Synthetic Lies: Understanding AI-Generated Misinformation and Evaluating Algorithmic and Human Solutions

Large language models have abilities in creating high-volume human-like texts and can be used to generate persuasive misinformation. However, the risks remain under-explored. To address the gap, this work first examined characteristics of AI-generated misinformation (AI-misinfo) compared with human creations, and then evaluated the applicability of existing solutions. We compiled human-created COVID-19 misinformation and ed it into narrative prompts for a language model to output AI-misinfo. We found significant linguistic differences within human-AI pairs, and patterns of AI-misinfo in enhancing details, communicating uncertainties, drawing conclusions, and simulating personal tones. While existing models remained capable of classifying AI-misinfo, a significant performance drop compared to human-misinfo was observed. Results suggested that existing information assessment guidelines had questionable applicability, as AI-misinfo tended to meet criteria in evidence credibility, source transparency, and limitation acknowledgment. We discuss implications for practitioners, researchers, and journalists, as AI can create new challenges to the societal problem of misinformation. © 2023 Owner/Author.

Effective Messaging Strategies for Colorectal Cancer Screening: The Development of the 2022 NCCRT Messaging Guidebook for Black and African American People

Introduction: Colorectal cancer (CRC) is the second leading cause of cancer-related deaths in the United States (US) and disproportionately impacts Black individuals. The US Preventive Services Taskforce began recommending CRC screening for individuals aged 45-49 in 2021, however effective strategies to increase screening participation in Black individuals in this age group are unknown. Thus, the National Colorectal RoundTable (NCCRT) used a mixed methods approach to identify barriers and facilitators to CRC screening in Black individuals, with specific focus on those age 45-49. Results informed the development of the 2022 NCCRT Messaging Guidebook for Black & African American People. Method(s): We conducted a mixed-methods study in a large, nationally representative sample of unscreened Black individuals. We first conducted semi-structured qualitative interviews with Black individuals over age 45, recruited from the Schlesinger Group qualitative research platform. Findings informed content for a subsequent survey to understand barriers and facilitators, administered broadly via the Prodege online research platform. Messages to encourage screening participation were developed based on learnings from prior ACS and NCCRT work. Message were tested using MaxDiff analytic methods and reviewed by a multidisciplinary advisory committee for inclusion in the Guidebook. Result(s): There were 10 qualitative interview and 490 survey participants. The average age of participants was 52.7 (s.d.56.1) for interviews and 55.3 (s.d.57.3) for surveys. 40.0% were female and 38.2% lived in the Southeast US (Table). The most frequently reported barrier to screening was procrastination (40.0% in age 45-49;42.8% in age 50-65;34.2% in age .55). Procrastination was often attributed to financial concerns (20.8% in age 45-49) and COVID-19 (27.0% in age 50-54;21.8% in age .55) (Figure). Of those age 45-49, the majority preferred to receive screening information from a health care provider (57.5%), however only 31.7% reported that a provider had initiated a screening conversation. Several messages rated as highly effective in encouraging screening were included in the NCCRT Guidebook. Conclusion(s): We identified several age-specific barriers to CRC screening and developed unique messaging to motivate screening among unscreened Black individuals age 45 and over. Messages that tested positively are publicly available as a resource for organizations and institutions that aim to increase screening rates.

Shifting Trust: Examining How Trust and Distrust Emerge, Transform, and Collapse in COVID-19 Information Seeking

During crises like COVID-19, individuals are inundated with conflicting and time-sensitive information that drives a need for rapid assessment of the trustworthiness and reliability of information sources and platforms. This parallels evolutions in information infrastructures, ranging from social media to government data platforms. Distinct from current literature, which presumes a static relationship between the presence or absence of trust and people's behaviors, our mixed-methods research focuses on situated trust, or trust that is shaped by people's information-seeking and assessment practices through emerging information platforms (e.g., social media, crowdsourced systems, COVID data platforms). Our findings characterize the shifts in trustee (what/who people trust) from information on social media to the social media platform(s), how distrust manifests skepticism in issues of data discrepancy, the insufficient presentation of uncertainty, and how this trust and distrust shift over time. We highlight the deep challenges in existing information infrastructures that influence trust and distrust formation.

