Thread by @DiseaseEcology

New SARS-CoV-2 variant will outrun best control US states achieved in 2020.
W/out rapid action we'll lose race b/w vaccination & virus by much more than we are now.
Much faster vaccine rollout & spacing doses as UK is doing is urgently needed.

Thread.

Background
New variant of SARS-CoV-2 was detected in UK in Sept & spread rapidly since. Estimates suggest it is more transmissible with reproductive number R (# of cases/case) ~50% higher (see tweets in thread for details)

https://twitter.com/trvrb/status/1344031146808647683

More details on estimates of higher R:

https://twitter.com/jcbarret/status/1344607063658405888

Variant has been detected in many countries, including in 2 states in US (CO, CA) so far. It is likely already in most states, at low (<1%) prevalence. How will this impact next 3-6 months? Likely enormously, unfortunately.
nytimes.com/2020/12/30/hea…

It's easy to say, qualitatively, that higher R will require more control to keep R<1 (i.e. infections decreasing), or higher herd immunity by vaccination or infection before R<1.
But quantitative aspect is critical & this is what I am worried about. Here's why.

If we want R<1, & new R is ~50% higher, then we have to reduce transmission of initial virus so that R is <0.66 so that R will still be <1 if variant R is 50% higher, as data suggest. Can we do that? Data suggest Europe, Asia can, but US hasn't been able to (yet).

Unlike many European countries, US has never been able to get cases to drop fast (i.e. R far below 1). Even when R<1, only just barely. Graph shows trends in whole country cases data. What about at state level?

Crude estimates from rt.live suggest that NO US STATE has had R<0.66 at any point in time since March.
Here are data from 32 states.

Here are remaining 14 states.

More rigorous calculations of R by @sbfnk et al (that sadly don't show R estimates before Oct) are slightly more hopeful for a few states recently, but holiday reporting issues makes me less confident in these estimates.
epiforecasts.io/covid/posts/na…

Note: Even states are wrong scale for analysis. Transmission occurs within communities of interacting people. US Counties (one estimate: santacruzhealth.org/HSAHome/HSADiv…) would be far better & even finer spatial scale would be best. But what can we say based on state data?

We can say that we need to reduce R more than most ever have to push R<0.66 or else new variant will spread (R>1) & lead to cases, hospitalizations & deaths rising exponentially until large fraction of pop becomes infected. This outcome is heartbreaking given new vaccines.

What CAN we do?
-Limit spread as much as possible to delay spread of variant &
-Get vaccine to as many people as fast as possible

Limiting spread requires combination of safe workplaces, support for isolating & quarantining, opportunities to facilitate safer personal choices for gathering (e.g. outdoor heated spaces in cold places).

Rapid vaccine rollout requires resources to do so. Given urgency of situation (hospitals bursting, holiday surge impending), need to vaccinate many.

Several prominent people are recognizing this urgency & re-assessing current vaccination plan of 2 doses given 21-8d apart.

@Bob_Wachter. Here he lays out why he thinks spreading out doses as UK is doing, is worth considering, despite risks:

https://twitter.com/Bob_Wachter/status/1344667655324585986

@michaelmina_lab also weighs pros-cons of this approach.

https://twitter.com/michaelmina_lab/status/1344542385695580166

@SRileyIDD also reasons through the issues (read replies)

https://twitter.com/SRileyIDD/status/1344187338545618944

Doing so does NOT come w/out risks. 2 big ones:
-Efficacy rapidly wanes w/out 2nd boost; a double worry: immunity wanes enough to allow transmission & illness
-Waning immunity facilitates virus evolution (see @edwardcholmes reply):

https://twitter.com/PaulBieniasz/status/1344277738576900097

I'd lean towards vaccine spacing as UK is doing, but have subset w/ normal timing & collect data on efficacy to adjust strategy as needed. Also monitor viral evolution much better than we currently are in US.

Summary
-Higher transmissibility of new variant will outrun past control strategies in most places in US that couldn't get R<0.66, leading to runaway epidemic
-Maximally effective control + rapid vaccine rollout w/ increased spacing urgently needed to prevent huge loss of life

Addition:
Here's a new thread by @VirusesImmunity on impacts of spacing out 2 vaccine doses by more than 3-4 wks used in Pfizer & Moderna trials.
tl;dr She is supportive given urgency w/ spread of new variant.

https://twitter.com/VirusesImmunity/status/1345086669607890945

Addition 2: Sadly, it looks like US won't use vaccine spacing. IMO this is a huge mistake that will lead to 10000s of avoidable deaths. I hope we'll change policy soon & hope other countries will use spacing approach to save lives.

https://twitter.com/EricTopol/status/1345080202964029442