Validating model output in the absence of ground truth data: A COVID-19 case study using the Simulator of Infectious Disease Dynamics in North Carolina (SIDD-NC) model.

As the COVID-19 pandemic has progressed, various models have been developed to forecast changes in the outbreak and assess intervention strategies. In this study we validate the Simulator of Infectious Disease Dynamics in North Carolina (SIDD-NC) model against an ensemble of proxy-ground truth infections datasets. We assess the performance of SIDD-NC using Spearman Rank Correlation, RMSE, and percent RMSE at a state and county level. We conduct the analysis for the period of March 2020 through November 2020 as well as in shorter time increments to assess both the recreation of the pandemic curve as well as day-to-day transmission of SARS-CoV-2 within the population. We find that SIDD-NC performs well against the datasets in the ensemble, generating an estimate of infections that is robust both spatially and temporally.

NC-COVID: A Time-Varying Compartmental Model for Estimating SARS-CoV-2 Infection Dynamics in North Carolina, US.

Efforts to track and model SARS-CoV-2 infection dynamics in the population have been complicated by certain aspects of the transmission characteristics, which include a pre-symptomatic infectious phase as well as asymptomatic infectious individuals. Another problem is that many models focus on case count, as there has been (and is) limited data regarding infection status of members of the population, which is the most important aspect for constructing transmission models. This paper describes and explains the parameterization, calibration, and revision of the NC-COVID model, a compartmental model to estimate SARS-CoV-2 infection dynamics for the state of North Carolina, US. The model was developed early in the pandemic to provide rapid, up-to-date state-level estimates of the number of people who were currently infected, were immune from a prior infection, and remained susceptible to infection. As a post modeling exercise, we assessed the veracity of the model by comparing its output to SARS-CoV-2 viral particle concentrations detected in wastewater data and to estimates of people infected using COVID-19 deaths. The NC-COVID model was highly correlated with these independently derived estimates, suggesting that it produced accurate estimates of SARS-CoV-2 infection dynamics in North Carolina.

The impact of heat exposure on reduced gestational age in pregnant women in North Carolina, 2011-2015.

Research on the impact of heat on pregnant women has focused largely on outcomes following extreme temperature events, such as particular heat waves or spells of very cold weather on pregnant women. Consistently, the literature has shown a statistically significant relationship between heat with shortened gestational age with studies concentrated largely in the western states of the USA or other nations. The association between heat and shortened gestational age has not been examined in the Southeastern US where maternal outcomes are some of the most challenging in the nation. Unlike previous studies that focus on the impacts of a single heat wave event, this study seeks to understand the impact of high heat over a 5-year period during the annual warm season (May-September). To achieve this goal, a case-crossover study design is employed to understand the impact of heat on preterm labor across regions in North Carolina (NC). Temperature thresholds for impact and the underlying relationships between preterm labor and heat are investigated using generalized additive models (GAM). Gridded temperature data (PRISM) is used to establish exposure classifications. The results reveal significant impacts to pregnant women exposed to heat with regional variations. The exposure variable with the most stable and significant result was minimum temperature, indicating high overnight temperatures have the most impact on preterm birth. The magnitude of this impact varies across regions from a 1% increase in risk to 6% increase in risk per two-degree increment above established minimum temperature thresholds.

Hospital surge capacity for an influenza pandemic in the triangle region of North Carolina.

This research investigates the geographic aspects of health care delivery in the event of a sudden increase in the need for care. We constructed an integrated disease outbreak and surge capacity model to evaluate the ability of a region's healthcare system to provide care in the event of a pandemic. In a case study, we implement the model to investigate how an influenza pandemic similar to the 1918 Spanish Flu pandemic would affect the population of the Raleigh-Durham-Chapel Hill metropolitan statistical area and the ability of the region's hospital system to respond to such an event. Under varying scenarios for hospital capacity, we found that the population needing care would overwhelm the system's ability to provide care in the case study. Our model is presented as a framework that can be augmented and expanded to suit the needs of the particular event and healthcare system or services required. By integrating concepts and models from epidemiology, geography, and health care services research, we provide a valuable tool for potential use in disaster planning, hospital system evaluation, and pandemic preparedness.