Longitudinal study of an emerging COVID-19 stigma: Media exposure, danger appraisal, and stress

Media coverage of coronavirus disease (COVID-19) has played a critical role throughout the pandemic: sharing news about the novel virus, policies and practices to mitigate it, and the race to create and distribute vaccines. The media coverage, however, has been critiqued as stigmatizing. Although this critique is not new, there is limited understanding of how and why new stigmas emerge from exposure to media coverage. Drawing upon the model of stigma communication (Smith et al., 2019) and the attribution model of stigma (Corrigan et al., 2003), we investigated a novel model of stigma emergence that delineates two kinds of longitudinal processes: (a) a message-effects process, in which exposure to mediated messages about COVID-19 leads to public stigma through danger appraisal and (b) a coping process in which stress and rumination shape later perceptions of public stigma. To test the model, we tracked an emerging COVID-19 stigma with a two-wave survey of a prospective, longitudinal cohort living in one county in a mid-Atlantic state (N = 883). The results supported this model. The longitudinal processes of stigma emergence and implications for COVID-19 stigma are discussed. (PsycInfo Database Record (c) 2023 APA, all rights reserved)

A mixed methods study exploring requests for unproven COVID therapies such as ivermectin and healthcare distrust in the rural South.

The COVID-19 pandemic has led to contentious discourse regarding unproven COVID-19 therapies (UCTs),(e.g. ivermectin). Despite recommendations against it, ivermectin remains, in some areas, highly demanded. The goal of this study is to understand patient and provider perspectives about UCTs (e.g., ivermectin) and how responses to requests for UCTs impact healthcare distrust. This mixed methods observational study was conducted in a rural healthcare system in the Southern United States. Adults (n = 26) with a history of COVID-19 or clinicians (n = 8) from the same system were interviewed using questionnaires assessing healthcare distrust and qualitatively interviewed exploring perceptions about UCTs. Patient themes were: 1) Importance of anecdotal stories for decision-making; 2) Use of haphazard approaches to 'research'; 3) Strong distrust of government and healthcare organizations; 4) Inherent trust in local healthcare; 5) Decision-making as weighing pros/cons; and 6) Feeling a right to try medications. High survey medians indicated high distrust with differences of 8.5 points for those who requested/used ivermectin versus those who did not (p = 0.027). Clinician themes were: 1) Frustration when patients trust social media over clinicians; 2) Acceptance of community beliefs about UCTs; 3) Distrust originating outside of the healthcare system; 4) Feeling torn about prescribing UCTs to build trust; and 5) Variable educational strategies. When clinicians are perceived as aligned with government, this may void patients' trust of clinicians. Clinicians should leverage trust in local healthcare and distance themselves from distrusted information sources. Ethical questions arise regarding appropriateness of acquiescing to patient requests for ivermectin for building trust.

Extending the Theory of Normative Social Behavior: Collective Norms, Opinion Leadership, and Masking During the COVID-19 Pandemic.

Novel, public behaviors, such as masking, should be susceptible to normative influence. This paper advances the theory of normative social behavior by considering a new set of moderators of normative influence - superdiffuser traits - and by clarifying the antecedents and consequences of exposure to collective norms. We use data from a two-wave survey of a cohort living in one U.S. county during the pandemic (N = 913) to assess normative effects on masking. We also used a bipartite network (based on people shopping for food in the same stores) to examine exposure to collective norms. The results show different superdiffuser traits have distinct effects on the relationship between perceived injunctive norms and masking intentions. Exposure to collective norms influences masking, but this influence depends on how people interact with their social environments. Network analysis shows that behavioral homophily is a significant predictor of selective exposure to collective norms earlier (but not later) in the pandemic. Implications for understanding normative influence in a context where opinion leadership matters are discussed.

Mixed-methods Analysis of Adults Likely versus Unlikely to Get a COVID-19 Vaccination.

Objectives: This mixed-methods study compared perspectives of those 'very likely' versus 'very unlikely' to receive a hypothetical COVID-19 vaccine. Methods: We used an explanatory, sequential, mixed- methods design to analyze quantitative data from a rural Pennsylvania sample. Of the 976 participants, 67 selected 'very unlikely' to get the COVID-19 vaccine. Responses to open-ended questions: "What worries you the most about the COVID 19 pandemic?" and "What are your thoughts about a potential COVID 19 vaccine?" were qualitatively compared to answers from the 67 participants who selected 'very likely' to get the COVID-19 vaccine. We used descriptive content analysis to compare themes across the 2 groups. Results: Both groups had thematic commonalities related to their concerns. Themes that were more common among those 'very unlikely' to get vaccinated included concern for politics overriding vaccine safety and rushed vaccine development timeline, whereas themes related to hope and optimism about vaccination were exclusive to the 'very likely' group. Conclusions: Shared beliefs existed across groups with different intents to vaccinate; yet, identification with vaccine spokespersons differed. Messaging campaigns can use these commonalities to address vaccine hesitancy.

