Thread by @peripatetical

Why would we use a vaccine that mightn't prevent transmission? Should we use one particular vaccine, or roll out the vaccines we have as broadly and quickly as possible?

I'd argue that we should use all available vaccines that prevent COVID, even if they don't reduce transmission as much as we'd like.

Vaccines can protect people both directly and indirectly. If you get an effective vaccine, you directly benefit. You have a reduced risk of getting the disease.

However, if enough people get vaccinated, then even if you don't (or can't) get vaccinated, you have a reduced risk of getting infected. This is because you aren't likely to get infected if everyone around you doesn't have infection.

This is known as "herd immunity". The proportion of people that need to be immune to achieve herd immunity depends on the infectiousness of the disease.

(We're familiar with R0 - the average number of secondary cases that result from a primary case. The herd immunity threshold is approximately 1 - 1/R0, but can be higher or lower).

So, can COVID vaccines give us herd immunity? We currently don't know. The AZ vaccine reduces symptomatic infection by 70% (62% in those that received the standard dose).
thelancet.com/journals/lance…

However, in a small subgroup of participants that had routine tests while asymptomatic, it only reduced asymptomatic infection by 8%.

As overall infection was reduced, this would suggest that the AZ vaccine does reduce infectiousness so some degree. But even if all adults were vaccinated, it probably wouldn't achieve herd immunity.

In the Moderna vaccine study they only did asymptomatic swabs routinely before the second dose and found a lower number of asymptomatic infections in the vaccine group (14/14134) than in the control group (38/14073)
fda.gov/media/144453/d…

In the Pfizer vaccine study, they are planning to do N serology to see if participants get asymptomatic infection, but the results aren't available yet.
pfe-pfizercom-d8-prod.s3.amazonaws.com/2020-09/C45910…

So, these vaccines prevent disease, and some may reduce infectiousness to an uncertain and varying degree. This is not unlike the flu vaccine - it reduces infection by about 50% but we don't get herd immunity.

But there is still benefit in getting a vaccine that protects you, even if it may not block transmission. Getting the vaccine means that you have a reduced risk of getting sick.

Australia has access to 10 million doses of the Pfizer vaccine that is being delivered over the year. We also have access to 53 million doses of the AstraZeneca vaccine (3.8 million doses available soon, and roughly 1 million doses a week).
health.gov.au/initiatives-an…

We also may have some more doses of different vaccines via the COVAX facility. There is also another vaccine that is under development (Novovax) but the phase 3 studies have only just commenced
nih.gov/news-events/ne…

The Pfizer vaccine looks about as good as it gets - it appears to reduce symptomatic infection by about 95%, even if we don't know yet whether it reduces asymptomatic infection

But we if only used this vaccine, we could only vaccinate 5 million people (~20% of the population) over the next year. This wouldn't achieve herd immunity even if it completely prevented infection and infectiousness (which we don't know yet).

The AZ vaccine may not be as good (it reduces infection by about 62-70%) but this can be rolled out more quickly. Even if doesn't protect against transmission, it does protect against disease and that's a benefit.

The choice we have isn't whether to use one vaccine or the other. Our choice is whether to offer everything we have now to protect as many people as we can, or to leave some effective vaccines in the warehouse.

That said, I understand that there are ongoing discussions with vaccine manufacturers. But if you were in charge of a vaccine manufacturer, would you send your supply to a country where there are thousands of deaths a day or to Australia?

A few follow ups to this thread:
1. I made an error in the tweet about the effectiveness of the AZ vaccine against asymptomatic infection - overall, the reduction was 27%, but was 3.8% in the standard dose group (thanks @EmmaTruswell for pointing this out)

2. Can you get another vaccine of a different type after getting an initial vaccine course? There aren't any data yet, but from first principles a later booster dose should augment protection after the primary two doses.

The exception is that if you get two doses of the AZ vaccine first, you probably can't get a booster with that same vaccine later. However, the use of "mixed" schedules (with different vaccine types) needs to be studied.