Social Media Use and COVID-19 Vaccination Intent: An Exploratory Study on the Mediating Role of Information Exposure

We stumble upon new and repeating information daily. As information comes from many sources, social media continues to play a predominant role in disseminating information, ultimately impacting individuals' perceptions and behaviors. A prime example of this impact was observed during the COVID-19 pandemic, in which social media use was influencing willingness to receive the COVID-19 vaccine. While studies on this relationship between social media use and vaccination intent have been widely investigated, less is known about the mechanisms that link these two variables, specifically the types of information seen on social media platforms and the effects of these different types of information. In this exploratory study, we demonstrate the mediator role of information exposure (to include both types of information and frequency) between social media use and vaccination intent. Our results show that different types of information mediate this relationship differently and demonstrate how these relationships were further moderated by the income level of the participant. We conclude with the implications of these findings and how our findings can inform the direction of future research within the field of human-computer interaction.

Investigating Older Adults' Atitudes towards Crisis Informatics Tools: Opportunities for Enhancing Community Resilience during Disasters

The world population is projected to rapidly age over the next 30 years. Given the increasing digital technology adoption amongst older adults, researchers have investigated how technology can support aging populations. However, little work has examined how technology can support older adults during crises, despite increasingly common natural disasters, public health emergencies, and other crisis scenarios in which older adults are especially vulnerable. Addressing this gap, we conducted focus groups with older adults residing in coastal locations to examine to what extent they felt technology could support them during emergencies. Our fndings characterize participants' desire for tools that enhance community resilience-local knowledge, preparedness, community relationships, and communication, that help communities withstand disasters. Further, older adults' crisis technology preferences were linked to their sense of control, social relationships, and digital readiness. We discuss how a focus on community resilience can yield crisis technologies that more efectively support older adults.

Perceptions on the use of distance learning by families of children with visual impairment and deafblindness

The purpose of our qualitative study was to explore what distance-based teaching and learning practices have been supportive to students with visual impairments and their families. Using purposive sampling, interviews, and qualitative analysis, we found that supportive approaches for distance learning (DL) included parental involvement and participation, as well as tailored instructional approaches and accommodations for the student. In some instances, DL was identified as being more supportive for immune-compromised children. Negative facets of the practice included diminished richness in socializing, and the lack of certain strengths of in-person education. Families' experiences ranged from finding DL helpful, to considering the practice as unfit for their child's education, as well as a poor fit for family life. Flags for future research include family preparation for future DL needs, including culturally-diverse families in research opportunities, and evaluating what DL supports lead to improved outcomes for children and families. © The Author(s) 2023.

Investigating sialic acid usage in species D human adenoviruses

Human adenoviruses are phylogenetically divided across seven species, A-G, causing transient mild illnesses, except in immunocompromised individuals. Their double stranded DNA genome is amenable to genetic manipulations, enabling development of highly engineered virotherapies. Species D adenoviruses have naturally low seroprevalence rates, an important trait in avoiding neutralising anti-vector immunity. We previously demonstrated that HAdV-D26, the platform of the Janssen SARS-CoV2 vaccine, uses sialic acid as a primary cell entry receptor. Here, we structurally and biologically investigated sialic acid usage across species D. We solved multiple structures of species D adenovirus fiber knob proteins alone and in complex with sialic acid, identifying a conserved binding pocket common with known sialic acid binders HAdV-D26 and 37. Using fiber-knob pseudotyped viruses, we demonstrate significantly reduced transduction in cells treated with neuraminidase to remove sialic acid residues in HAdV-D26 and 53, with HAdV-D15, 24 and 29 also demonstrating non-significant reductions. IC50 data also showed highlighted binding to CAR, although at a significantly lower affinity compared to the CAR-binding HAdV-C5. Improved understanding of the usage of sialic acid as a receptor will enable better exploitation of the species D adenoviruses as therapeutic vectors. Our findings raise the possibility of a conserved sialic acid binding pocket within species D adenoviruses resulting in varying affinity levels. Further evaluation of specific glycan binding patterns used by these viruses, as observed between HAdV-D37 and GD1a glycan, will better inform the design of appropriate antivirals to contain adenovirus outbreaks as well as the engineering of targeted vectors for translational applications.

