Preprints with The Lancet is part of SSRN´s First Look, a place where journals identify content of interest prior to publication. Authors have opted in at submission to The Lancet family of journals to post their preprints on Preprints with The Lancet. The usual SSRN checks and a Lancet-specific check for appropriateness and transparency have been applied. Preprints available here are not Lancet publications or necessarily under review with a Lancet journal. These preprints are early stage research papers that have not been peer-reviewed. The findings should not be used for clinical or public health decision making and should not be presented to a lay audience without highlighting that they are preliminary and have not been peer-reviewed. For more information on this collaboration, see the comments published in The Lancet about the trial period, and our decision to make this a permanent offering, or visit The Lancet´s FAQ page, and for any feedback please contact email@example.com.
Analysing the Effect of Containment and Mitigation Measures on COVID-19 Infection Rates Using Machine Learning on Data of 95 Countries: An Observational Study
27 Pages Posted: 18 May 2020More...
Background: Insights on the effect of containment and mitigation measures to control the global COVID-19 outbreak are mostly derived from simulation studies, which work with assumptions based on fragmented knowledge, and a high level of uncertainty. We propose a complementary approach to evaluate the effectiveness of various measures on COVID-19 infection rates and the time these measures need to become effective.
Methods: A non-linear machine learning model was used on data of 95 countries to assess the effect of 31 different containment and mitigation measures on the infection rate (i.e., growth rate of daily cumulative confirmed COVID-19 cases). Accumulated local effect (ALE) plots were used to visualise the changes in the growth rate starting from 14 days prior to measure implementation to 40 days after.
Findings: The model identified three important measures to reduce infection rates: school closures, limit public gatherings, and public services closures. On average, growth rates started to decrease about seven to ten days after implementing the measure.
Interpretation: Our analysis of empirical data indicates that strict reduction of social contact is the most effective way to reduce infection rates of COVID-19. School closure is highlighted as an important measure by the model, which sheds new light on this scientifically and politically controversially discussed measure.
Funding Statement: The authors received no specific funding for this work.
Declaration of Interests: The authors declare no competing interests.
Keywords: COVID-19; coronavirus; mitigation measures; containment measures; government measures; non-pharmaceutical interventions; machine learning; accumulated local effect plots; infection rate; cross-country study
Suggested Citation: Suggested Citation