The Impact of Non-Pharmaceutical Interventions on the Second Wave of COVID-19: Insights from an Artificial Intelligence-Based, Cross-Country Study

Objective: To determine the relative influences of various non-pharmaceutical interventions (NPIs) put in place during the first wave of COVID-19 across countries on the spreading rate of COVID-19 during the second wave, taking into account national-level climatic, environmental, clinical, health, economic, pollution, social, and demographic factors. Methods: We first estimate the growth of the first and second wave across countries by fitting a logistic model to reported daily case numbers, up to the first and second epidemic peaks. Using the growth rate, we estimate the basic and effective reproduction number (second wave) Re across countries. Next, we use Random Forest, an “off-the-shelf” machine learning algorithm, to study the association between the growth rate of the second wave of COVID-19 and NPIs as well as pre-existing country characteristics (climatic, environmental, clinical, health, economic, pollution, social, and demographic factors). Lastly, we compare the growth rate of the first and second waves of COVID-19. Findings: Our findings reveal that the mean R0 and Re were respectively 2.02 (S.D 1.09) and 1.07 (S.D. 0.41). R0 has the highest value in Israel (R0 = 6.93) and lowest in Senegal (R0 = 1.13) whereas Re (second wave) had the highest value in Mexico ( Re = 3.08) and lowest in Bangladesh (Re = 1.07). The top three factors associated with the growth of the second wave are body mass index, the number of days that the government sets restrictions on requiring facial coverings outside the home at all times regardless of location or presence of other people in some areas, and restrictions on gatherings of 10 people or less. We found a statistically significant difference between the means of the first and second waves. Conclusion: Artificial intelligence techniques can enable scholars as well as public health decision- and policy-makers to estimate the effectiveness of public health policies and mitigation strategies to counteract the toll of the outbreak in terms of infections and deaths, enforcing and implementing “smart” interventions, which are as efficacious as drastic and stringent ones.