Virus-cell interactions of the ChAdOx1 viral vector give insights into vaccine-induced immune thrombotic thrombocytopenia

The ChAdOx1 nCoV-19 vaccine (AZD1222/Vaxzervia) adapted from the chimpanzee adenovirus Y25 (ChAd-Y25) has been critical in combatting the severe acute respiratory coronavirus 2 (SARS-CoV-2) pandemic. However, as part of the largest vaccination campaign in history, a potentially lifethreatening clotting disorder, thrombosis with thrombocytopenia, resembling heparin-induced thrombocytopenia (HIT), has been observed in a minority of AZD1222 patients following the first but not the second dose. Vaccine-induced immune thrombotic thrombocytopenia (VITT) is characterised by development of thromboses at uncommon sites such as the cerebral venous sinuses and the splanchnic veins, with concomitant thrombocytopaenia. Therefore, to determine how ChAdOx1 may contribute to this novel disorder, it is critical to investigate the vector-host interactions of ChAdOx1. Structural and in vitro analysis of the fiber knob responsible for the primary virus-cell interaction suggests that coxsackie and adenovirus receptor (CAR) is the primary ChAdOx1 receptor. However, ChAdOx1 infection of CAR(-) human vascular endothelial cells has been demonstrated in vitro, suggesting ChAdOx1 may be using additional receptors. Dual tropism has been demonstrated in other human adenoviruses, with HAdV-D26 and HAdV-D37 both using sialic acid and CAR for transduction. Furthermore, coagulation factor X (FX), a factor demonstrated to bind to the hexon and facilitate human adenovirus type 5 (HAdV-C5) transduction via a CARindependent pathway does not increase ChAdOx1 infection, with amino acid alignment between the hexon proteins suggesting ChAdOx1 is unable to bind FX. Taken together, these findings suggest ChAdOx1 uses additional as yet unknown mechanisms for transduction, which may further contribute to the pathogenesis of VITT.

Circumventing anti-vector immunity towards adenoviral vectored vaccines

Replication deficient (RD) adenoviruses (Ad) are the most widely administered viral vectors, with licensed SARS-CoV-2 vaccines using vectors derived from human Ad type 5 (Ad5) and 26 (Ad26), and chimpanzee Ad ''ChAdOx1''. Ad vectored vaccines generate robust cellular and humoral immunity, against both the transgene-encoded protein and the Ad vector itself. It's unclear how many times a single Ad vector can be readministered before this anti-vector immunity impairs generation of the desired transgene-specific adaptive responses. Antivector immunity also arises from naturally acquired Ad infections. In the absence of anti-Ad5 immunity, Ad5 is a goldstandard vector with robust vaccine immunogenicity, however widespread Ad5 seroprevalence hampers its use as a vector for the global population. We developed novel pseudotyped Ads as RD vectored vaccines encoding SARS-CoV-2 spike protein. These vectors exhibit fiber knob swaps from low seroprevalence Ads grafted onto an Ad5 backbone. We characterised innate immune responses following administration of these vectors in mice, and spikespecific adaptive responses three weeks later. Furthermore, we quantified the effects of anti-vector humoral immunity against these vectors in an in vitro transduction assay using human plasma. The pseudotyped vectors exhibit many desirable vaccine characteristics as the equivalent Ad5 vector, including CD4+ and CD8+ T cell responses against multiple spike epitopes. Importantly, fiber knob pseudotyping can substantially circumvent the direct, humoral, anti-vector immunity induced through Ad exposure in humans. These data indicate the fiber knob plays an important role in anti-vector immunity, and can be manipulated for evasion of such responses without hampering vaccine immunogenicity.

