Missing infectiousness information for vaccines

Welcome to Plugging the Gap (my email newsletter about Covid-19 and its economics). In case you don’t know me, I’m an economist and professor at the University of Toronto. I have written lots of books including, most recently, on Covid-19. You can follow me on Twitter (@joshgans) or subscribe to this email newsletter here.


As I have written about previously, one of the big unknowns regarding vaccines is whether they prevent people from being contagious as well as protecting people themselves. There is hope that vaccines will do it all but it is important to know this. For instance, it is not crucial that everyone is vaccinated with a vaccine that can prevent spreading. But if that is not the case, and certainly if people believe erroneously that it is the case, there are bigger risks as vaccinated people don’t socially distance or wear masks and therefore put unvaccinated people at bigger risk. This also means that we cannot return to normal economic and social life until prevalence levels fall.

The question is: what should we be doing now to find out this critical piece of information for each vaccine candidate? One possibility would have been to collect that information along with trials. That would mean subjecting trial subjects to frequent testing. Testing can detect the presence of viral loads in individuals even if they are not showing symptoms. Why this is/was not done I don’t know. It didn’t necessarily have to be done on every subject but given that only a 100 or so per 30,000 tested actually were confirmed to have had Covid-19 during the trial period, maybe that would be a fruitless task and the entire pool needed to be tested.

That said, it will take a while for vaccines to be distributed. So there is still time to understand this and adjust our expectations in the process. This is particularly important as we really do want to know if some vaccine candidates prevent spreading and others do not.

There is one thing we could do with trial participants. We could contact trace all of them. For instance, suppose a trial participant known to have the ‘real’ vaccine rather than a placebo, was found to have close associates with high prevalence of Covid-19 during the trial period. What could this mean? On the one hand, it is further evidence of the vaccine’s efficacy in protecting you from the disease. On the other hand, it may be that the vaccine subject had Covid-19 and spread it to others. This is perhaps because they thought their risk of getting it had dropped by 50%. This seems like a conundrum but drilling down, if it was determined that the spread was asymmetric around the vaccine subject — for instance, there was an outbreak at work but not at home or vice versa — then this is evidence they weren’t a spreader. There won’t be many clusters to investigate in this way but the returns to even that small amount of information are high.

The second thing we could do is insist that those who receive the vaccine early are subject to frequent testing. These will be long term care residents and health care workers — precisely the population where it is relatively easy to test frequently. In other words, we could treat the early recipients as a new trial. The only problem is that it is not a blind trial. But it is information. If we want to guide against some distortions we could randomise the frequent testing itself.

The final thing would be to consider this issue as part of the roll-out of the vaccine itself. Perhaps vaccinate a hard-hit region first to see if this impacts on prevalence. Yes, there are distortions here — vaccinated people will take less risks — but they go in the right direction from the point of view of determining quickly if vaccinated people can still be spreaders. I’ll leave the ethics to others but since we are vaccinating anyway, at least one criterion should be to consider what we can learn from early recipients.

My point here is that this isn’t over. The above represents things that I could imagine and, I don’t need to remind anyone here, I am an economist and not someone who has any expertise in how to detect viral properties. It is critical governments get on board with, once again, finding out information quickly. There are real risks if our hope for the vaccine ends up being false.


What did I miss?