Dataset Information

Social media study of public opinions on potential COVID-19 vaccines: informing dissent, disparities, and dissemination.

ABSTRACT: Background: The current development of vaccines for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is unprecedented. Little is known, however, about the nuanced public opinions on the vaccines on social media. Methods: We adopt a human-guided machine learning framework using more than six million tweets from almost two million unique Twitter users to capture public opinions on the vaccines for SARS-CoV-2, classifying them into three groups: pro-vaccine, vaccine-hesitant, and anti-vaccine. After feature inference and opinion mining, 10,945 unique Twitter users are included in the study population. Multinomial logistic regression and counterfactual analysis are conducted. Results: Socioeconomically disadvantaged groups are more likely to hold polarized opinions on coronavirus disease 2019 (COVID-19) vaccines either pro-vaccine ( B=0.40,SE=0.08,P<.001,OR=1.49;95%CI=[1.26,1.75] ) or anti-vaccine ( B=0.52,SE=0.06,P<.001,OR=1.69;95%CI=[1.49,1.91] ). People who have the worst personal pandemic experience are more likely to hold the anti-vaccine opinion ( B=-0.18,SE=0.04,P<.001,OR=0.84;95%CI=[0.77,0.90] ). The U.S. public is most concerned about the safety, effectiveness, and political issues regarding vaccines for COVID-19, and improving personal pandemic experience increases the vaccine acceptance level. Conclusion: Opinion on COVID-19 vaccine uptake varies across people of different characteristics.

SUBMITTER: Lyu H

PROVIDER: S-EPMC8384764 | biostudies-literature | 2021 Aug

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Social media study of public opinions on potential COVID-19 vaccines: informing dissent, disparities, and dissemination.

Lyu Hanjia H Wang Junda J Wu Wei W Duong Viet V Zhang Xiyang X Dye Timothy D TD Luo Jiebo J

Intelligent medicine 20210825 1

<b>Background</b> The current development of vaccines for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is unprecedented. Little is known, however, about the nuanced public opinions on the vaccines on social media. <b>Methods</b> We adopted a human-guided machine learning framework using more than six million tweets from almost two million unique Twitter users to capture public opinions on the vaccines for SARS-CoV-2, classifying them into three groups: pro-vaccine, vaccine-hesita ...[more]

PMID: 34457371

Similar Datasets

Project description:BackgroundSocial media public health campaigns have the advantage of tailored messaging at low cost and large reach, but little is known about what would determine their feasibility as tools for inducing attitude and behavior change.ObjectiveThe aim of this study was to test the feasibility of designing, implementing, and evaluating a social media-enabled intervention for skin cancer prevention.MethodsA quasi-experimental feasibility study used social media (Twitter) to disseminate different message "frames" related to care in the sun and cancer prevention. Phase 1 utilized the Northern Ireland cancer charity's Twitter platform (May 1 to July 14, 2015). Following a 2-week "washout" period, Phase 2 commenced (August 1 to September 30, 2015) using a bespoke Twitter platform. Phase 2 also included a Thunderclap, whereby users allowed their social media accounts to automatically post a bespoke message on their behalf. Message frames were categorized into 5 broad categories: humor, shock or disgust, informative, personal stories, and opportunistic. Seed users with a notable following were contacted to be "influencers" in retweeting campaign content. A pre- and postintervention Web-based survey recorded skin cancer prevention knowledge and attitudes in Northern Ireland (population 1.8 million).ResultsThere were a total of 417,678 tweet impressions, 11,213 engagements, and 1211 retweets related to our campaign. Shocking messages generated the greatest impressions (shock, n=2369; informative, n=2258; humorous, n=1458; story, n=1680), whereas humorous messages generated greater engagement (humorous, n=148; shock, n=147; story, n=117; informative, n=100) and greater engagement rates compared with story tweets. Informative messages, resulted in the greatest number of shares (informative, n=17; humorous, n=10; shock, n=9; story, n=7). The study findings included improved knowledge of skin cancer severity in a pre- and postintervention Web-based survey, with greater awareness that skin cancer is the most common form of cancer (preintervention: 28.4% [95/335] vs postintervention: 39.3% [168/428] answered "True") and that melanoma is most serious (49.1% [165/336] vs 55.5% [238/429]). The results also show improved attitudes toward ultraviolet (UV) exposure and skin cancer with a reduction in agreement that respondents "like to tan" (60.5% [202/334] vs 55.6% [238/428]).ConclusionsSocial media-disseminated public health messages reached more than 23% of the Northern Ireland population. A Web-based survey suggested that the campaign might have contributed to improved knowledge and attitudes toward skin cancer among the target population. Findings suggested that shocking and humorous messages generated greatest impressions and engagement, but information-based messages were likely to be shared most. The extent of behavioral change as a result of the campaign remains to be explored, however, the change of attitudes and knowledge is promising. Social media is an inexpensive, effective method for delivering public health messages. However, existing and traditional process evaluation methods may not be suitable for social media.