Implications of novel Adenovirus-Host interactions in Thrombosis with Thrombocytopenia Syndrome

In this study we investigated a link between adenovirus-based vaccines, deployed to fight the SARS-CoV-2 pandemic, and lifethreatening thromboembolisms after vaccination. Post-marketing surveillance showed that, following vaccination, Vaxzevria (ChAdOx1 based, AstraZeneca) and Jcovden (Adenovirus type 26 based, Johnson & Johnson) are associated with reduced platelet counts (thrombocytopenia) and blood clots (thrombosis) in some individuals. This extremely rare condition, with a rate between 1:50,000 - 1:350,000 cases per vaccinated individual, is above background rates of thrombosis in the population and can lead to fatal ischemic events including cerebral venous thrombosis, intracranial haemorrhage, and pulmonary embolism. It has been termed vaccine induced thrombotic thrombocytopenia (VITT) or thrombosis with thrombocytopenia syndrome (TTS). Heparin induced thrombocytopenia (HIT) is another condition with a similar clinical presentation to TTS. In HIT, immunoaggregates are formed due to the presence of strong anti-selfantibodies directed against Platelet Factor 4 (PF4). When similar anti-PF4 antibodies were detected in TTS patients, we investigated whether there could be a link between the adenovirus vectors used in the vaccines and PF4. This study demonstrates a direct interaction between adenovirus capsids and PF4 using surface plasmon resonance. We then utilized an integrative structural biology workflow including cryo-electron microscopy and molecular dynamics to characterize and demonstrate the mechanism of this interaction. These results demonstrate a previously unknown adenovirushost interaction and provide critical clues as to the underlying mechanism which causes TTS, including how these pathogenic anti-PF4 antibodies may be induced. We are therefore able to present a hypothesis as to the route of pathogenesis in TTS.

Bleeding and thrombosis outcomes in hospitalised COVID-19 patients on low-molecular-weight heparin and antiplatelet therapy.

BACKGROUND: An increased incidence of thromboembolic events in hospitalised COVID­19 patients has been demonstrated despite the use of low-molecular-weight heparin (LMWH). Antiplatelet therapy prior to admission and early in the disease course has been hypothesised to be protective against thrombosis. OBJECTIVES: To describe the bleeding and thrombosis outcomes in hospitalised patients with confirmed COVID­19 receiving LMWH, with and without concomitant antiplatelet therapy. Secondary objectives were to explore predictors of bleeding and thrombosis outcomes, and dosing practices of antiplatelet therapy and LMWH. METHODS: We conducted a descriptive, cross-sectional study of bleeding and thrombosis outcomes at Tygerberg Academic Hospital, Cape Town, South Africa, during the first COVID­19 wave, in 808 hospitalised patients with confirmed COVID­19 receiving LMWH with and without concomitant antiplatelet therapy. Multivariate logistic regression analysis was performed if predictors were deemed statistically and clinically significant. RESULTS: Patients receiving both LMWH and antiplatelet therapy had similar bleeding outcomes compared with patients only receiving LMWH (odds ratio (OR) 1.5; 95% confidence interval (CI) 0.6 - 4.0). Patients receiving both LMWH and antiplatelet therapy had increased odds of developing thrombosis compared with patients only receiving LMWH (OR 4.8; 95% CI 2.1 - 10.7). CONCLUSION: The bleeding risk in COVID­19 patients receiving both LMWH and antiplatelet therapy was not significantly increased. A potentially higher risk of thrombosis in patients receiving LMWH and antiplatelet therapy was observed. However, this could reflect confounding by indication. Randomised studies are required to further evaluate the use of antiplatelet therapy to treat hospitalised patients with COVID­19.

Undiagnosed comorbidities among individuals hospitalised with COVID-19 in South African public hospitals.

