Thread by @jljcolorado

1/ A big COVID outbreak takes 2 ingredients:

- someone highly infective (most infected people don't seem to be infective or little; the highly infective are so during a short period)

- The right indoor conditions for accumulating and breathing in the virus.

https://twitter.com/charliesmithvcr/status/1485730807360548866

2/ In this paper we showed that the outbreaks that have complete data for analysis are explained by airborne transmission in shared indoor air

[Detailed explanation on that thread]

https://twitter.com/jljcolorado/status/1479250147330637825

3/ The level of virus in the air is analogous to the level of water in a sink, depends on:

- faucet: emission rate of virus

- size of sink: volume of room

- size of drains: ventilation, filtration etc.

[Analogy not perfect but good for illustration]

https://twitter.com/jljcolorado/status/1479260955959832576

4/ The emission rate of the virus depends on:

- Viral load / aerosol emission rate of the infected person(s). Very highly variable in time and between people.

- How much time they spend in space

- Vocalization: far more emission if talking, even more if yelling singing

5/ Virus emission also:

- seems to increase with heavy breathing, e.g. exercise in gyms.

- And decreases with masking (some with cloth, more with surgical, a lot more with well-fit N95)

6/ Then how much virus is breathed in by the susceptible people (SP) present depends on:

- The level of virus in room (sink analogy in prev. tweet)

- How much time SP spend in the space

- Breathing rate of SP (e.g. low if sedentary, high if dancing, gym...)

- masking of SP

7/ You can combine all those effects rigorously (*) into single number: Relative Infection Risk Parameter (Hr)

*: within approximation that the air mixes faster than people enter and leave the space, as e.g. with someone lighting a cigarette in a room

https://twitter.com/jljcolorado/status/1479262364046397443

8/ The attack rate on the outbreaks from the scientific literature (famous restaurant, choirs, buses, call center, meat packing, schools...) falls in line with Hr.

9/ So not a mystery where the outbreaks happen, we know the conditions that favor them (same as 3 Cs of Japan or many other studies: nature.com/articles/s4158…).

But very random whether an outbreak will happen, depending on infrequent highly infective individuals being present

10/ Definitely not surprising that cars and buses can have a lot of outbreaks, since they are small "rooms" with significant time spent there, and many of them are poorly ventilated

https://twitter.com/mustafahirji/status/1485738083526000640

11/ This is explained in our recent @ConversationUK article (led by @trishgreenhalgh) where we provide this chart:

[You can see numbers and modify chart at tinyurl.com/COVID-Tables]

theconversation.com/heres-where-an…

12/ We have a short paper under review where we propose that Hr should be calculated for all indoor public spaces and posted (for typical use without mitigations, with various mitigations). And then use that to guide mitigations, occupancy etc.

13/ Now, given how slow Public Health has been to accept airborne transmission during the pandemic, I won't hold my breath till this approach or something like it is put in place widely.