October 8, 2020

expert reaction to analysis of asymptomatic SARS-CoV-2 infection from English Household Survey data

A study, published in Clinical Epidemiology, looked at asymptomatic SARS-CoV-2 infection by analysing English Household Survey data.

Prof Patrick H Maxwell FMedSci, Regius Professor of Physic & Head of the School of Clinical Medicine, said:

“This looks to me to be rigorous and it’s certainly an important message. It underscores that many, many people who are infected are asymptomatic. So social distancing and other measures are very important, and there will be great public health benefit in terms of reducing transmission if we can reliably identify asymptomatic individuals and they then self-isolate. Testing to identify asymptomatic individuals is not as simple as it sounds and poses a range of logistic and other challenges, including requiring lots of tests. In Cambridge we are piloting an approach that involves pooling samples that we hope will enable us to run a mass asymptomatic testing programme for our students.”

Prof Paul Hunter, Professor in Medicine at the University of East Anglia, said:

“The study by Petersen and Philips cannot be used to determine the proportion of people who get infected by COVID become symptomatic or remain asymptomatic at some stage during their infection.

“This study is based on results of the ONS Coronavirus study which is a prevalence survey for people who yield a positive result and combined with whether people were symptomatic on the day of the test. So anyone who was previously symptomatic and had now recovered or who were currently incubating the infection and would develop symptoms within the following hours would not be included as being symptomatic in this study. We know that the PCR test can become positive a day or so prior to symptoms and remain positive for days or weeks after symptom recovery. So let us consider an individual who becomes infected on day 0, their PCR test becomes positive on day 4, they have mild symptoms for 3 days and then continue to shed virus for a further 14 days after recovery. This person is definitely a symptomatic case but a single random sample and symptom questionnaire would suggest asymptomatic infection on 15/18 (83%) of occasions. This does not mean that the person remains infectious for that whole time and people probably stop being infections after about a week or so.

“I think Petersen and Philip’s conclusions on the percentage of people who have an asymptomatic COVID-19 infection could be misleading. Many people reading this paper or its abstract will think that about 3/4s of people infected will remain asymptomatic and that is definitely not the case. Nor is it the case that many of these so called “asymptomatic” positive cases will be infectious. It is disappointing that the authors did not raise these really important issues in their discussion and conclusion.” 

’Three quarters of people with SARS-CoV-2 Infection are Asymptomatic: Analysis of English Household Survey data’ by Irene Petersen and Andrew Philips was published in Clinical Epidemiology at 00:01 UK time Thursday 8 October.

All our previous output on this subject can be seen at this weblink:

www.sciencemediacentre.org/tag/covid-19

Declared interests

None received.