Estimation and worldwide monitoring of the effective reproductive number of SARS-CoV-2

Abstract

The effective reproductive number Re is a key indicator of the growth of an epidemic. Since the SARS-CoV-2 pandemic started, many methods and online dashboards have sprung up to monitor this number. However, these methods are not always thoroughly tested or are applied only to a limited geographic range. Here, we present a method for near real time monitoring of Re, applied to epidemic data from 170 countries. We thoroughly validate the method on simulated data, and present an intuitive web interface for interactive data exploration. We show that in the majority of countries the estimated Re dropped below 1 only after the introduction of major non-pharmaceutical interventions. For Europe, Asia, and North America we found that the implementation of non-pharmaceutical interventions was associated with reductions in the effective reproductive number. Globally, we found that relaxing non-pharmaceutical interventions did not fully revert Re values to their original levels. Generally, our framework is useful both to inform governments and the general public on the status of the epidemic in their country, as well as a source for detailed comparison between countries and in relation to local public health policies and external covariates such as mobility, behavioural, or weather data.Significance statementDuring the SARS-CoV-2 pandemic, governments need a way to monitor the epidemiological situation in their country. A key indicator is the effective reproductive number Re. It describes the average number of secondary infections caused by a primary infected individual. Here, we present a method to estimate Re from case report data. We thoroughly validate the method on simulated data, and present Re estimates for 170 countries on an interactive web interface. We then use this method to investigate the impact of non-pharmaceutical interventions on reducing Re worldwide. We find that the estimated Re was significantly above 1 prior to the introduction of major non-pharmaceutical interventions, and that relaxing these interventions does not fully revert Re estimates to their prior levels.

Practical considerations for measuring the effective reproductive number, Rt

“…This method was applied to AIDS cases, which feature long delays from infection to observation [ 48 ], and in the reconstruction of incidence from mortality times series for the 2009 H1N1 (swine flu) pandemic [ 46 ]. It is now being applied to COVID-19 [ 8 , 49 ]. Fig 5 shows an example of deconvolution applied following the methods in [ 46 ].…”

“…The simplest approach is to drop estimates on the last few dates or to flag them as unreliable [ 8 ]. But many methods are available to adjust for right truncation, which can improve real-time analysis [ 42 , 54 , 59 ].…”

“…If changes in ascertainment are well understood and measurable, they can be incorporated into or used to adjust R t estimates. For example, several R t estimation dashboards currently adjust for testing effort [ 6 , 8 , 11 ]. Observed hospitalizations and deaths may be less sensitive to changes in test availability and testing effort, but may be more sensitive to other factors, such as hospital bed availability.…”

Dynamics of COVID-19 under social distancing measures are driven by transmission network structure

“…In many regions around the world, the effect of social distancing interventions is monitored in real-time using estimates of the time-dependent reproduction number R t (e.g. [ 22 ]). We applied standard procedures for calculating R t [ 23 ] to the incidence data from our simulations, and using the time at which R t first crossed the threshold of 1 as a measure of the delay, we found that the trends agreed with those reported for the epidemic peak ( S8 Fig ).…”

Detection of a SARS-CoV-2 variant of concern in South Africa

“…We obtained the data from the Report on Weekly Deaths from the South Africa Medical Research Council Burden of Disease Research Unit 40 . We generated estimates for the effective reproduction number (Re) of SARS-CoV-2 in South Africa from the covid-19-re data repository (https://github.com/covid-19-Re/ dailyRe-Data) as of 14 December 2020 41 .…”