ABSTRACT
Colleges and universities in the US struggled to provide safe in-person education throughout the COVID-19 pandemic. Testing coupled with isolation is a nimble intervention strategy that can be tailored to mitigate health and economic costs, as the virus and our arsenal of medical countermeasures continue to evolve. We developed a decision-support tool to aid in the design of university-based testing strategies using a mathematical model of SARS-CoV-2 transmission. Applying this framework to a large public university reopening in the fall of 2021 with a 60% student vaccination rate, we find that the optimal strategy, in terms of health and economic costs, is twice weekly antigen testing of all students. This strategy provides a 95% guarantee that, throughout the fall semester, case counts would not exceed the CDCs original high transmission threshold of 100 cases per 100k persons over 7 days. As the virus and our medical armament continue to evolve, testing will remain a flexible tool for managing risks and keeping campuses open. We have implemented this model as an online tool to facilitate the design of testing strategies that adjust for COVID-19 conditions, university-specific parameters, and institutional goals. Author SummaryAs a part of the COVID-19 response team at a large public university in the US, we performed an analysis that considered together, the potential health and economic costs of different testing policies for the student body. University administrators had to weigh the up-front effort needed to implement wide scale testing against the potential costs of responding to high levels of disease on campus in the Fall of 2021, after vaccines were widely available but vaccination rates among college students were uncertain. The results presented here are applied to this specific instance, but the online tool provided can be tailored to university specific parameters, the epidemiological conditions, and the goals of the university. As we confront newly emerging variants of COVID-19 or novel pathogens, consideration of both the health and economic costs of proactive testing may serve as a politically tractable and cost-effective disease mitigation strategy.
ABSTRACT
COVID-19 has disproportionately impacted individuals depending on where they live and work, and based on their race, ethnicity, and socioeconomic status. Studies have documented catastrophic disparities at critical points throughout the pandemic, but have not yet systematically tracked their severity through time. Using anonymized hospitalization data from March 11, 2020 to June 1, 2021, we estimate the time-varying burden of COVID-19 by age group and ZIP code in Austin, Texas. During this 15-month period, we estimate an overall 16.9% (95% CrI: 16.1-17.8%) infection rate and 34.1% (95% CrI: 32.4-35.8%) case reporting rate. Individuals over 65 were less likely to be infected than younger age groups (8.0% [95% CrI: 7.5-8.6%] vs 18.1% [95% CrI: 17.2-19.2%]), but more likely to be hospitalized (1,381 per 100,000 vs 319 per 100,000) and have their infections reported (51% [95% CrI: 48-55%] vs 33% [95% CrI: 31-35%]). Children under 18, who make up 20.3% of the local population, accounted for only 5.5% (95% CrI: 3.8-7.7%) of all infections between March 1 and May 1, 2020 compared with 20.4% (95% CrI: 17.3-23.9%) between December 1, 2020 and February 1, 2021. We compared ZIP codes ranking in the 75th percentile of vulnerability to those in the 25th percentile, and found that the more vulnerable communities had 2.5 (95% CrI: 2.0-3.0) times the infection rate and only 70% (95% CrI: 61%-82%) the reporting rate compared to the less vulnerable communities. Inequality persisted but declined significantly over the 15-month study period. For example, the ratio in infection rates between the more and less vulnerable communities declined from 12.3 (95% CrI: 8.8-17.1) to 4.0 (95% CrI: 3.0-5.3) to 2.7 (95% CrI: 2.0-3.6), from April to August to December of 2020, respectively. Our results suggest that public health efforts to mitigate COVID-19 disparities were only partially effective and that the CDCs social vulnerability index may serve as a reliable predictor of risk on a local scale when surveillance data are limited.
ABSTRACT
Claims that in-person schooling has not amplified SARS-CoV-2 transmission are based on similar infection rates in schools and their surrounding communities and limited numbers of documented in-school transmission events. Simulations assuming high in-school transmission suggest that these metrics cannot exclude the possibility that transmission in schools exacerbated overall pandemic risks.
ABSTRACT
Recent identification of the highly transmissible novel SARS-CoV-2 variant in the United Kingdom (B.1.1.7) has raised concerns for renewed pandemic surges worldwide 1,2. B.1.1.7 was first identified in the US on December 29, 2020 and may become dominant by March 2021 3. However, the regional prevalence of B.1.1.7 is largely unknown because of limited molecular surveillance for SARS-CoV-2 4. Quantitative PCR data from a surveillance testing program on a large university campus with roughly 30,000 students provides local situational awareness at a pivotal moment in the COVID-19 pandemic.