Essential and non-essential US workers' health behaviors during the COVID-19 pandemic.

We seek to quantify the relationship between health behaviors and work-related experiences during the COVID-19 pandemic by predicting health behaviors as a function of essential worker status, job loss, change in work hours, and COVID-19 experiences. We use multivariate models and survey data from 913 employed adults in a semi-rural mid-Atlantic US county, and test whether essential worker results vary by gender, parenthood, and/or university employment. Multivariate models indicate that essential workers used tobacco on more days (4.5; p <.01) and were less likely to sleep 8 h (odds ratio [OR] 0.6; p <.01) than non-essential workers. The risk of sleeping less than 8 h is concentrated among essential workers in the service industry (OR 0.5; p <.05) and non-parents (OR 0.5; p <.05). Feminine essential workers exercised on fewer days (-0.8; p <.05) than feminine non-essential workers. Workers with reduced work hours consumed more alcoholic drinks (0.3; p <.05), while workers with increased work hours consumed alcohol (0.3; p <.05) and exercised (0.6; p <.05) on more days. Essential worker status and changes in work hours are correlated with unhealthy behaviors during the COVID-19 pandemic.

Developing and testing an automated qualitative assistant (AQUA) to support qualitative analysis.

Qualitative research remains underused, in part due to the time and cost of annotating qualitative data (coding). Artificial intelligence (AI) has been suggested as a means to reduce those burdens, and has been used in exploratory studies to reduce the burden of coding. However, methods to date use AI analytical techniques that lack transparency, potentially limiting acceptance of results. We developed an automated qualitative assistant (AQUA) using a semiclassical approach, replacing Latent Semantic Indexing/Latent Dirichlet Allocation with a more transparent graph-theoretic topic extraction and clustering method. Applied to a large dataset of free-text survey responses, AQUA generated unsupervised topic categories and circle hierarchical representations of free-text responses, enabling rapid interpretation of data. When tasked with coding a subset of free-text data into user-defined qualitative categories, AQUA demonstrated intercoder reliability in several multicategory combinations with a Cohen's kappa comparable to human coders (0.62-0.72), enabling researchers to automate coding on those categories for the entire dataset. The aim of this manuscript is to describe pertinent components of best practices of AI/machine learning (ML)-assisted qualitative methods, illustrating how primary care researchers may use AQUA to rapidly and accurately code large text datasets. The contribution of this article is providing guidance that should increase AI/ML transparency and reproducibility.

Voices from the Pandemic: A Qualitative Study of Family Experiences and Suggestions regarding the Care of Critically Ill Patients.

Rationale: Intensive care unit (ICU) visitation restrictions during the coronavirus disease (COVID-19) pandemic have drastically reduced family-engaged care. Understanding the impact of physical distancing on family members of ICU patients is needed to inform future policies. Objectives: To understand the experiences of family members of critically ill patients with COVID-19 when physically distanced from their loved ones and to explore ways clinicians may support them. Methods: This qualitative study of an observational cohort study reports data from 74 family members of ICU patients with COVID-19 at 10 United States hospitals in four states, chosen based on geographic and demographic diversity. Adult family members of patients admitted to the ICU with COVID-19 during the early phase of the pandemic (February-June 2020) were invited to participate in a phone interview. Interviews followed a semistructured guide to assess four constructs: illness narrative, stress experiences, communication experiences, and satisfaction with care. Interviews were transcribed verbatim and analyzed using an inductive approach to thematic analysis. Results: Among 74 interviewees, the mean age was 53.0 years, 55% were white, and 76% were female. Physical distancing contributed to substantial stress and harms (nine themes). Participants described profound suffering and psychological illness, unfavorable perceptions of care, and weakened therapeutic relationship between family members and clinicians. Three communication principles emerged as those most valued by family members: contact, consistency, and compassion (the 3Cs). Family members offered suggestions to guide clinicians faced with communicating with physically distanced families. Conclusions: Visitation restrictions impose substantial psychological harms upon family members of critically ill patients. Derived from the voics of family members, our findings warrant strong consideration when implementing visitation restrictions in the ICU and advocate for investment in infrastructure (including staffing and videoconferencing) to support communication. This study offers family-derived recommendations to operationalize the 3Cs to guide and improve communication in times of physical distancing during the COVID-19 pandemic and beyond.

Perceptions of Diseases of Despair by Members of Rural and Urban High-Prevalence Communities: A Qualitative Study.

