COVID-19: Do The Vaccines Prevent Transmission?
by Raywat Deonandan, PhD
Epidemiologist & Associate Professor
University of Ottawa
(I add my credentials to these COVID-19 blog posts in case they get shared. I want readers to know that my opinion is supposedly an educated and informed one)
Before we get into today’s topic, I want to thank Canada’s longest running comedy show, This Hour Has 22 Minutes, for a funny bit at the end of this week’s episode that mocked “TV pandemic doctors”. They were kind enough to include photos of yours truly, and made me look like the hardboiled lieutenant in a 1970s cop thriller:
Today’s topic was inspired by this tweet by Cornell’s Brian Wasik, virologist:
Dr Wasik was responding to a statement by Fox News that, “We now know that the vaccines do not prevent infection and transmission, so any community-spread benefit is negligible.” Dr Wasik’s reply: “Vaccination reduces risk of infection. There is more than Omicron out there. Infection VE in the real world is clear. Look at Omicron-hit NYC through 12/25, 4X reduced risk.”
So let’s talk about whether these vaccines do reduce infection and transmission, and how we measure that.
Back in December of 2020, when the COVID mRNA vaccines were about to be authorized by Health Canada, I was being interviewed on CBC radio. The host asked, “What questions would you like to have answered about these vaccines?”
My response, “Do they just prevent symptoms, or do they also prevent transmission.”
The host was shocked and audibly gasped. “That seems like a big deal!” she responded, and clarified that she’d never considered that possibility. I was quick to point out that we have examples of vaccines that do not prevent asymptomatic infection, but which succeed in making the disease mostly go away anyway. Rotavirus comes to mind.
As the months have progressed, the question of the vaccines’ ability to prevent or slow transmission has become politicized, especially as new variants have emerged that challenge vaccine efficacy. What is certain is that the vaccines remain astonishingly good at preventing serious disease, hospitalization and death. There is no credible argument against that, as per these data from Alberta, which show that the hospitalization rate among thrice-vaccinated 80+ year olds is lower than the hospitalization rate among unvaccinated 12-29-year-olds:
And these data from Switzerland, showing that the unvaccinated are far more likely to die. This is doubly astonishing since the unvaccinated are more likely to be younger and healthier than the vaccinated:
When looking at risk of infection, though, the story is not as clear. As noted in my previous post, “Let’s Talk Again of the Base Rate Fallacy“, since Christmas Ontario is showing a paradoxical increase in infection risk among the vaccinated, which I attribute to a combination of testing bias and exposure bias.
So Omicron might have confused the issue somewhat. Let’s try to disentangle it a bit more.
Infection vs Transmission
The first thing we have todo is to separate infection from transmission. It’s one thing to wonder whether becoming vaccinated will prevent you from getting infected with the SARS-CoV2 virus. But it’s another to wonder whether, once infected, the vaccine would prevent you from infecting others, i.e. blunting transmission.
These are not mutually exclusive constructs. Clearly, if the vaccine prevents you from becoming infected, then you can’t pass it on to others. But not so clearly, if you were to suffer a “breakthrough” infection despite being vaccinated, do you have the same probability of passing it on to your contacts?
Vaccine Protection Against Infection
The human body’s immune system essentially has two arms: the humoral arm and the cellular arm. Humoral immunity comes by way of antibodies, which we produce to keep infectious agents out of our cells. They are the first line of defence. If the antibodies can neutralize the invaders, then viral replication doesn’t happen; we don’t get infected; and we don’t get sick. Think of the antibodies as the front-line troops in a war. They repel invaders at the border.
But once the invaders penetrate that border, we have another line of defence, the “home guard”, or fifth column, or whatever military analogy you’d like to make. The cellular arm of the immune system has an array of warriors, including T-lymphocytes, that become active only once infection has occurred. But they’re the beasts that prevent the virus from causing serious disease.
Humoral immunity often wanes after a few months –as was observed in the COVID vaccines. But cellular immunity is more durable, and can often last decades or even a lifetime. So don’t be surprised when a vaccine that doesn’t offer much or lasting protection from initial infection nonetheless confers extraordinary and durable protection against symptomatic and/or serious disease and death.
Remember that the famous initial clinical trials assessing the Pfizer and Moderna shots found that they were about 95% effective at preventing symptomatic COVID.
But those trials were not designed to test whether the vaccinated were less likely to become infected at all, nor whether breakthrough infections would result in diminished transmission. The reasons are that: (a) the goal of the vaccination program was to save lives and remove strain from the health care system, so symptomatic disease is what is most important; and (b) to properly answer the questions of whether the vaccines affect risk of transmission and infection, the thousands of trial participants would have to be PCR tested on a nearly daily basis, as would their household contacts, and that’s just financially and practically untenable.
Do the Vaccines Lower Your Chances of Becoming Infected?
