Stay Home

Life COVID-19

Stay home.

I’ve been talking to folks 1:1 about this, but from a scroll through the feed today, I don’t think the general community has caught on. COVID-19 is not fucking around. If we don’t contain or dramatically slow it, we are going to run out of health care workers, hospital beds, and equipment. People are going to die for want of care. This is not a problem of the distant future: recent modeling suggests that without a significant reduction in social contact, Seattle will exhaust healthcare capacity around two weeks from now. Other regions will not be far behind.

This doesn’t mean panic. This means we need to take calm, decisive action to reduce transmission.

COVID-19 is containable. Surveillance data from China, South Korea, and Taiwan suggests their efforts are working, but the US has not implemented these kinds of aggressive measures yet. Our surveillance is limited by test shortages, and we have not performed the kind of contact tracing, isolation, and social distancing they have. I hope we get there, but before that happens, it’s up to us to do our part voluntarily–and we have to act NOW. What we do in the early days of an outbreak has an outsized impact on later trajectories.

Cancel your events. I know. Cancel them. Stay home, or away from people outdoors, as much as possible. Wash your hands frequently and thoroughly with soap and water. Stay six feet apart. Now is not the time for bar events, for contests, for house parties, for travel. Tell your friends and neighbors to stay home too.

This means you. Even if you’re healthy, even if you’re young and likely to survive: you may unwittingly infect others, and this is about everyone’s health. Current models suggest each case infects roughly two more–and those infect four more, eight, sixteen, thirty-two–doubling every six days. Social distancing–staying home, avoiding contact, etc–reduces that ratio. It’s the key to flattening the epidemic curve, and making sure that as people get sick, there are beds, supplies, and health workers to take care of them.

For practical guidance from epidemiologists on what social distancing looks like in real life, see The Atlantic’s reporting.

You can read and sign up for daily situation reports from John Hopkins and the WHO.

Geographic and timeseries visualizations of case reporting are available from the University of Washington, the Center for Systems Science and Engineering, and the COVID-19 Tracking project.

When you read this reporting, keep in mind that confirmed cases are an underestimate: in the US, we suspect there could be orders of magnitude more cases going undetected. Also note that deaths lag roughly three weeks behind initial infection.

For projections of Seattle’s caseload, see the Institute for Disease Modeling’s working draft.

For a clinical retrospective of hospitalized-case outcomes in China, see this article in the Lancet.

Best wishes, everyone. Be kind & conscientious with each other. <3

Albert
Albert on

Instead of tucking us in like little children, can you please write a good technical article on a subject of your choosing, jut like you did way before?

I promise, I will sit down at my computer, read it thoroughly, and play with the code, all of that at home.

Aphyr on

Dear Albert: I could not possibly roll my eyes any harder.

Aphyr
Justin
Justin on

I wish my Country, India had taken this advise seriously. C19 is not fucking around, it is killing people in the tens of thousands overwhelming every aspect of healthcare and even cremation of bodies. What initially protected us was preventive measures, then everyone became complacent and now are paying a heavy price.

anonymous on

I don’t think the general community has caught on.

Sorta depressing that over a year, 600k dead, and several vaccines later that this is still far too often the case…

Erich
Erich on

That first comment aged really well

anonymous on

So how did that “two weeks to flatten the curve” thing work out?

Aphyr on

So how did that “two weeks to flatten the curve” thing work out?

Modeling this kind of thing is actually a really tough research problem, but there has been a fair bit of observational analysis done! You might start with Fowler, Hill, Levin, & Obradovich 2021 which tries to model the impact of stay-at-home orders. I’d read the introduction & discussion sections in depth–they talk at length about the challenges of causal inference, associated non-pharmaceutical interventions, individual action including voluntary contact reduction and noncompliance, and so on. Then I’d look at their timeseries plots for states which implemented stay-at-home vs aggregate states not implementing those controls. They suggest (very loosely):

These results suggest that stay-at-home orders reduced confirmed cases nationwide by 390,000 (170,000 to 680,000) and fatalities by 41,000 (27,000 to 59,000) within the first three weeks. These estimates suggest reductions in cases of 25% and in fatalities of 35%, respectively. As with all counterfactual projections, our estimates here rely on the estimating assumptions built into our empirical analysis and should be interpreted accordingly.

Aphyr

Post a Comment

Comments are moderated. Links have nofollow. Seriously, spammers, give it a rest.

Please avoid writing anything here unless you're a computer. This is also a trap:

Supports Github-flavored Markdown, including [links](http://foo.com/), *emphasis*, _underline_, `code`, and > blockquotes. Use ```clj on its own line to start an (e.g.) Clojure code block, and ``` to end the block.