Assessment of Sociodemographic Factors Associated with Time to Self-reported COVID-19 Infection Among a Large Multi-Center Prospective Cohort Population in the Southeastern United States (preprint)

Objective: We aimed to investigate sociodemographic factors associated with self-reported COVID-19 infection. Methods: The study population is a multicenter prospective cohort of adult volunteers recruited from healthcare systems located in the mid-Atlantic and southern United States. Between April 2020 and October 2021 participants completed daily online questionnaires about symptoms, exposures, and risk behaviors related to COVID-19, including self-reports of positive SARS CoV-2 detection tests and COVID-19 vaccination. Analysis of time from study enrollment to self-reported COVID-19 infection used a time-varying mixed effects Cox-proportional hazards framework. Results: Overall, 1,603 of 27,214 study participants (5.9%) reported a positive COVID-19 test during the study period. The adjusted hazard ratio demonstrated lower risk for women, those with a graduate level degree, and smokers. A higher risk was observed for healthcare workers, those aged 18-34, those in rural areas, those from households where a member attends school or interacts with the public, and those who visited a health provider in the last year. Conclusions: Increased risk of self-reported COVID-19 was associated with specific demographic characteristics, which may help to inform targeted interventions for future pandemics.

Association between Self-reported Masking Behavior and SARS-CoV-2 Infection Wanes from Pre-Delta to Omicron-Predominant Periods - North Carolina COVID-19 Community Research Partnership (preprint)

We assessed the association between self-reported mask use during non-household interactions and COVID-19 infection during three pandemic periods. Odds of infection for those who did not always compared to those who always wore a mask was 66% higher during pre-Delta, 53% higher during Delta, declining to 16% higher during Omicron.

COVID-19 Utilization and Resource Visualization Engine (CURVE) to Forecast In-Hospital Resources (preprint)

Background: The emergence of COVID-19 has created an urgent threat to public health worldwide. With rapidly evolving demands on healthcare resources, it is imperative that healthcare systems have the ability to access real-time local data to predict, plan, and effectively manage resources. Objective: To develop an interactive COVID-19 Utilization and Resource Visualization Engine (CURVE) as a data visualization tool to inform decision making and guide a large health system's proactive pandemic response. Methods: We designed and implemented CURVE using R Shiny to display real-time parameters of healthcare utilization at Atrium Health with projections based upon locally derived models for the COVID-19 pandemic. We used the CURVE app to compare predictions from two of our models: one created before and one after the statewide stay-at-home and social distancing orders (denoted before- and after-SAH-order model). We established parameter settings for best-, moderate-, and worst-case scenarios for pandemic spread and resource use, leveraging two locally developed forecasting models to determine peak date trajectory, resource use, and root mean square error (RMSE) between observed and predicted results. Results: CURVE predicts and monitors utilization of hospital beds, ICU beds, and number of ventilators in the context of up-to-date local resources and provides Atrium Health leadership with timely, actionable insights to guide decision-making during the COVID-19 pandemic. The after-SAH-order model demonstrated the lowest RMSE in total bed, ICU bed, and patients on ventilators. Conclusions: CURVE provides a powerful, interactive interface that provides locally relevant, dynamic, timely information to guide health system decision making and pandemic preparedness.

Modeling COVID-19 latent prevalence to assess a public health intervention at a state and regional scale (preprint)

BackgroundEmergence of COVID-19 caught the world off-guard and unprepared, initiating a global pandemic. In the absence of evidence, individual communities had to take timely action to reduce the rate of disease spread and avoid overburdening their healthcare systems. Although a few predictive models have been published to guide these decisions, most have not taken into account spatial differences and have included assumptions that do not match the local realities. Access to reliable information that is adapted to local context is critical for policymakers to make informed decisions during a rapidly evolving pandemic. ObjectiveThe goal of this study was to develop an adapted susceptible-infected-removed (SIR) model to predict the trajectory of the COVID-19 pandemic in North Carolina (NC) and the Charlotte metropolitan region and to incorporate the effect of a public health intervention to reduce disease spread, while accounting for unique regional features and imperfect detection. MethodsUsing the software package R, three SIR models were fit to infection prevalence data from the state and the greater Charlotte region and then rigorously compared. One of these models (SIR-Int) accounted for a stay-at-home intervention and imperfect detection of COVID-19 cases. We computed longitudinal total estimates of the susceptible, infected, and removed compartments of both populations, along with other pandemic characteristics (e.g., basic reproduction number). ResultsPrior to March 26, disease spread was rapid at the pandemic onset with the Charlotte region doubling time of 2.56 days (95% CI: (2.11, 3.25)) and in NC 2.94 days (95% CI: (2.33, 4.00)). Subsequently, disease spread significantly slowed with doubling times increased in the Charlotte region to 4.70 days (95% CI: (3.77, 6.22)) and in NC to 4.01 days (95% CI: (3.43, 4.83)). Reflecting spatial differences, this deceleration favored the greater Charlotte region compared to NC as a whole. A comparison of the efficacy of intervention, defined as 1 - the hazard ratio of infection, gave 0.25 for NC and 0.43 for the Charlotte region. Also, early in the pandemic, the initial basic SIR model had good fit to the data; however, as the pandemic and local conditions evolved, the SIR-Int model emerged as the model with better fit. ConclusionsUsing local data and continuous attention to model adaptation, our findings have enabled policymakers, public health officials and health systems to proactively plan capacity and evaluate the impact of a public health intervention. Our SIR-Int model for estimated latent prevalence was reasonably flexible, highly accurate, and demonstrated the efficacy of a stay-at-home order at both the state and regional level. Our results highlight the importance of incorporating local context into pandemic forecast modeling, as well as the need to remain vigilant and informed by the data as we enter into a critical period of the outbreak.