In a word: yes.
Once the vaccines started being used, post-market approaches could be used to determine the reduction in infection risk. One method is to randomly select households in a population and test everyone therein, irrespective of vaccination or symptom status.
UK data, assessed during a time period that included when the Delta variant was circulating, showed that the vaccines reduce the chances of getting infected by 50%-75%. Another UK study found that two doses of the Pfizer vaccine were 80% effective against preventing all Delta infections.
And the NYC data posted above by Dr Wasik further shows that the vaccinated have overwhelmingly shown a lesser probability of becoming infected.
Do the Vaccines Lower Your Chances of Transmitting the Virus?
In a word: yes.
Now, if you give this some thought, you’ll realize there isn’t an obvious and clear-cut way of practically answering this question. But epidemiologists have some concepts that can help. The first is something called an “attack rate”, which is roughly the proportion of a defined population that becomes newly infected with a disease over a define time interval. This doesn’t necessarily mean that all the people in the denominator were exposed to the disease, especially if the population we’re looking at is a province or country and not, say, a daycare or company picnic. But this is what we have to work with.
Comparing the attack rates among the vaccinated and unvaccinated can tell us how good the vaccines are at preventing actual infection.
Then there’s a thing called a “secondary attack rate” or SAR, which is the proportion of the contacts of the infected people in the preceding paragraph who end up becoming infected. This is an important measure because it tells us how the disease spreads in different settings, the extent to which the disease moves person-to-person once it’s been introduced into the population, and possibly how effective mitigation tools are at changing transmission.
Comparing the SARs of two groups of people (like the vaccinated infected and unvaccinated infected) tells us which group is more or less likely to transmit the disease once infected.
There’s also a thing called “vaccine effectiveness against transmission” or VET, which is more-or-less a ratio of attack rates, though it’s often computed using more complicated statistics. The important point is that the higher the VET (which is a percent), the more likely it is that the vaccine prevents transmission.
So what do we know? A Dutch study from early 2021 computed a VET of 70% for the Pfizer jab and 88% for Moderna. Similarly, a Chinese pre-print study in the summer of 2021 (when the Delta variant was circulating) found that, by comparing SARs, the unvaccinated had almost three times higher odds of transmitting to household contacts than did vaccinated people with breakthrough infections. And an Israeli pre-print study from December of 2021, also when Delta was dominant, computed a VET of 88.5% for the Pfizer vaccine.
How is it possible that an unvaccinated infected person and a vaccinated infected person can be so differently able to transmit? With the Delta variant, there was some panic that breakthrough infections resulted in the same amount of viral load as in infected unvaccinated people, leading to the speculation that they were equally as likely to transmit. However, many studies show that vaccinated people clear that viral load faster, so are infectious for a shorter period of time. Thus, at a population level, vaccinated infected people transmit less than do unvaccinated infected people.
So a combination of lesser likelihood to carry viral load for long durations (hence lowering SAR), combined with a lesser likelihood of getting infected in the first place, means an impressive VET for these vaccines.
But What About Omicron?
The Omicron variant has changed the game in many ways. The very high rate of breakthrough infections presents a significant wallop to vaccination’s ability to reduce initial infection —though it is still excellent at preventing serious disease and death!
Danish data from the end of December, 2021, found that individuals in unvaccinated households had 1.41 times the odds of being infected with any variant, as compared to those with two doses; whereas those with three doses had 39% lesser odds of being infected than those with only two doses.
But when the researchers only looked at Omicron infections, the odds of infection in the unvaccinated households were only 4% more than in the double-dosed households.
However, they also found that booster shots halved the risk of being infected. So the takeaway for me is that Omicron has appreciably compromised the ability of two doses of vaccine to slow transmission. But that third dose ratchets it back up there. In other words, the booster shots work to significantly reduce the chances of being infected by Omicron!
UPDATE: The sense that 3rd doses return vaccines’ ability to prevent infection to a very high level is confirmed by this pre-print study that measured efficacy against infection to be 70% after the booster shot. I would also direct my gentle readers to Eric Topol’s substack article, which includes this graphic:
It’s pretty clear that prior to the dominance of the Omicron variant, two doses of an mRNA vaccine conferred significant dampening effect on the ability of the virus that causes COVID-19 to spread in a community. If enough people had become vaccinated prior to the Omicron era, the epidemic would have been over.
However, Omicron has changed that calculus considerably. Data is severely limited. But what of it exists suggests that the ability of two doses to significantly quell the spread of Omicron has been greatly compromised.
But but but…. two things are becoming clear. First is that two doses are still extraordinarily good at preventing hospitalization and death. And second, it looks like that third dose, the booster, will probably once again reduce transmission significantly. Vaccination remains our best tool of quashing this pandemic in a way that minimizes human suffering and death.