[ad_1]
Throughout the COVID-19 pandemic, countries have relied on a variety of ad hoc border control protocols to allow non-essential travel while protecting public health: from quarantining all travelers to restriction of entry to certain countries based on epidemiological measures at the population level such as cases, deaths or test positivity rates1.2. Here we report the design and performance of a reinforcement learning system, nicknamed “Eva”. In the summer of 2020, Eva was deployed across all Greek borders to limit the influx of asymptomatic travelers infected with SARS-CoV-2 and to inform border policies through real-time estimates of the prevalence of COVID -19. Unlike national protocols, Eva allocated Greece’s limited testing resources based on the demographic information of inbound travelers and the test results of previous travelers. By comparing Eva’s performance to modeled counterfactuals, we show that Eva identified 1.85 times more asymptomatic infected travelers than random surveillance tests, with up to 2-4 times more during peak trips, and 1.25-1.45 times more asymptomatic travelers, infected travelers as testing policies that only use epidemiological measures. We demonstrate that this latter benefit stems, at least partially, from the fact that population-level epidemiological metrics had limited predictive value for the actual prevalence of SARS-CoV-2 in asymptomatic travelers and exhibited strong country-specific idiosyncrasies. in the summer of 2020. Our results raise serious concerns about the effectiveness of the border control policies proposed at the international level and independent of the countries.3 based on epidemiological parameters at the population level. Instead, our work represents a successful example of the potential of reinforcement learning and real-time data to protect public health.
[ad_2]
Source link