Socia Media and Healthcare Technology

Permanent URI for this collectionhttps://hdl.handle.net/10125/107490

Browse

Now showing 1 - 9 of 9

Influence of Twitter on Medication Prescriptions
(2024-01-03) Gorthi, Sai Soundarya; Ganju , Kartik; Pinsonneault, Alain
Social media is playing a critical role in disseminating health information, and anecdotal evidence suggests it is influencing the perceptions of healthcare providers. To understand the influence of social media on healthcare providers' clinical decisions, we examine the influence of Twitter discourse regarding Hydroxychloroquine for treating COVID-19 patients on the proportion of hydroxychloroquine prescriptions prescribed for COVID-19 patients in the USA in 2020. We collected Twitter data from Twitter API v2 and Hydroxychloroquine prescriptions in the USA from Symphony Health data on the COVID-19 research database. Econometric analysis of our data indicates that Twitter discourse has a positive influence on the hydroxychloroquine prescriptions prescribed in the USA in 2020. Our results are robust to time-invariant location-specific metrics. We discuss the possible pathways through which Twitter could have influenced healthcare providers' prescribing decisions.
Utilizing the Fediverse and AI-bots for Youth Engagement During COVID-19 in a Hybrid Preventative Intervention
(2024-01-03) Cole, Mason; Gary, Kevin; Meier, Matt; Gonzales, Nancy; Pina, Armando; Stoll, Ryan
Social media is a pervasive platform for delivering targeted interventions, albeit with cautionary ethical consequences. Recently AI robots or “bots” have been combined with social media platforms to enhance interaction, and enact behavior change through increased engagement and adherence to intervention protocols. This paper presents a customized social media platform for promoting engagement and adherence to a prevention intervention protocol. The protocol was originally developed in a group workshop format, and then online during COVID-19. A social media platform was utilized to connect group participants and deliver protocol activities. Bots encouraged participation via positive reinforcement mechanism for the entire group, and to remind a participant of protocol activities. While not a formal study, our exploratory results demonstrate that bots and a social media context support a group leader in increased engagement and adherence to the protocol. Our principal contribution in this paper is demonstrating that a personalized, adaptive instance of a Control Systems Engineering model may improve engagement-related outcomes in brief protocols.
Classifying Vaccine Misinformation in Online Social Media Videos using Natural Language Processing and Machine Learning
(2024-01-03) Schmidt, Sarah; Thoms, Brian; Eryilmaz, Evren; Isaacs, Jason
The spread of information through online social media videos is one of the most popular ways to share and obtain information, while at the same time the spread of misinformation across these same social spaces has become a significant concern affecting human well-being. Being able to detect this misinformation before it spreads is becoming more and more desirable for many social media platforms. This research focuses on exploring the accuracy of detecting misinformation across two social media platforms, YouTube and BitChute. This involves the classification of video data into two types: genuine information or misinformation. More specifically, this research generates additional metadata embedded within online videos related to the COVID-19 vaccination. Using natural language processing (NLP) we extract medical subject headings (MeSH) terms from video transcripts and classify videos using four machine learning techniques including naïve Bayes, random forest, support vector machine, and logistic regression. Implementation of each classifier is presented, and the accuracy of each technique is compared and discussed.
Pro-social framing and sentiment in U.S. broadcast networks’ Instagram posts about the COVID-19 vaccine
(2024-01-03) Ritchart, Amy; Britt, Rebecca
The study uses health risk frames to conduct 1) a content analysis and 2) a computational sentiment analysis to analyze the framing of the COVID-19 vaccine in Instagram posts by major broadcast news networks during its first year of availability in the United States. The analysis focused on the portrayal of the vaccine as a pro-social solution to the pandemic, with emphasis on the consequences of the pandemic, individual solutions from medical professionals grounded in science, and the vaccine as a preventive measure. Findings from qualitative content analysis (N = 45) included a reduced presence of alarmist, loss, conflict, and economic consequences frames, and the absence of uncertainty,indicating an explanatory framework and support for the vaccine. The computational sentiment analysis (N = 178) revealed a slightly positive and variable sentiment, suggesting an overall affirmative picture of the vaccine in the networks’ Instagram coverage.
Association Between Exposure to E-Cigarette Content Across 10 Social Media Sites and Youth E-Cigarette Use
