective in detecting misinformation regarding COVID-19 vaccines on social media platforms.

Exploring public perceptions of the COVID-19 vaccine online from a cultural perspective: Semantic network analysis of two social media platforms in the United States and China

The development and uptake of the COVID-19 (coronavirus disease 2019) vaccine is a top priority in stifling the COVID-19 pandemic. How the public perceives the COVID-19 vaccine is directly associated with vaccine compliance and vaccination coverage. This study takes a cultural sensitivity perspective and adopts two well-known social media platforms in the United States (Twitter) and China (Weibo) to conduct a public perception comparison around the COVID-19 vaccine. By implementing semantic network analysis, results demonstrate that the two countries' social media users overlapped in themes concerning domestic vaccination policies, priority groups, challenges from COVID-19 variants, and the global pandemic situation. However, Twitter users were prone to disclose individual vaccination experiences, express anti-vaccine attitudes. In comparison, Weibo users manifested evident deference to authorities and exhibited more positive feelings toward the COVID-19 vaccine. Those disparities were explained by the cultural characteristics' differences between the two countries. The findings provide insights into comprehending public health issues in cross-cultural contexts and illustrate the potential of utilizing social media to conduct health informatics studies and investigate public perceptions during public health crisis time.

Spatial Analysis of COVID-19 Vaccination: A Scoping Review

Spatial analysis of COVID-19 vaccination research is increasing in recent literature due to the availability of COVID-19 vaccination data that usually contain location components. However, to our knowledge, no previous study has provided a comprehensive review of this research area. Therefore, in this scoping review, we examined the breadth of spatial and spatiotemporal vaccination studies to summarize previous findings, highlight research gaps, and provide guidelines for future research. We performed this review according to the five-stage methodological framework developed by Arksey and O'Malley. We screened all articles published in PubMed/MEDLINE, Scopus, and Web of Science databases, as of 21 September 2021, that had employed at least one form of spatial analysis of COVID-19 vaccination. In total, 36 articles met the inclusion criteria and were organized into four main themes: disease surveillance (n = 35); risk analysis (n = 14); health access (n = 16); and community health profiling (n = 2). Our findings suggested that most studies utilized preliminary spatial analysis techniques, such as disease mapping, which might not lead to robust inferences. Moreover, few studies addressed data quality, modifiable areal unit problems, and spatial dependence, highlighting the need for more sophisticated spatial and spatiotemporal analysis techniques.

Differences in Sources of Information, Risk Perception, and Cognitive Appraisals between People with Various Latent Classes of Motivation to Get Vaccinated against COVID-19 and Previous Seasonal Influenza Vaccination: Facebook Survey Study with Latent Profile Analysis in Taiwan

The present study aimed (1) to identify distinct latent classes of motivation to get vaccinated against coronavirus disease 2019 (COVID-19) and previous seasonal influenza vaccination among people in Taiwan and (2) to examine the roles of sources of information, risk perception, and cognitive appraisals of vaccination against COVID-19 in these classes. We recruited 1047 participants through a Facebook advertisement. The participants' motivation to get vaccinated against COVID-19, previous seasonal influenza vaccination, sources of information about COVID-19 vaccination, risk perception of COVID-19, and cognitive appraisals of vaccination against COVID-19 were determined. We examined the participants' motivation for COVID-19 vaccination and previous seasonal influenza vaccination through latent profile analysis. Four latent classes of motivation were identified: participants with high motivation for COVID-19 vaccination and high seasonal influenza vaccination, those with high motivation for COVID-19 vaccination but low seasonal influenza vaccination, those with low motivation for COVID-19 vaccination but high seasonal influenza vaccination, and those with low motivation for COVID-19 vaccination and low seasonal influenza vaccination. Compared with participants in the latent class of high motivation for COVID-19 vaccination and high seasonal influenza vaccination, those in the other three latent classes had lower levels of positive appraisals of COVID-19 vaccination; participants in the latent class of low motivation for COVID-19 vaccination and low seasonal influenza vaccination had lower risk perception of COVID-19 and were also less likely to obtain information about COVID-19 vaccination from the internet, friends, and family members. The various motivations and behaviors for vaccination, sources of information, risk perception, and cognitive appraisals of vaccination against COVID-19 should be considered in intervention programs aiming to increase people's motivation to get vaccinated against COVID-19.