ABSTRACT
Short-term probabilistic forecasts of the trajectory of the COVID-19 pandemic in the United States have served as a visible and important communication channel between the scientific modeling community and both the general public and decision-makers. Forecasting models provide specific, quantitative, and evaluable predictions that inform short-term decisions such as healthcare staffing needs, school closures, and allocation of medical supplies. Starting in April 2020, the US COVID-19 Forecast Hub (https://covid19forecasthub.org/) collected, disseminated, and synthesized tens of millions of specific predictions from more than 90 different academic, industry, and independent research groups. A multi-model ensemble forecast that combined predictions from dozens of different research groups every week provided the most consistently accurate probabilistic forecasts of incident deaths due to COVID-19 at the state and national level from April 2020 through October 2021. The performance of 27 individual models that submitted complete forecasts of COVID-19 deaths consistently throughout this year showed high variability in forecast skill across time, geospatial units, and forecast horizons. Two-thirds of the models evaluated showed better accuracy than a naive baseline model. Forecast accuracy degraded as models made predictions further into the future, with probabilistic error at a 20-week horizon 3-5 times larger than when predicting at a 1-week horizon. This project underscores the role that collaboration and active coordination between governmental public health agencies, academic modeling teams, and industry partners can play in developing modern modeling capabilities to support local, state, and federal response to outbreaks. Significance StatementThis paper compares the probabilistic accuracy of short-term forecasts of reported deaths due to COVID-19 during the first year and a half of the pandemic in the US. Results show high variation in accuracy between and within stand-alone models, and more consistent accuracy from an ensemble model that combined forecasts from all eligible models. This demonstrates that an ensemble model provided a reliable and comparatively accurate means of forecasting deaths during the COVID-19 pandemic that exceeded the performance of all of the models that contributed to it. This work strengthens the evidence base for synthesizing multiple models to support public health action.
ABSTRACT
BACKGROUNDIn 2020, U.S schools closed due to SARS-CoV-2 but their role in transmission was unknown. In fall 2020, national guidance for reopening omitted testing or screening recommendations. We report the experience of 2 large independent K-12 schools (School-A and School-B) that implemented an array of SARS-CoV-2 mitigation strategies that included periodic universal testing. METHODSSARS-CoV-2 was identified through periodic universal PCR testing, self-reporting of tests conducted outside school, and contact tracing. Schools implemented behavioral and structural mitigation measures, including mandatory masks, classroom disinfecting, and social distancing. RESULTSOver the fall semester, School-A identified 112 cases in 2320 students and staff; School-B identified 25 cases (2.0%) in 1200 students and staff. Most cases were asymptomatic and none required hospitalization. Of 69 traceable introductions, 63(91%) were not associated with school-based transmission, 59 cases (54%) occurred in the 2 weeks post-Thanksgiving. In 6/7 clusters, clear noncompliance with mitigation protocols was found. The largest outbreak had 28 identified cases and was traced to an off-campus party. There was no transmission from students to staff. CONCLUSIONSAlthough school-age children can contract and transmit SARS-CoV-2, rates of COVID-19 infection related to in-person education were significantly lower than those in the surrounding community. However, social activities among students outside of school undermined those measures and should be discouraged, perhaps with behavioral contracts, to ensure the safety of school communities. In addition, introduction risks were highest following extended school breaks. These risks may be mitigated with voluntary quarantines and surveillance testing prior to re-opening.
ABSTRACT
As COVID-19 vaccination begins worldwide, policymakers face critical trade-offs. Using a mathematical model of COVID-19 transmission, we find that timing of the rollout is expected to have a substantially greater impact on mortality than risk-based prioritization and uptake and that prioritizing first doses over second doses may be life saving.
ABSTRACT
The recent publication of the Great Barrington Declaration (GBD), which calls for relaxing all public health interventions on young, healthy individuals, has brought the question of herd immunity to the forefront of COVID-19 policy discussions, and is partially based on unpublished research that suggests low herd immunity thresholds (HITs) of 10-20%. We re-evaluate these findings and correct a flawed assumption leading to COVID-19 HIT estimates of 60-80%. If policymakers were to adopt a herd immunity strategy, in which the virus is allowed to spread relatively unimpeded, we project that cumulative COVID-19 deaths would be five times higher than the initial estimates suggest. Our re-estimates of the COVID-19 HIT corroborate strong signals in the data and compelling arguments that most of the globe remains far from herd immunity, and suggest that abandoning community mitigation efforts would jeopardize the welfare of communities and integrity of healthcare systems.
ABSTRACT
Letter textThe prevalence of asymptomatic COVID-19 infections is largely unknown and may determine the course of future pandemic waves and the effectiveness of interventions. Using an epidemiological model fit to COVID-19 hospitalization counts from New York City, New York and Austin, Texas, we found that the undocumented attack rate in the first pandemic wave depends on the proportion of asymptomatic infections but not on the infectiousness of such individuals. Based on a recent report that 22.7% of New Yorkers are seropositive for SARS-CoV-2, we estimate that 56% (95% CI: 53-59%) of COVID-19 infections are asymptomatic. Given uncertainty in the case hospitalization rate, however, the asymptomatic proportion could be as low as 20% or as high as 80%. We find that at most 1.26% of the Austin population was infected by April 27, 2020 and conclude that immunity from undetected infections is unlikely to slow future pandemic spread in most US cities in the summer of 2020.
ABSTRACT
We propose a Bayesian model for projecting first-wave COVID-19 deaths in all 50 U.S. states. Our models projections are based on data derived from mobile-phone GPS traces, which allows us to estimate how social-distancing behavior is "flattening the curve" in each state. In a two-week look-ahead test of out-of-sample forecasting accuracy, our model significantly outperforms the widely used model from the Institute for Health Metrics and Evaluation (IHME), achieving 42% lower prediction error: 13.2 deaths per day average error across all U.S. states, versus 22.8 deaths per day average error for the IHME model. Our model also provides an accurate, if slightly conservative, assessment of forecasting accuracy: in the same look-ahead test, 98% of data points fell within the models 95% credible intervals. Our models projections are updated daily at https://covid-19.tacc.utexas.edu/projections/.