Importance: Diseases of despair (ie, mortality or morbidity from suicidality, drug abuse, and alcoholism) were first characterized as increasing in rural White working-class populations in midlife with low educational attainment and associated with long-term economic decline. Excess mortality now appears to be associated with working-class citizens across demographic and geographic boundaries, but no known qualitative studies have engaged residents of rural and urban locales with high prevalence of diseases of despair to learn their perspectives. Objective: To explore perceptions about despair-related illness and potential intervention strategies among diverse community members residing in discrete rural and urban hotspots. Design, Setting, and Participants: In this qualitative study, high-prevalence hotspots for diseases of despair were identified from health insurance claims data in Central Pennsylvania. Four focus groups were conducted with 60 community members in organizations and coalitions from 3 census block group hotspot clusters in the health system between September 2019 and January 2020. Focus groups explored awareness and beliefs about causation and potential intervention strategies. Main Outcomes and Measures: A descriptive phenomenological approach was applied to thematic analysis, and a preliminary conceptual model was constructed to describe how various factors may be associated with perpetuating despair and with public health. Results: In total, 60 adult community members participated in 4 focus groups (44 women, 16 men; 40 White non-Hispanic, 17 Black, and 3 Hispanic/Latino members). Three focus groups with 43 members were held in rural areas with high prevalence of diseases of despair, and 1 focus group with 17 members in a high-prevalence urban area. Four themes emerged with respect to awareness and believed causation of despair-related illness, and participants identified common associated factors, including financial distress, lack of critical infrastructure and social services, deteriorating sense of community, and family fragmentation. Intervention strategies focused around 2 themes: (1) building resilience to despair through better community and organizational coordination and peer support at the local level and (2) encouraging broader state investments in social services and infrastructure to mitigate despair-related illness. Conclusions and Relevance: In this qualitative study, rural and urban community members identified common factors associated with diseases of despair, highlighting the association between long-term political and economic decline and public health and a need for both community- and state-level solutions to address despair. Health care systems participating in addressing community health needs may improve processes to screen for despair (eg, social history taking) and codesign primary, secondary, and tertiary interventions aimed at addressing factors associated with distress. Such actions have taken on greater urgency with the COVID-19 pandemic.

Exploring Behavioral Typologies to Inform COVID-19 Health Campaigns: A Person-Centered Approach.

As the United States continues to be ravaged by COVID-19, it becomes increasingly important to implement effective public health campaigns to improve personal behaviors that help control the spread of the virus. To design effective campaigns, research is needed to understand the current mitigation intentions of the general public, diversity in those intentions, and theoretical predictors of them. COVID-19 campaigns will be particularly challenging because mitigation involves myriad, diverse behaviors. This study takes a person-centered approach to investigate data from a survey (N = 976) of Pennsylvania adults. Latent class analysis revealed five classes of mitigation: one marked by complete adherence with health recommendations (34% of the sample), one by complete refusal (9% of the sample), and three by a mixture of adherence and refusal. Statistically significant covariates of class membership included relatively positive injunctive norms, risk due to essential workers in the household, personal knowledge of someone who became infected with COVID-19, and belief that COVID-19 was a leaked biological weapon. Additionally, trait reactance was associated with non-adherence while health mavenism was associated with adherence. These findings may be used to good effect by local healthcare providers and institutions, and also inform broader policy-making decisions regarding public health campaigns to mitigate COVID-19.

Unique Predictors of Intended Uptake of a COVID-19 Vaccine in Adults Living in a Rural College Town in the United States.

PURPOSE: To explore public confidence in a COVID-19 vaccine. DESIGN: Cross-sectional survey. SETTING: A rural college town in central Pennsylvania. SUBJECTS: Adult residents without minor children. MEASURES: The primary outcome was COVID-19 vaccination intention. Secondary measures included vaccination attitudes, norms, efficacy, past behavior, trust in the vaccination process, and sociodemographic variables of education, financial standing, political viewpoint, and religiosity. ANALYSIS: Descriptive statistics were used to describe quantitative data. Multivariate ordinal regression was used to model predictors of vaccine intention. RESULTS: Of 950 respondents, 55% were "very likely" and 20% "somewhat likely" to take a coronavirus vaccine, even though 70% had taken the flu vaccine since September 2019. The strongest predictors of vaccine acceptance were trust in the system evaluating vaccines and perceptions of local COVID-19 vaccination norms. The strongest predictors of negative vaccine intentions were worries about unknown side-effects and positive attitudes toward natural infection. Sociodemographic factors, political views, and religiosity did not predict vaccine intentions. CONCLUSION: Fewer adults intend to take a coronavirus vaccine than currently take the flu vaccine. Traditional sociodemographic factors may not be effective predictors of COVID-19 vaccine uptake. Although based on a small sample, the study adds to our limited understanding of COVID-19-specific vaccine confidence among some rural Americans and suggests that traditional public health vaccination campaigns based on sociodemographic characteristics may not be effective.