How can we distinguish the effects of a lockdown from those of a vaccine? My question is related to Covid-19 specifically, but it can be generalized to any infectious disease. I am asking this since I am reading a lot of information on several Social Media regarding one specific vaccine which is widely used in my home country.
Suppose 4 weeks ago a country decides to go into lockdown; suppose also that at the same time the same country starts vaccinating the citizens with a vaccine. If now (4 weeks later), we see a drastic reduction of positive cases and hospitalizations, how can we attribute that to the effectiveness of the vaccine or the obvious effects of a lockdown? Is it even possible?
 A: If two interventions are rolled out concurrently with no control condition, I can't see how you could dissect the effect of each on a distal outcome. There might be subtle ways of teasing this out if the two interventions had regional differences in extent or speed of rollout or compliance. Or if you could link local or temporal differences in the outcome to differences in the modes of action of these two interventions.
Which brings us to your overall question - can you distinguish these two interventions in general. So some extent we would expect to be able to. A lockdown decreases the number of contacts each person has, and therefore that any new case will have. A drop in close contacts should be a mediator of the effect of restrictions on movement. Vaccines might actually be associated with an increase in contacts, but their mode of action is (depending on the vaccine I suppose) to reduce infectivity, and thereby reduce transmissions. If a vaccine is affecting incidence, we should observe a lower proportion of close contacts who then test positive.
A: The first question that needs to be asked is: have you been able to detect any impact from lockdowns before vaccines were ever even rolled out? I have come across ~30 odd published papers/journals that have concluded that lockdowns do not have any positive impact in delivering any reduction in number of deaths due to the virus, number of hospitalisations, number of infections etc. What I have observed in official government data (depending on which country we're referring to) is the strategic implementation of lockdowns whenever the curves were already naturally dropping down again and subsequently using those observed reductions in the numbers as a means of justifying the lockdowns.
I am going to guess that we also won't be able to detect any major difference in the number of overall deaths due to the rollout of the vaccine. The vaccines might be effective, but since the people who are prone to dying from COVID are also very prone to dying from any other infection and/or natural causes, this claim can be backed up by looking at the QALYs lost for those who died from COVID. The average age of death for COVID deaths are in almost all instances approximately 1 year above that of the country's life expectancy.
