Predicting and interpreting COVID-19 transmission rates from the ensemble of government policies

Abstract

Several questions resonate as the governments relax their COVID-19 mitigation policies - is it too early to relax them, were the policies as effective as they could have been. Answering these questions about the past or crafting newer policy decisions in the future requires a quantification of how policy choices affect the spread of the infection. Policy landscape as well as the infection trajectories from different states and countries diverged so fast that comparing and learning from them has not been easy. In this work, we standardize and pool together the ensemble of lockdown and graded re-opening policies adopted by the 50 states of USA in any given week between 9th March and 9th August. Using artificial intelligence (AI) on this pooled data, we build a predictive model (R2training=0.79, R2test=0.76) for the weekly-averaged transmission rate of infections. Predictability conceptually raises the possibility of an evidence-based or data-driven mitigation policy-making by evaluating the relative merits of the different policy scenarios. Probing the predictions with interpretable AI highlights how factors such as the closing of bars or the use of masks influence transmission, effects which have been hard to decouple from the ensemble of policy instrument combinations. While acknowledging the limitations of our predictions as well as of the infection testing, we ask the theoretical question if the observed transmission rates in the states were as efficient as they could have been under various levels of restrictions, and if the mitigation policies of the states are overdesigned. The model can be further refined with a more detailed inclusion of geographies and policy compliances, as well as expanded as newer policies emerge.

Human Mobility-Based Features to Analyse the Impact of COVID-19 on Power System Operation of Ireland

“…Shapley [44] initially proposed SHAP in 1953 based on the concept of game theory which has been used in several domains. In healthcare studies, Sruthi et al [45] predicted how different mitigation policies affect the COVID-19 transmission rate, using XGBboost combined with SHAP. Chen et al [38] examined feature importance using SHAP values to predict extubation failure based on LGBM results.…”

Machine Learning on the COVID-19 Pandemic, Human Mobility and Air Quality: A Review

“…[47] Cracow, Poland Investigating changes in pedestrian activities in public places (e.g., tourist spots, residential areas, and places with mixed land uses) before and during COVID-19. [48] 50 states of the US Assessing the impacts of policy instruments (e.g., closing and reopening of retail stores, workplaces, businesses, places of entertainment and worship, and restriction on mobility) on the COVID-19 pandemic. [49] US, Italy, Spain, Germany, France, and South Korea Understand the impacts of social distancing measures (i.e., mobility) on the transmission of COVID-19.…”

“…Travel mode Driving [45], [53], [62], [64] Transit [45], [52], [53], [62] Walking [45], [53], [62] Others (e.g., bike, air flight) [62], [64] Travel purpose Retail and recreation [7], [45], [48], [49], [53], [55], [57], [59], [61] Grocery store and pharmacy [7], [45], [48]- [50], [53], [55], [57], [59], [61] Park [45], [48]- [50], [53], [57], [61] Transit station [45], [48]- [50], [53], [57], [61] Workplace [45], [48], [49], [53], [57], [61] Residential …”

“…By building a predictive model, researchers in [48] assessed the impacts of different policy instruments (e.g., graded closing and reopening of retail stores, workplaces, businesses, places of entertainment and worship, and restriction on mobility) on the COVID-19 transmission rate in the 50 US states. With data on policy instruments collected from press releases and with data from Google's mobility reports for the period of March 9 to August 2, 2020, they used XGBboost to predict the transmission of COVID-19 under different policy scenarios.…”

How Policies on Restaurants, Bars, Nightclubs, Masks, Schools, and Travel Influenced Swiss COVID-19 Reproduction Ratios

“…In this work, we apply a recently developed formalism [13] to decipher the role of the different NPI policies adopted by Switzerland. Until April, Switzerland had the highest per-capita COVID-19 infections globally, however Switzerland could open up most of its activities staring May.…”