November 27, 2020

expert reaction to systematic review and meta-analysis on setting-specific transmission rates of SARS-CoV-2

A systematic review and meta-analysis from the Imperial College London COVID-19 Response Team, in collaboration with multiple partners, looks at SARS-CoV-2 setting-specific transmission rates.

Prof Paul Hunter, Professor in Medicine, The Norwich School of Medicine, University of East Anglia, said:

“This report from Thompson and colleagues is a valuable contribution to our understanding of the transmission risks. The authors conducted a systematic review of the available literature and estimated the secondary attack rates and R values in different settings. The secondary attack rate (SAR) can be thought of as the probability that an individual in a group acquires the infection from the initial case.

“The first important observation is that the authors estimate that SAR for asymptomatic index cases was approximately two thirds (I think this should have read one third) of those for symptomatic index (3.5% vs. 12.8%, p<0.001). So they conclude that asymptomatic infections does occur but much less efficiently that symptomatic infection.

“Perhaps the more important observation is that SAR is dependent on the setting with big differences in SAR between settings. SAR is most common in household contacts (21.2%), then social gatherings with family and friends (5.9%) and casual close contacts (1.2%).  What this means is that if you are infectious, someone who you live with is about 20 times more likely to get the infection than would a casual close contact. This is in line with suggestions that many/most infections in the UK are acquired from a household contact. However, this does not mean that household contacts are what are driving the epidemic. We generally have many more casual contacts then we have household contacts. So social gatherings between family and friends and casual close contacts are probably what are still driving community transmission. After all if the disease only spread within household members it would die out pretty quickly. It has to spread between households to maintain itself and the only way we can effectively control the epidemic until a vaccine is available is to reduce spread between households.

“Thompson also reports that the risk of infection was about twice as much in household contacts who were together for more than five days compared to those who were together for less than 5 days. This should not be taken to mean that there is a step change in risk at the five day point. It is much more likely that the risk of infection increases gradually with increasing time together but the authors found it easier to analyse the results if they artificially allocated cases to these two time periods, something that is common in epidemiological studies, when it would be difficult to analyse the number of days otherwise.”

‘Report 38 – SARS-CoV-2 setting-specific transmission rates: a systematic review and meta-analysis’ by Hayley A Thompson et al.

https://www.imperial.ac.uk/mrc-global-infectious-disease-analysis/covid-19/report-38-transmission/

Declared interests

None received