Intended for healthcare professionals

Rapid response to:


Covid-19: Eight day quarantine is as good as 14 for returning travellers, study finds

BMJ 2020; 370 doi: (Published 30 July 2020) Cite this as: BMJ 2020;370:m3047

Read our latest coverage of the coronavirus outbreak

Rapid Response:

Study finding 8 day quarantine may be "as good" (but not the same) as 14 for returning travellers, is based on assumptions most likely to change, potentially and rapidly rendering it invalid

Dear Editors

I am responding to the favourable reporting of this study by Mr Day in which he enthusiastically declared that the reasearchers had shown "quarantining people who return to the UK from the US or the EU for eight days, with a polymerase chain reaction (PCR) test on day 7, is virtually as effective at preventing new community COVID-19 infections as the current two week quarantine policy,"

Despite repeated reading and text search of the non-peer-reviewed paper, I am unable to find the portion of the article eluding to the idea that an eight day quarantine and a 7th day PCR testing is “virtually as effective at preventing new community COVID-19 infections” as the current two week quarantine policy. The focus of the study was in fact whether a shorter quarantine period and exit testing will have any difference, compared to current 14 day quarantine, in reducing the number of infectious arrival being released into the community following the quarantine. This does not directly translate to “preventing new community infection” as I will explain why later.

I accept that being a theoretical study (rather than ad-hoc analysis of observational data), multiple assumptions have to be made by the researchers with respect to COVID-19 prevalence in UK community (9 per 10,000) and travellers from US (40 per 10,000) and EU (2.8 per 10,000) as of 20 July 2020, PCR testing sensitivity and specificity of symptomatic, presymptomatic and asymptomatic travellers over various stages of COVID-19 disease.

Mr Day did not appear to put any emphasis on the caveats voiced by the researchers themselves about these assumptions, nor the fluid nature of the estimation of numbers of infectious travellers released into the community after quarantine (the authors actually have to update their paper data from June 2020, to July 2020 (being aware of the importance of rapidly changing prevelance in various comparison regions)

I would strongly recommend readers read the paper, regarding the researchers’ risk mitigation strategies and detection models, 5 paragraphs (900 words) of assumptions explained by the author,s if they want to consider using this paper in any advocacy on quarantine matters.

Notwithstanding the trust in screening abilities and vigour in excluding symptomatic travellers from departing ports in US and UK, the optimism in individual ability to self-diagnose/self-isolate and adhere to public health directions when symptomatic, and reliability of data about infectious duration and “infectiousness” based on other studies on this newly-identified infection, it is clear to any reader that the conclusion about the 8-day quarantine can only be valid if the travel volume (estimated at 1% of 2019 volume) and COVID-19 prevalence of US, EU and UK. As the international travel is expected to increase when COVID-19 cases appears to reduce (albeit temporarily) and COVID-19 prevalence will change when nations experience a second wave of transmission, the suggested quarantine for 8 days for all incoming travellers will not likely be as vigorous as 14 days quarantine in limiting infectious travellers into UK.

Furthermore, for countries with very low or no prevalence of COVID-19, the risks of infection transmission by infected travellers released into the community will be likely to be unacceptable socially and politically regardless of what the scientific community considers acceptable. As UK’s caseload regresses over the next few months and travel restrictions within UK relaxes, we may find that UK residents will have no appetite for a resumption of lockdown when additional handful of infected travellers (however statistically “marginal” in numbers) result in a resurgence of community transmission; the latter a possible scenario in the current spike of cases in the Australian state of Victoria in July 2020, singlehandedly responsible for doubling national tally of confirmed cases (3.1 per 10,000) to 7.0 per 10,000 for the entire Australia in 2020 so far.

Hence the risk-mitigation assessment of 8 day quarantine may be statistically acceptable as 14 days as of 20 July 2020, but as the international situation remains fluid, so will be the validity of this assessment as expected.

Competing interests: No competing interests

03 August 2020
Shyan Goh
Orthopaedic Surgeon
Sydney, Australia