Abstract-Corbin Quick - Harvard
Title: Effects of Interventions on COVID-19 Transmission in the United States
Title: Effects of Interventions on COVID-19 Transmission in the United States
Abstract:
Abstract:
The United States (US) has experienced the largest COVID-19 outbreak in the world, with >5.4M and >170K confirmed cases and deaths respectively as of August 2020. US state and local governments have deployed multifaceted measures to contain the virus, including stay-home orders, travel restrictions, face covering mandates, and contact tracing; in recent months, many of these interventions have been relaxed and re-introduced following infection resurgence. Evaluating the impact COVID-19 interventions is critical to inform future policy.
The United States (US) has experienced the largest COVID-19 outbreak in the world, with >5.4M and >170K confirmed cases and deaths respectively as of August 2020. US state and local governments have deployed multifaceted measures to contain the virus, including stay-home orders, travel restrictions, face covering mandates, and contact tracing; in recent months, many of these interventions have been relaxed and re-introduced following infection resurgence. Evaluating the impact COVID-19 interventions is critical to inform future policy.
However, estimating the effects of interventions on transmission rates is challenging, as epidemics follow a naturally non-linear trajectory characterized by more rapid early growth, and public behavior is influenced by myriad cooccurring news reports and health recommendations over time. Also, interventions vary in implementation, adherence, and other factors, highlighting a need for comprehensive analysis across diverse regions and time periods.
However, estimating the effects of interventions on transmission rates is challenging, as epidemics follow a naturally non-linear trajectory characterized by more rapid early growth, and public behavior is influenced by myriad cooccurring news reports and health recommendations over time. Also, interventions vary in implementation, adherence, and other factors, highlighting a need for comprehensive analysis across diverse regions and time periods.
Here, we developed a Frequentist statistical framework to estimate the effective reproductive number (Rt) of COVID-19 over time across multiple geographic units as a function of baseline and time-varying covariates. We aggregated data sets on COVID-19 cases, government interventions, movement trends, and demographics across 3,073 US counties over the period March 1 - July 30, 2020 to systematically evaluate differences in COVID-19 Rt associated with 3 categories of interventions: 1) stay-home orders, 2) business closures and social gathering restrictions, and 3) face-covering mandates.
Here, we developed a Frequentist statistical framework to estimate the effective reproductive number (Rt) of COVID-19 over time across multiple geographic units as a function of baseline and time-varying covariates. We aggregated data sets on COVID-19 cases, government interventions, movement trends, and demographics across 3,073 US counties over the period March 1 - July 30, 2020 to systematically evaluate differences in COVID-19 Rt associated with 3 categories of interventions: 1) stay-home orders, 2) business closures and social gathering restrictions, and 3) face-covering mandates.
Each category of intervention was associated with significantly lower transmission rate, and face-covering mandates were associated with the largest reduction among the categories considered (rate ratio = 0.9, p < 3e-4). Finally, we present simulation studies to characterize changes in epidemic curves as a function of intervention timing, duration, and efficacy, and discuss implications for effective COVID-19 mitigation.
Each category of intervention was associated with significantly lower transmission rate, and face-covering mandates were associated with the largest reduction among the categories considered (rate ratio = 0.9, p < 3e-4). Finally, we present simulation studies to characterize changes in epidemic curves as a function of intervention timing, duration, and efficacy, and discuss implications for effective COVID-19 mitigation.