Project description:BACKGROUND:Society always has limited resources to expend on health care, or anything else. What are the unmet medical needs? How do we allocate limited resources to maximize the health and welfare of the people? These challenging questions might be re-examined systematically within an infodemiological frame on a much larger scale, leveraging the latest advancement in information technology and data science. OBJECTIVE:We expanded our previous work by investigating news media data to reveal the coverage of different diseases and medical conditions, together with their sentiments and topics in news articles over three decades. We were motivated to do so since news media plays a significant role in politics and affects the public policy making. METHODS:We analyzed over 3.5 million archive news articles from Reuters media during the periods of 1996/1997, 2008 and 2016, using summary statistics, sentiment analysis, and topic modeling. Summary statistics illustrated the coverage of various diseases and medical conditions during the last 3 decades. Sentiment analysis and topic modeling helped us automatically detect the sentiments of news articles (ie, positive versus negative) and topics (ie, a series of keywords) associated with each disease over time. RESULTS:The percentages of news articles mentioning diseases and medical conditions were 0.44%, 0.57% and 0.81% in the three time periods, suggesting that news media or the public has gradually increased its interests in medicine since 1996. Certain diseases such as other malignant neoplasm (34%), other infectious diseases (20%), and influenza (11%) represented the most covered diseases. Two hundred and twenty-six diseases and medical conditions (97.8%) were found to have neutral or negative sentiments in the news articles. Using topic modeling, we identified meaningful topics on these diseases and medical conditions. For instance, the smoking theme appeared in the news articles on other malignant neoplasm only during 1996/1997. The topic phrases HIV and Zika virus were linked to other infectious diseases during 1996/1997 and 2016, respectively. CONCLUSIONS:The multi-dimensional analysis of news media data allows the discovery of focus, sentiments and topics of news media in terms of diseases and medical conditions. These infodemiological discoveries could shed light on unmet medical needs and research priorities for future and provide guidance for the decision making in public policy.

Project description:BackgroundDuring an infodemic, timely, reliable, and accessible information is crucial to combat the proliferation of health misinformation. While message testing can provide vital information to make data-informed decisions, traditional methods tend to be time- and resource-intensive. Recognizing this need, we developed the rapid message testing at scale (RMTS) approach to allow communicators to repurpose existing social media advertising tools and understand the full spectrum of audience engagement.ObjectiveWe had two main objectives: (1) to demonstrate the use of the RMTS approach for message testing, especially when resources and time are limited, and (2) to propose and test the efficacy of an outcome variable that measures engagement along a continuum of viewing experience.MethodsWe developed 12 versions of a single video created for a vaccine confidence project in India. We manipulated video length, aspect ratio, and use of subtitles. The videos were tested across 4 demographic groups (women or men, younger or older). We assessed user engagement along a continuum of viewing experience: obtaining attention, sustaining attention, conveying the message, and inspiring action. These were measured by the percentage of video watched and clicks on the call-to-action link.ResultsThe video advertisements were placed on Facebook for over 4 consecutive days at the cost of US $450 and garnered a total of 3.34 million impressions. Overall, we found that the best-performing video was the shorter version in portrait aspect ratio and without subtitles. There was a significant but small association between the length of the video and users' level of engagement at key points along the continuum of viewing experience (N=1,032,888; χ24=48,261.97; P<.001; V=.22). We found that for the longer video, those with subtitles held viewers longer after 25% video watch time than those without subtitles (n=15,597; χ21=7.33; P=.007; V=.02). While we found some significant associations between the aspect ratio, the use of subtitles, and the number of users watching the video and clicking on the call-to-action link, the effect size for those were extremely small.ConclusionsThis test served as a proof of concept for the RMTS approach. We obtained rapid feedback on formal message attributes from a very large sample. The results of this test reinforce the need for platform-specific tailoring of communications. While our data showed a general preference for a short video in portrait orientation and without subtitles among our target audiences on Facebook, that may not necessarily be the case in other social media platforms such as YouTube or TikTok, where users go primarily to watch videos. RMTS testing highlights nuances that communication professionals can address instead of being limited to a "one size fits all" approach.

