Read on Twitter
Comment on https://jamanetwork.com/journals/jama/fullarticle/2764658 with @sarahcobey @KeyaJoshi3 shows conclusions change dramatically if use more appropriate Wallinga-Teunis method for R(t) estimation. Comment on the site and at https://github.com/keyajoshi/Pan_response with figure
This paper reached the alarming conclusion that COVID-19 containment requires involuntary out-of-home quarantine, a view now repeated by many who should know better. The accompanying editorial questioned this extreme interpretation https://jamanetwork.com/journals/jama/fullarticle/2764656 @eliowa
We add to these questions by noting the conclusion sensitively depends on use of Cori et al. method rather than Wallinga-Teunis -- both useful for different purposes, but the choice of method here makes decline in R(t) appear later than it really is.
Bottom line: Pan et al. JAMA paper's call for mandatory out of home quarantine as a policy solution is based on faulty calculations. Do them differently and you get the opposite answer.
Also text of comment now available at https://dash.harvard.edu/handle/1/42660128
Overclaimed science makes bad policy. Based on preprint of Pan et al., prominent public health exoerts called recently for mandatory, outside-the-home quarantine of individuals exposed to COVID-19, separating families involuntarily to stem the epidemic https://www.nytimes.com/2020/04/07/opinion/coronavirus-smart-quarantine.html.
This alarming blow to Americans’ liberties is not justified by existing evidence and could have unintended negative consequences. As noted in our comment, the Wuhan study has significant methodological limitations.
NYT oped also cites examples from Korea, and Singapore to justify this policy. But actually Korea used in-home quarantine https://www.technologyreview.com/2020/03/06/905459/coronavirus-south-korea-smartphone-app-quarantine/
So did Singapore, mainly https://www.gov.sg/article/everything-you-need-to-know-about-quarantine-orders.
Fineberg, Kim, and Shlain’s recommendation also ignores the risk that individuals may avoid getting tested so that they aren’t separated from their families.
Enacting intrusive policies in the absence of strong evidence for their effectiveness sets a worrying precedent of weakly justified government over-reach.
Some interventions may be needed which exceed the powers of the state that are acceptable in normal times. But the dubious justification offered for out-of-home, mandatory quarantine is doubly dangerous.
Not only is the plan harmful to civil liberties and human rights, but it would also be highly resource-intensive, which could distract from more effective measures.
This tweet-string was composed with @yhgrad and @caroline_of_b (apart from the first sentence, an editorial comment I added).
Links provided above for Korea and Singapore say that cases usually are in homes but clearly not always
I agree w @DiseaseEcology that Voluntary centralized isolation or quarantine may be valuable, especially in densely populated areas.
The literature I have read on household transmission is quite variable, and may depend a lot on whether it is a proportion obtained during extreme social distancing, when household is the main venue left for transmission.