BACKGROUND: Previous studies have reported comorbid disease, including hypertension, diabetes mellitus, chronic cardiac and renal disease, malignancy, HIV, tuberculosis (TB) and obesity, to be associated with COVID­19 mortality. National demographic surveys have reported a high proportion of undiagnosed and untreated comorbid disease in South Africa (SA). OBJECTIVES: To determine the number of individuals with previously undiagnosed HIV, TB and non-communicable diseases (NCDs) among patients hospitalised with COVID­19, and the level of medical control of these chronic diseases. METHODS: We conducted a sentinel surveillance study to collect enhanced data on HIV, TB and NCDs among individuals with COVID­19 admitted to 16 secondary-level public hospitals in six of the nine provinces of SA. Trained surveillance officers approached all patients who met the surveillance case definition for inclusion in the study, and consenting patients were enrolled. The data collection instrument included questions on past medical history to determine the self-reported presence of comorbidities. The results of clinical and laboratory testing introduced as part of routine clinical care for hospitalised COVID­19 patients were collected for the study, to objectively determine the presence of hypertension, diabetes, HIV and TB and the levels of control of diabetes and HIV. RESULTS: On self-reported history, the most prevalent comorbidities were hypertension (n=1 658; 51.5%), diabetes (n=855; 26.6%) and HIV (n=603; 18.7%). The prevalence of self-reported active TB was 3.1%, and that of previous TB 5.5%. There were 1 254 patients admitted with COVID­19 (39.0%) who met the body mass index criteria for obesity. On clinical and laboratory testing, 87 patients were newly diagnosed with HIV, 29 with TB, 215 with diabetes and 40 with hypertension during their COVID­19 admission. There were 151/521 patients living with HIV (29.0%) with a viral load >1 000 copies/mL and 309/570 (54.2%) with a CD4 count <200 cells/µL. Among 901 patients classified as having diabetes, 777 (86.2%) had a glycated haemoglobin (HbA1c) level ≥6.5%. CONCLUSION: The study revealed a high prevalence of comorbid conditions among individuals with COVID­19 admitted to public hospitals in SA. In addition, a significant number of patients had previously undiagnosed hypertension, diabetes, HIV and active TB, and many and poorly controlled chronic disease, as evidenced by high HbA1c levels in patients with diabetes, and high viral loads and low CD4 levels in patients with HIV. The findings highlight the importance of strengthening health systems and care cascades for chronic disease management, which include prevention, screening for and effectively treating comorbidities, and ensuring secure and innovative supplies of medicines in primary healthcare during the COVID­19 pandemic.

Use of inhaled nitric oxide in ICU - When is a therapy trial enough?

Introduction: Inhaled nitric oxide (iNO) can be utilized as a rescue treatment option in refractory hypoxaemia and in the potential reversal of pulmonary vascular resistance by pulmonary vasodilation.1 Its use remains controversial due to limited evidence regarding efficacy and potential side effects.2 Furthermore, it requires additional equipment and consumables and its infrequent use means staff may be relatively unfamiliar with the treatment. Objectives: Investigate use of iNO within our adult critical care unit in order to identify potential quality improvements that could be made to the delivery of this therapy. Clarify the proportion of patients who demonstrated a favourable PaO2/FiO2 (PF) response to iNO. Methods: We conducted a single-centre retrospective analysis of consecutive patients treated with iNO on the General & Cardiac Intensive Care Unit at Manchester Royal Infirmary between 01/01/2018 and 25/06/2021. Data was extracted from electronic patient records on patient characteristics, indication for iNO, starting dose, ventilatory characteristics, change in PF ratio and ICU outcome. Results were recorded at iNO initiation, 2, 6, 12 and 24 hours. A responder was defined by improvement in the PF ratio of ≥20% at any point up to 6 hours after initiation. Results: 37 patients were identified, mean age 51 years (SD 14). 84% were male and 16% were female. 27 patients (73%) had a diagnosis of COVID-19 pneumonitis (Figure 1). Primary indication for iNO was acute respiratory failure (ARF) in32 (86%) and right ventricular failure in 5 (14%)patients. Prior to iNO, patients had been in ICU for 11 days (SD 11) and had received invasive mechanical ventilation for 163 hours (SD 188). PF ratios at initiation were 12.8kPa (SD 5.0), in keeping with severe ARDS and 13 (39%) were already proned at initiation. Median troponin result prior to therapy was 21[IQR 14-81]. 12 patients (32%) had ECHO evidence of raised pulmonary artery pressures. A starting dose of 20ppm iNO was observed for each patient. By 6 hours, 14/37 patients (37%) were classified as positive responders. There was a significant increase in PF ratios at 6 hours compared to baseline in responders (15.9kPa vs 12.0kPa, p=0.04) but no significant difference at any other time point. There was no difference in the duration of therapy in responders vs non responders (76 vs 85 hours, p=0.72). Conclusion: iNO therapy may offer short-term improvement in oxygenation in <40% of patients at 6 hours. Duration of therapy was similar regardless of response. This may suggest reluctance to discontinue therapy once started possibly through fear of precipitating deterioration, belief that iNO has halted decline or reluctance to recognize futility when there are few other therapeutic options. iNO is typically commenced when patients have already been invasively ventilated for several days. It is unknown whether earlier initiation would affect response. Use of iNO has increased markedly since the start of the COVID-19 pandemic which should prompt units to evaluate their own practice with this therapy.

