Accelerating guideline dissemination in nursing homes during the COVID-19 pandemic: A patient-centered randomized controlled trial

Background Nursing homes were ill-equipped for the pandemic;though facilities are required to have infection control staff, only 3% have taken a basic infection control course. Little is known about the implementation of effective practices outside of the acute care setting. We proposed an intervention utilizing Project ECHO, to connect Penn State University experts with nursing home staff and administrators to explore how infection control guidelines can be implemented effectively. Methods A stratified cluster randomized design was used to assign nursing homes to either AHRQ-funded COVID-19 ECHO or AHRQ-funded COVID-19 ECHO+ Results: 136 nursing homes participated. There were no significant differences in COVID-19 infection rate, hospitalization, deaths, or influenza, between ECHO or ECHO+.Discussion: The ECHO model has significant strengths when compared to traditional training, as it allows for remote learning delivered by a multidisciplinary team of experts and utilizes case discussions that match the context of nursing homes.

Skeptical Health Mavens May Limit COVID-19 Vaccine Diffusion: Using the Innovation Diffusion Cycle to Interpret Results of a Cross-sectional Survey among People Who are Socially Vulnerable.

We sought to identify barriers to COVID-19 vaccine uptake among persons who are socially vulnerable in light of the natural cycle of innovation diffusion. Widespread adoption of a health innovation requires a cadre of opinion leaders to build on successes experienced by early adopters. One type of opinion leader in healthcare are health mavens: members of a community who maintain up-to-date health knowledge and share their knowledge others. We surveyed 139 persons who are socially vulnerable regarding their COVID-19 vaccination intention, and evaluated their responses based on psychological traits captured by two scales: innovativeness and health mavenism. Health mavenism was not strongly correlated with COVID-19 vaccine intention. Health mavens often relied on their own healthcare providers (n = 46) and health agency websites (n = 42) for vaccine information. Those who relied on their faith leaders (n = 4) reported a lower likelihood of getting vaccinated (31.5% vs. 76.0%, p < .05). The observed lack of support by health mavens represents a critical barrier to COVID-19 vaccine uptake; targeting campaigns to health mavens may increase COVID-19 vaccine uptake in socially vulnerable communities.

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.

School Nurse Perspectives on COVID-19.

Pennsylvania responded to the COVID-19 pandemic by closing schools and moving to online instruction in March 2020. We surveyed Pennsylvania school nurses (N = 350) in May 2020 to assess the impact of COVID-19 on nurses' concerns about returning to school and impact on practice. Data were analyzed using χ2 tests and regression analyses. Urban school nurses were more concerned about returning to the school building without a COVID-19 vaccine than rural nurses (OR = 1.58, 95% CI [1.05, 2.38]). Nurses in urban locales were more likely to report being asked for guidance on COVID-19 (OR = 1.69, 95% CI [1.06, 2.68]), modify communication practices (OR = 2.33, 95% CI [1.42, 3.82]), and be "very/extremely concerned" about their safety (OR = 2.16, 95% CI [1.35, 3.44]). Locale and student density are important factors to consider when resuming in-person instruction; however, schools should recognize school nurses for their vital role in health communication to assist in pandemic preparedness and response.

Reaching the Hispanic Community About COVID-19 Through Existing Chronic Disease Prevention Programs.

Publicly available data on racial and ethnic disparities related to coronavirus disease 2019 (COVID-19) are now surfacing, and these data suggest that the novel virus has disproportionately sickened Hispanic communities in the United States. We discuss why Hispanic communities are highly vulnerable to COVID-19 and how adaptations were made to existing infrastructure for Penn State Project ECHO (Extension for Community Healthcare Outcomes) and Better Together REACH (a community-academic coalition using grant funds from Racial and Ethnic Approaches to Community Health) to address these needs. We also describe programming to support COVID-19 efforts for Hispanic communities by using chronic disease prevention programs and opportunities for replication across the country.