Anything that depends on the weather and which is run in a businesslike manner will be heavily reliant on statistics and in particular the probability of various scenarios and the known extremes. Examples include agriculture, water supply, short term operations of electricity supply and so on.
A share trader trading a mechanical system will ultimately be relying on some set of statistical data, in particular the probability of various scenarios and the known extremes, in order to determine that the system gives them an edge and doesn't blow up. The details used will vary according to the situation but stand back and that's what it comes down to.
When it comes to vaccination the same applies. It's possible that a vaccine might kill you yes but based on available knowledge the probability massively favours vaccination being less likely to harm you than the disease it seeks to prevent.
That's much like saying that under certain scenarios it's possible that someone dies in a car crash because they were wearing a seat belt or because of the airbags deploying. That is possible indeed there are reported occurrences. Likewise at a guess it probably has happened that a building burnt down because the fire alarm caught fire. Theoretically it's possible and odds are it has happened somewhere at least once. Statistically however, you're far less likely to be killed by the belt, bag or alarm than that same belt, bag or alarm is to save your life. Statistical probability favours having it despite the new risk it creates.
In the absence of certainty I'll pick proper statistical data over random thoughts any day. There's no guarantee that turns out to be the right decision but most likely it will be and there's no "do nothing" option here, there's no option to simply say we'll do nothing right now and will collect some more data before we make a decision as someone might do with a proposed share trading system or the construction of a dam. In this case there's no option to do nothing, the virus is already up and running.
The economic impacts will continue rising until we've got a critical mass of people vaccinated. Only then, and I'm basing this on reports of the overseas experience, can we get back to business with any real certainty.