Communicating Science, Technology, and Environmental Issues: A Case Study of an Intercultural Learning Experience

This science communication case study analyzes an online international co-taught course where students practiced blog article conceptualization and production covering a wide variety of science and technology related issues. Students had an international experience during the COVID-19 pandemic, and gained experience in communicating science and technology to intercultural audiences. Through student article reviews, course evaluations and project reflections students demonstrated an adoption of new science communication skills and some key examples of changing perspective on issues such as environment and technology. They also enjoyed the opportunity to learn about new cultures, reflect on their own, and bond over life experiences. Copyright © 2022 van Kempen, Kristiansen and Feldpausch-Parker.

Comparison of patients with severe COVID-19 admitted to an intensive care unit in South Africa during the first and second wave of the COVID-19 pandemic.

Background: The second wave of coronavirus disease 2019 (COVID-19), dominated by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Beta variant, has been reported to be associated with increased severity in South Africa (SA). Objectives: To describe and compare clinical characteristics, management and outcomes of COVID-19 patients admitted to an intensive care unit (ICU) in SA during the first and second waves. Methods: In a prospective, single-centre, descriptive study, we compared all patients with severe COVID-19 admitted to ICU during the first and second waves. The primary outcomes assessed were ICU mortality and ICU length of stay (LOS). Results: In 490 patients with comparable ages and comorbidities, no difference in mortality was demonstrated during the second compared with the first wave (65.9% v. 62.5%, p=0.57). ICU LOS was longer in the second wave (10 v. 6 days, p<0.001). More female admissions (67.1% v. 44.6%, p<0.001) and a greater proportion of patients were managed with invasive mechanical ventilation than with non-invasive respiratory support (39.0% v. 14%, p<0.001) in the second wave. Conclusion: While clinical characteristics were comparable between the two waves, a higher proportion of patients was invasively ventilated and ICU stay was longer in the second. ICU mortality was unchanged.

Mapping the Landscape of COVID-19 Crisis Visualizations

In response to COVID-19, a vast number of visualizations have been created to communicate information to the public. Information exposure in a public health crisis can impact people's attitudes towards and responses to the crisis and risks, and ultimately the trajectory of a pandemic. As such, there is a need for work that documents, organizes, and investigates what COVID-19 visualizations have been presented to the public. We address this gap through an analysis of 668 COVID-19 visualizations. We present our findings through a conceptual framework derived from our analysis, that examines who, (uses) what data, (to communicate) what messages, in what form, under what circumstances in the context of COVID-19 crisis visualizations. We provide a set of factors to be considered within each component of the framework. We conclude with directions for future crisis visualization research.