Disparities in air quality downscaler model uncertainty across socioeconomic and demographic indicators in North Carolina.

Studies increasingly use output from the Environmental Protection Agency's Fused Air Quality Surface Downscaler ("downscaler") model, which provides spatial predictions of daily concentrations of fine particulate matter (PM2.5) and ozone (O3) at the census tract level, to study the health and societal impacts of exposure to air pollution. Downscaler outputs have been used to show that lower income and higher minority neighborhoods are exposed to higher levels of PM2.5 and lower levels of O3. However, the uncertainty of the downscaler estimates remains poorly characterized, and it is not known if all subpopulations are benefiting equally from reliable predictions. We examined how the percent errors (PEs) of daily concentrations of PM2.5 and O3 between 2002 and 2016 at the 2010 census tract centroids across North Carolina were associated with measures of racial and educational isolation, neighborhood disadvantage, and urbanicity. Results suggest that there were socioeconomic and demographic disparities in surface concentrations of PM2.5 and O3, as well as their prediction uncertainties. Neighborhoods characterized by less reliable downscaler predictions (i.e., higher PEPM2.5 and PEO3) exhibited greater levels of aerial deprivation as well as educational isolation, and were often non-urban areas (i.e., suburban, or rural). Between 2002 and 2016, predicted PM2.5 and O3 levels decreased and O3 predictions became more reliable. However, the predictive uncertainty for PM2.5 has increased since 2010. Substantial spatial variability was observed in the temporal changes in the predictive uncertainties; educational isolation and neighborhood deprivation levels were associated with smaller increases in predictive uncertainty of PM2.5. In contrast, racial isolation was associated with a greater decline in the reliability of PM2.5 predictions between 2002 and 2016; it was associated with a greater improvement in the predictive reliability of O3 within the same time frame.

Data-driven testing program improves detection of COVID-19 cases and reduces community transmission.

COVID-19 remains a global threat in the face of emerging SARS-CoV-2 variants and gaps in vaccine administration and availability. In this study, we analyze a data-driven COVID-19 testing program implemented at a mid-sized university, which utilized two simple, diverse, and easily interpretable machine learning models to predict which students were at elevated risk and should be tested. The program produced a positivity rate of 0.53% (95% CI 0.34-0.77%) from 20,862 tests, with 1.49% (95% CI 1.15-1.89%) of students testing positive within five days of the initial test-a significant increase from the general surveillance baseline, which produced a positivity rate of 0.37% (95% CI 0.28-0.47%) with 0.67% (95% CI 0.55-0.81%) testing positive within five days. Close contacts who were predicted by the data-driven models were tested much more quickly on average (0.94 days from reported exposure; 95% CI 0.78-1.11) than those who were manually contact traced (1.92 days; 95% CI 1.81-2.02). We further discuss how other universities, business, and organizations could adopt similar strategies to help quickly identify positive cases and reduce community transmission.