Project description:BACKGROUND:Depression is a common disorder that still remains underdiagnosed and undertreated in the UK National Health Service. Charities and voluntary organizations offer mental health services, but they are still struggling to promote these services to the individuals who need them. By analyzing social media (SM) content using machine learning techniques, it may be possible to identify which SM users are currently experiencing low mood, thus enabling the targeted advertising of mental health services to the individuals who would benefit from them. OBJECTIVE:This study aimed to understand SM users' opinions of analysis of SM content for depression and targeted advertising on SM for mental health services. METHODS:A Web-based, mixed methods, cross-sectional survey was administered to SM users aged 16 years or older within the United Kingdom. It asked participants about their demographics, their usage of SM, and their history of depression and presented structured and open-ended questions on views of SM content being analyzed for depression and views on receiving targeted advertising for mental health services. RESULTS:A total of 183 participants completed the survey, and 114 (62.3%) of them had previously experienced depression. Participants indicated that they posted less during low moods, and they believed that their SM content would not reflect their depression. They could see the possible benefits of identifying depression from SM content but did not believe that the risks to privacy outweighed these benefits. A majority of the participants would not provide consent for such analysis to be conducted on their data and considered it to be intrusive and exposing. CONCLUSIONS:In a climate of distrust of SM platforms' usage of personal data, participants in this survey did not perceive that the benefits of targeting advertisements for mental health services to individuals analyzed as having depression would outweigh the risks to privacy. Future work in this area should proceed with caution and should engage stakeholders at all stages to maximize the transparency and trustworthiness of such research endeavors.

Project description:Foodborne illness is prevented by inspection and surveillance conducted by health departments across America. Appropriate restaurant behavior is enforced and monitored via public health inspections. However, surveillance coverage provided by state and local health departments is insufficient in preventing the rising number of foodborne illness outbreaks. To address this need for improved surveillance coverage we conducted a supplementary form of public health surveillance using social media data: Yelp.com restaurant reviews in the city of San Francisco. Yelp is a social media site where users post reviews and rate restaurants they have personally visited. Presence of keywords related to health code regulations and foodborne illness symptoms, number of restaurant reviews, number of Yelp stars, and restaurant price range were included in a model predicting a restaurant's likelihood of health code violation measured by the assigned San Francisco public health code rating. For a list of major health code violations see (S1 Table). We built the predictive model using 71,360 Yelp reviews of restaurants in the San Francisco Bay Area. The predictive model was able to predict health code violations in 78% of the restaurants receiving serious citations in our pilot study of 440 restaurants. Training and validation data sets each pulled data from 220 restaurants in San Francisco. Keyword analysis of free text within Yelp not only improved detection of high-risk restaurants, but it also served to identify specific risk factors related to health code violation. To further validate our model we applied the model generated in our pilot study to Yelp data from 1,542 restaurants in San Francisco. The model achieved 91% sensitivity 74% specificity, area under the receiver operator curve of 98%, and positive predictive value of 29% (given a substandard health code rating prevalence of 10%). When our model was applied to restaurant reviews in New York City we achieved 74% sensitivity, 54% specificity, area under the receiver operator curve of 77%, and positive predictive value of 25% (given a prevalence of 12%). Model accuracy improved when reviews ranked highest by Yelp were utilized. Our results indicate that public health surveillance can be improved by using social media data to identify restaurants at high risk for health code violation. Additionally, using highly ranked Yelp reviews improves predictive power and limits the number of reviews needed to generate prediction. Use of this approach as an adjunct to current risk ranking of restaurants prior to inspection may enhance detection of those restaurants participating in high risk practices that may have gone previously undetected. This model represents a step forward in the integration of social media into meaningful public health interventions.

Dataset Information

Social media study of public opinions on potential COVID-19 vaccines: informing dissent, disparities, and dissemination.

Publications

Social media study of public opinions on potential COVID-19 vaccines: informing dissent, disparities, and dissemination.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets