Tweeting for Health Using Real-time Mining and Artificial Intelligence-Based Analytics: Design and Development of a Big Data Ecosystem for Detecting and Analyzing Misinformation on Twitter.

BACKGROUND: Digital misinformation, primarily on social media, has led to harmful and costly beliefs in the general population. Notably, these beliefs have resulted in public health crises to the detriment of governments worldwide and their citizens. However, public health officials need access to a comprehensive system capable of mining and analyzing large volumes of social media data in real time. OBJECTIVE: This study aimed to design and develop a big data pipeline and ecosystem (UbiLab Misinformation Analysis System [U-MAS]) to identify and analyze false or misleading information disseminated via social media on a certain topic or set of related topics. METHODS: U-MAS is a platform-independent ecosystem developed in Python that leverages the Twitter V2 application programming interface and the Elastic Stack. The U-MAS expert system has 5 major components: data extraction framework, latent Dirichlet allocation (LDA) topic model, sentiment analyzer, misinformation classification model, and Elastic Cloud deployment (indexing of data and visualizations). The data extraction framework queries the data through the Twitter V2 application programming interface, with queries identified by public health experts. The LDA topic model, sentiment analyzer, and misinformation classification model are independently trained using a small, expert-validated subset of the extracted data. These models are then incorporated into U-MAS to analyze and classify the remaining data. Finally, the analyzed data are loaded into an index in the Elastic Cloud deployment and can then be presented on dashboards with advanced visualizations and analytics pertinent to infodemiology and infoveillance analysis. RESULTS: U-MAS performed efficiently and accurately. Independent investigators have successfully used the system to extract significant insights into a fluoride-related health misinformation use case (2016 to 2021). The system is currently used for a vaccine hesitancy use case (2007 to 2022) and a heat wave-related illnesses use case (2011 to 2022). Each component in the system for the fluoride misinformation use case performed as expected. The data extraction framework handles large amounts of data within short periods. The LDA topic models achieved relatively high coherence values (0.54), and the predicted topics were accurate and befitting to the data. The sentiment analyzer performed at a correlation coefficient of 0.72 but could be improved in further iterations. The misinformation classifier attained a satisfactory correlation coefficient of 0.82 against expert-validated data. Moreover, the output dashboard and analytics hosted on the Elastic Cloud deployment are intuitive for researchers without a technical background and comprehensive in their visualization and analytics capabilities. In fact, the investigators of the fluoride misinformation use case have successfully used the system to extract interesting and important insights into public health, which have been published separately. CONCLUSIONS: The novel U-MAS pipeline has the potential to detect and analyze misleading information related to a particular topic or set of related topics.

Identifying the Socioeconomic, Demographic, and Political Determinants of Social Mobility and Their Effects on COVID-19 Cases and Deaths: Evidence From US Counties.

Background: The spread of COVID-19 at the local level is significantly impacted by population mobility. The U.S. has had extremely high per capita COVID-19 case and death rates. Efficient nonpharmaceutical interventions to control the spread of COVID-19 depend on our understanding of the determinants of public mobility. Objective: This study used publicly available Google data and machine learning to investigate population mobility across a sample of US counties. Statistical analysis was used to examine the socioeconomic, demographic, and political determinants of mobility and the corresponding patterns of per capita COVID-19 case and death rates. Methods: Daily Google population mobility data for 1085 US counties from March 1 to December 31, 2020, were clustered based on differences in mobility patterns using K-means clustering methods. Social mobility indicators (retail, grocery and pharmacy, workplace, and residence) were compared across clusters. Statistical differences in socioeconomic, demographic, and political variables between clusters were explored to identify determinants of mobility. Clusters were matched with daily per capita COVID-19 cases and deaths. Results: Our results grouped US counties into 4 Google mobility clusters. Clusters with more population mobility had a higher percentage of the population aged 65 years and over, a greater population share of Whites with less than high school and college education, a larger percentage of the population with less than a college education, a lower percentage of the population using public transit to work, and a smaller share of voters who voted for Clinton during the 2016 presidential election. Furthermore, clusters with greater population mobility experienced a sharp increase in per capita COVID-19 case and death rates from November to December 2020. Conclusions: Republican-leaning counties that are characterized by certain demographic characteristics had higher increases in social mobility and ultimately experienced a more significant incidence of COVID-19 during the latter part of 2020.

The Role UX Design Attributes Play in the Perceived Persuasiveness of Contact Tracing Apps

Contact tracing apps (CTAs) were deployed worldwide in 2020 to combat COVID-19. Due to their low uptake, a growing amount of empirical research is being conducted to understand the factors that drive their adoption. For CTAs to be adopted, users must, first and foremost, find them persuasive. However, there is little research to understand the role user experience (UX) plays in their perceived persuasiveness. Consequently, we conducted an online study on Amazon Mechanical Turk among Canadian and American residents (n = 446) to investigate the most important UX design attributes associated with the perceived persuasiveness of CTAs. The study was based on two app designs (control and persuasive), each of which comprises three use cases: no exposure, exposure, and diagnosis report interfaces. One interface (screenshot) was randomly presented to a participant to view and provide their responses on the perceived UX design attributes and perceived persuasiveness of the interface. In the overall model, we found that perceived usefulness is the most important and consistent UX design attribute that influences perceived persuasiveness (β= 0.29, p < 0.001), followed by perceived trustworthiness (β= 0.24, p < 0.001) and perceived privacy protection (β= 0.16, p < 0.05). Respectively, the three predictors were consistently significant in two-thirds, half, and one-third of the 12 submodels based on app design, adoption status, and country of residence. The relationships regarding the persuasive designs are more likely to be significant, with the variance of the target construct explained by the predictors ranging from 71% to 89% compared with 54% to 69% for the control designs. The three significant attributes will help designers know which UX design attributes to focus on when designing CTAs for future epidemics. More importantly, in predictive modeling, if their ratings are known, they hold potential in predicting new users' responsiveness to multiple persuasive strategies/messages featured in behavior-change support systems.

The Effect of Persuasive Design on the Adoption of Exposure Notification Apps: Quantitative Study Based on COVID Alert.

BACKGROUND: The adoption of contact tracing apps worldwide has been low. Although considerable research has been conducted on technology acceptance, little has been done to show the benefit of incorporating persuasive principles. OBJECTIVE: This research aimed to investigate the effect of persuasive features in the COVID Alert app, created by Health Canada, by focusing on the no-exposure status, exposure status, and diagnosis report interfaces. METHODS: We conducted a study among 181 Canadian residents, including 65 adopters and 116 nonadopters. This study was based on screenshots of the 3 interfaces, of which each comprised a persuasive design and a control design. The persuasive versions of the first two interfaces supported self-monitoring (of exposure levels), and that of the third interface supported social learning (about how many other users have reported their diagnosis). The 6 screenshots were randomly assigned to 6 groups of participants to provide feedback on perceived persuasiveness and adoption willingness. RESULTS: A multivariate repeated-measure ANOVA showed that there is an interaction among interface, app design, and adoption status regarding the perceived persuasiveness of the interfaces. This resulted in a 2-way ANOVA for each interface. For the no-exposure interface, there was an interaction between adoption status and app design. Among adopters, there was no significant difference P=.31 between the persuasive design (mean 5.36, SD 1.63) and the control design (mean 5.87, SD 1.20). However, among nonadopters, there was an effect of app design (P<.001), with participants being more motivated by the persuasive design (mean 5.37, SD 1.30) than by the control design (mean 4.57, SD 1.19). For the exposure interface, adoption status had a main effect (P<.001), with adopters (mean 5.91, SD 1.01) being more motivated by the designs than nonadopters (mean 4.96, SD 1.43). For the diagnosis report interface, there was an interaction between adoption status and app design. Among nonadopters, there was no significant difference P=.99 between the persuasive design (mean 4.61, SD 1.84) and the control design (mean 4.77, SD 1.21). However, among adopters, there was an effect of app design (P=.006), with participants being more likely to report their diagnosis using the persuasive design (mean 6.00, SD 0.97) than using the control design (mean 5.03, SD 1.22). Finally, with regard to willingness to download the app, pairwise comparisons showed that nonadopters were more likely to adopt the app after viewing the persuasive version of the no-exposure interface (13/21, 62% said yes) and the diagnosis report interface (12/17, 71% said yes) than after viewing the control versions (3/17, 18% and 7/16, 44%, respectively, said yes). CONCLUSIONS: Exposure notification apps are more likely to be effective if equipped with persuasive features. Incorporating self-monitoring into the no-exposure status interface and social learning into the diagnosis report interface can increase adoption by >30%.

Factors Influencing the Adoption of Contact Tracing Applications: Systematic Review and Recommendations.

Background: The emergence of new variants of COVID-19 causing breakthrough infections and the endemic potential of the coronavirus are an indication that digital contact tracing apps (CTAs) may continue to be useful for the long haul. However, the uptake of these apps in many countries around the world has been low due to several factors militating against their adoption and usage. Objective: In this systematic review, we set out to uncover the key factors that facilitate or militate against the adoption of CTAs, which researchers, designers and other stakeholders should focus on in future iterations to increase their adoption and effectiveness in curbing the spread of COVID-19. Data Sources: Seven databases, including PubMed, CINAHL, Scopus, Web of Service, IEEE Xplore, ACM Digital Library, and Google Scholar, were searched between October 30 and January 31, 2020. A total of 777 articles were retrieved from the databases, with 13 of them included in the systematic review after screening. Study Eligibility Criteria Participants and Intervention: The criteria for including articles in the systematic review were that they could be user studies from any country around the world, must be peer-reviewed, written in English, and focused on the perception and adoption of COVID-19 contact tracing and/or exposure notification apps. Other criteria included user study design could be quantitative, qualitative, or mixed, and must have been conducted during the COVID-19 pandemic, which began in the early part of 2020. Study Appraisal and Synthesis Methods: Three researchers searched seven databases (three by the first author, and two each by the second and third authors) and stored the retrieved articles in a collaborative Mendeley reference management system online. After the removal of duplicates, each researcher independently screened one third of the articles based on title/abstract. Thereafter, all three researchers collectively screened articles that were in the borderline prior to undergoing a full-text review. Then, each of the three researchers conducted a full-text review of one-third of the eligible articles to decide the final articles to be included in the systematic review. Next, all three researchers went through the full text of each borderline article to determine their appropriateness and relevance. Finally, each researcher extracted the required data from one-third of the included articles into a collaborative Google spreadsheet and the first author utilized the data to write the review. Results: This review identified 13 relevant articles, which found 56 factors that may positively or negatively impact the adoption of CTAs. The identified factors were thematically grouped into ten categories: privacy and trust, app utility, facilitating conditions, social-cognitive factors, ethical concerns, perceived technology threats, perceived health threats, technology familiarity, persuasive design, and socio-demographic factors. Of the 56 factors, privacy concern turned out to be the most frequent factor of CTA adoption (12/13), followed by perceived benefit (7/13), perceived trust (6/13), and perceived data security risk (6/13). In the structural equation models presented by the authors of the included articles, a subset of the 56 elicited factors (e.g., perceived benefit and privacy concern) explains 16 to 77% of the variance of users' intention to download, install, or use CTAs to curb the spread of COVID-19. Potential adoption rates of CTA range from 19% (in Australia) to 75% (in France, Italy, Germany, United Kingdom, and United States). Moreover, actual adoption rates range from 37% (in Australia) to 50% (in Germany). Finally, most of the studies were carried out in Europe (66.7%), followed by North America (13.3%), and Australia, Asia, and South America (6.7% each). Conclusion: The results suggest that future CTA iterations should give priority to privacy protection through minimal data collection and transparency, improving contact tracing benefits (personal and social), and fostering trust through laudable gestures such as delegating contact tracing to public health authorities, making source code publicly available and stating who will access user data, when, how, and what it will be used for. Moreover, the results suggest that data security and tailored persuasive design, involving reward, self-monitoring, and social-location monitoring features, have the potential of improving CTA adoption. Hence, in addition to addressing issues relating to utility, privacy, trust, and data security, we recommend the integration of persuasive features into future designs of CTAs to improve their motivational appeal, adoption, and the user experience. Systematic Review Registration: https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42021259080 PROSPERO, identifier CRD42021259080.

COVID Alert: Factors Influencing the Adoption of Exposure Notification Apps Among Canadian Residents.

The continued emergence of new variants of COVID-19 such as the Delta and Omicron variants, which can cause breakthrough infections, indicates that contact tracing and exposure notification apps (ENAs) will continue to be useful for the long haul. However, there is limited work to uncover the strongest factors that influence their adoption. Using Canada's "COVID Alert" as a case study, we conducted an empirical, technology-acceptance study to investigate the key factors that account for users' intention to use ENAs and the moderating effect of important human and design factors. Our path model analysis shows that four factors significantly influence the adoption of COVID Alert among Canadian residents: perceived risk, perceived usefulness, perceived trust, and perceived compatibility. The overall model explains over 60% of intention to use, with type of design, use case (functional interface), and adoption status moderating the strength of the relationships between the four factors and intention to use. We discuss these findings and make recommendations for the design of future ENAs.

Identifying the Socioeconomic, Demographic, and Political Determinants of Social Mobility and Their Effects on COVID-19 Cases and Deaths: Evidence From US Counties

Background The spread of COVID-19 at the local level is significantly impacted by population mobility. The U.S. has had extremely high per capita COVID-19 case and death rates. Efficient nonpharmaceutical interventions to control the spread of COVID-19 depend on our understanding of the determinants of public mobility. Objective This study used publicly available Google data and machine learning to investigate population mobility across a sample of US counties. Statistical analysis was used to examine the socioeconomic, demographic, and political determinants of mobility and the corresponding patterns of per capita COVID-19 case and death rates. Methods Daily Google population mobility data for 1085 US counties from March 1 to December 31, 2020, were clustered based on differences in mobility patterns using K-means clustering methods. Social mobility indicators (retail, grocery and pharmacy, workplace, and residence) were compared across clusters. Statistical differences in socioeconomic, demographic, and political variables between clusters were explored to identify determinants of mobility. Clusters were matched with daily per capita COVID-19 cases and deaths. Results Our results grouped US counties into 4 Google mobility clusters. Clusters with more population mobility had a higher percentage of the population aged 65 years and over, a greater population share of Whites with less than high school and college education, a larger percentage of the population with less than a college education, a lower percentage of the population using public transit to work, and a smaller share of voters who voted for Clinton during the 2016 presidential election. Furthermore, clusters with greater population mobility experienced a sharp increase in per capita COVID-19 case and death rates from November to December 2020. Conclusions Republican-leaning counties that are characterized by certain demographic characteristics had higher increases in social mobility and ultimately experienced a more significant incidence of COVID-19 during the latter part of 2020.

Designing Better Exposure Notification Apps: The Role of Persuasive Design.

BACKGROUND: Digital contact tracing apps have been deployed worldwide to limit the spread of COVID-19 during this pandemic and to facilitate the lifting of public health restrictions. However, due to privacy-, trust-, and design-related issues, the apps are yet to be widely adopted. This calls for an intervention to enable a critical mass of users to adopt them. OBJECTIVE: The aim of this paper is to provide guidelines to design contact tracing apps as persuasive technologies to make them more appealing and effective. METHODS: We identified the limitations of the current contact tracing apps on the market using the Government of Canada's official exposure notification app (COVID Alert) as a case study. Particularly, we identified three interfaces in the COVID Alert app where the design can be improved. The interfaces include the no exposure status interface, exposure interface, and diagnosis report interface. We propose persuasive technology design guidelines to make them more motivational and effective in eliciting the desired behavior change. RESULTS: Apart from trust and privacy concerns, we identified the minimalist and nonmotivational design of exposure notification apps as the key design-related factors that contribute to the current low uptake. We proposed persuasive strategies such as self-monitoring of daily contacts and exposure time to make the no exposure and exposure interfaces visually appealing and motivational. Moreover, we proposed social learning, praise, and reward to increase the diagnosis report interface's effectiveness. CONCLUSIONS: We demonstrated that exposure notification apps can be designed as persuasive technologies by incorporating key persuasive features, which have the potential to improve uptake, use, COVID-19 diagnosis reporting, and compliance with social distancing guidelines.

Factors Influencing the Adoption of Contact Tracing Applications: Protocol for a Systematic Review.

BACKGROUND: Following the onset of the COVID-19 pandemic, digital contact tracing apps have become prevalent worldwide in a coordinated effort to curb the spread of COVID-19. However, their uptake has been low and slow due to privacy concerns, the lack of trust and motivational affordances, and their minimalist design. OBJECTIVE: The objective of this article is to present a protocol for a systematic review of the main factors, including facilitators and barriers, that influence the adoption of contact tracing apps. METHODS: We searched seven databases, namely, Scopus, CINAHL, PubMed (MEDLINE), IEEE Xplore Digital Library, Association for Computing Machinery (ACM) Digital Library, Web of Science, and Google Scholar, for relevant publications between October 30, 2020, and January 31, 2021. Three authors were involved in removing duplicates, screening, and selection of relevant articles according to the PRISMA-P (Preferred Reporting Items for Systematic Review and Meta-analysis Protocols) guidelines. RESULTS: Altogether, we retrieved 777 articles from the seven databases. As of May 14, 2021, we have completed the screening process and arrived at 13 eligible articles to be included in the systematic review. We hope to elicit, summarize, and report the main findings in the systematic review article by the end of August 2021. We expect to uncover facilitators and barriers related to app utility, data security, ease of use, and persuasive design that are deemed important to adoption of contact tracing apps. CONCLUSIONS: The findings of the systematic review will help researchers to uncover the gaps in the adoption of contact tracing apps, and decision makers and designers to focus on the principal adoption factors necessary to create better and more effective contact tracing apps. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/28961.

Considerations for an Individual-Level Population Notification System for Pandemic Response: A Review and Prototype.

The outbreak of the coronavirus disease (COVID-19), caused by severe acute respiratory syndrome coronavirus 2, spread worldwide after its emergence in China. Whether rich or poor, all nations are struggling to cope with this new global health crisis. The speed of the threat's emergence and the quick response required from public health authorities and the public itself makes evident the need for a major reform in pandemic surveillance and notification systems. The development and implementation of a graded, individual-level pandemic notification system could be an effective tool to combat future threats of epidemics. This paper describes a prototype model of such a notification system and its potential advantages and challenges for implementation. Similar to other emergency alerts, this system would include a number of threat levels (level 1-5) with a higher level indicating increasing severity and intensity of safety measures (eg, level 1: general hygiene, level 2: enhanced hygiene, level 3: physical distancing, level 4: shelter in place, and level 5: lockdown). The notifications would be transmitted to cellular devices via text message (for lower threat levels) or push notification (for higher threat levels). The notification system would allow the public to be informed about the threat level in real time and act accordingly in an organized manner. New Zealand and the United Kingdom have recently launched similar alert systems designed to coordinate the ongoing COVID-19 pandemic response more efficiently. Implementing such a system, however, faces multiple challenges. Extensive preparation and coordination among all levels of government and relevant sectors are required. Additionally, such systems may be effective primarily in countries where there exists at least moderate trust in government. Advance and ongoing public education about the nature of the system and its steps would be an essential part of the system, such that all members of the public understand the meaning of each step in advance, similar to what has been established in systems for other emergency responses. This educational component is of utmost importance to minimize adverse public reaction and unintended consequences. The use of mass media and local communities could be considered where mobile phone penetration is low. The implementation of such a notification system would be more challenging in developing countries for several reasons, including inadequate technology, limited use of data plans, high population density, poverty, mistrust in government, and tendency to ignore or failure to understand the warning messages. Despite the challenges, an individual-level pandemic notification system could provide added benefits by giving an additional route for notification that would be complementary to existing platforms.