(2024-01-03) Rajeshkumar, Lavanya; Morean, Meghan E.; Kong, Grace; Bold, Krysten; Krishnan-Sarin, Suchitra
Social media use is associated with youth vaping and susceptibility, but the relationship is not well-understood. We investigated the association between exposure to e-cigarette-related content on 10 social media sites and e-cigarette susceptibility/use via multinomial logistic regression. The analytic sample comprised 1863 students from 2 Connecticut high schools who completed an anonymous survey in 2020. Compared to youth who were not susceptible to e-cigarette use, those who were susceptible, had used e-cigarettes in their past, or currently using e-cigarettes endorsed significantly higher social media use. While exposure to e-cigarette content on most social media sites was positively associated with susceptibly to e-cigarettes and their use, it was negatively associated with e-cigarette use on one social media site. Further studies should examine the content of social media ads and posts to better understand their respective effects on youth vaping. This can inform marketing regulations and development of youth vaping prevention programs.
A Preliminary Examination of Alcohol-Related Social Networking Site Consequences Among Adolescents and Young Adults
(2024-01-03) Litt, Dana; Lewis, Melissa
This study investigated consequences of posting about alcohol on social networking sites and explore their associations with drinking behavior and social networking site-related cognitions. A sample of 306 participants aged 15-20 completed surveys from which the data is drawn. Results revealed that individuals who posted about alcohol on social networking sites experienced various consequences, including interpersonal conflicts, guilt, shame, and embarrassment. While legal and job repercussions were rare, unwanted attention and negative comments on posts were reported. Correlations demonstrated positive associations between social networking site consequences and drinking consequences, as well as with social norms and perceived vulnerability. Notably, the frequency of alcohol-related posts and willingness to engage in such behavior was not associated with consequences. Additional item-level correlations primarily supported the overall correlational findings. Findings highlight that further research is warranted to unravel the complex interplay between social networking site use, alcohol consumption, and related consequences.
Is Posting about Alcohol on Social Networking Sites Associated with Self-reported Alcohol-Induced Blackouts in a Diverse Sample of Non-College-Attending Young Adults?
(2024-01-03) Merrill, Jennifer; Davidson, Lily; Riordan, Benjamin; Logan, Zoey; Ward, Rose Marie
Research among young adults (YA), in samples of majority White college students, indicates links between posting about alcohol on social media and self-reported drinking behavior. We sought to extend this work by examining unique associations between public versus private posting about alcohol and the high-risk outcome of alcohol-related blackouts, among a sample of racially/ethnically diverse, non-college YA. A sample of 610 participants (74% non-White) completed an online survey about social media use and drinking behavior. Across three platforms (Twitter, Instagram, TikTok), public posting on Instagram was most common. A higher frequency of private posting about alcohol was associated with a higher frequency of past-month blackouts. Private posting on Instagram and TikTok (though not Twitter) were uniquely related to blackouts, as was public posting on TikTok. Results have implications for eventual online interventions, which can identify individuals potentially at risk for hazardous drinking, based on their social media posting behavior.
Enhancing Audiences’ Engagement with Health Education Videos: Visual Call or Textual Call?
(2024-01-03) Tong, Jingjing; Xu, Jingjun (David)
With the rising popularity of video-based social media, more and more doctors are utilizing them to provide health education videos and employing the call-to-action strategy to enhance video engagement. Although the effectiveness of the call-to-action has been established in customer-brand communication literature, it remains unknown how it affects engagement with health education videos. Drawing on multiple literature streams, we propose that visual call positively affects video engagement, while textual call has a negative effect. Additionally, the medical term use weakens the positive impact of visual call and strengthens the negative impact of textual call. Our analysis results support the positive impact of visual call and the negative effect of textual call on audiences’ liking and collecting behavior, which are contingent upon medical term use. Our study contributes to research on technology-mediated patient education and the literature on call-to-action and provides implications for doctors and platform managers.
Introduction to the Minitrack on Socia Media and Healthcare Technology
(2024-01-03) Bock, Beth; Braciszewski, Jordan

Browse

Recent Submissions