Why should/shouldn't one fear the long-term health effects of the Covid vaccines?

Question

I'm aware that some people don't want to take a vaccine for Covid-19, and in fact some countries have suspended rollout of some vaccines due to some young people getting blood clots. Also some people have had adverse reaction from the virus e.g. anaphylaxis. My question is not about any of these rare, short-term reactions to the vaccine (which I think are poor justification to refuse the vaccine, because the probability of death or serious illness from the virus is far greater than these rare reactions from the vaccine).

People in different countries are weighing the advantages vs the disadvantages of taking the vaccine. A large proportion of people in France are choosing to delay taking the vaccine, probably for various different reasons, whereas most people here in Britain are choosing to take the vaccine.

It seems to me that one of the main reasons that many people including myself are choosing to delay taking the vaccine, is: "fear of the (unknown) long-term potential health effects of taking the Covid-19 Vaccine." Many people express this opinion in the comments here:

https://www.reddit.com/r/science/comments/jpxv55/only_58_of_people_across_europe_were_willing_to/

and in the comments here:

https://www.ft.com/content/c576e15f-e5b1-4369-a5f0-073b4466036f

In fact, even in the above article itself, it says:

Among the 2,305 survey respondents, many feared long-term side effects...

I suspect that these fears are irrational and that scientists took the unknowns of the long-term effects of the vaccine into account before rolling out the millions of vaccines. Furthermore, people have to make decisions on incomplete information, and everyone does this all the time.

Now I am a layman when it comes to virology and medicine in general. I believe I have some vague understanding of how the vaccines work based on for example this video.

So my question is, what is the science behind why it is so unlikely (<1%?) that the long-term health effects of the vaccine will be harmful to a significant proportion of the population?

Or is this the wrong way of viewing things, and a better way of viewing things is: Whatever long-term harmful effects that you get from the vaccine if you take it, you are very likely to get even worse long-term effects of a similar kind the when you (inevitably at some point) catch the virus if you didn't take the vaccine. Is this correct?

Addendum: One point made by the only current answer is that, "There is no way to completely rule out the possibility of totally unexpected side effects that don't show up until much later." My response to this is that you can't completely rule out anything ever, so "completely ruling out" is irrelevant. I am asking why it is believed to be so unlikely to get side effects from vaccines - in particular the Covid ones - years later (in particular, less likely than getting seriously ill from Covid)?

This website says:

ARE THERE LONG-TERM SIDE EFFECTS? Long-term side effects following any vaccination are extremely rare. Millions of people have received COVID-19 vaccines safely.

Why does no serious side effects in the population after a few months imply no side serious side effects in the population after 10, 20, or 30+ years? Linear extrapolation of the data is surely insufficient justification (extrapolation of data is unreliable). Wouldn't one have to do long-term simulations of multi-variable analysis?

Answer 1

This general-audience article from Children's Hospital of Philadelphia gives a good overview of reasons to believe that long-term side effects are unlikely. To summarize its two main points:

• Previous vaccine research and development has found that most side effects occur within about two months. The FDA approval process for COVID vaccines required the trial to follow participants for eight weeks before submitting the vaccine for approval.
• Although the Pfizer and Moderna vaccines are the first mRNA vaccines in wide use, previous mRNA vaccines were developed and tested for Zika, influenza, and other diseases. Thus although the technology is "new" it is not as if this is the first time anyone has been injected with mRNA.

In assessing long-term side effects, one problem is you have to wait a long time, but another problem is that the longer you wait, the more difficult it becomes to track the results and determine which events can be attributed to the vaccine. Answers to this question discuss such issues with the oral polio vaccine.

There is no way to completely rule out the possibility of totally unexpected side effects that don't show up until much later. You can't really know what will happen in ten years until ten years have passed. However, all that is known from previous vaccines suggests it is unlikely that some catastrophic side effect will suddenly arise years later.

Your last question is also relevant: any decision about the use of the vaccine has to take into account the risks of not using it. Given the massive public health crisis and other societal damage caused by COVID-19, there would probably need to be significant justification for waiting, say, an extra year "just to be sure" that no side effects emerged in that time. Also, preliminary data on "long COVID" mean that it is also possible there will be long-term "side effects" of COVID-19 itself, meaning that delaying the vaccine to avoid its possible side effects could just result in people suffering different long-term effects from the disease. So even if a catastrophic vaccine side effect did arise years later, that could still be better than not getting the vaccine. (Of course, we won't know the long-term effects of COVID until time passes, just as we won't know the long-term effects of the vaccine.)

Another thing to remember is that people's subjective feelings about risk levels are often only vaguely connected to reality. I'm sure there are people out there who worry about the risk of the COVID vaccine but smoke a pack of cigarettes a day; three guesses which one of those is likely to pose a greater health risk. Some people worry about the risk of dying in a crash when they get on a commercial airline flight in the US, despite the fact that annual total deaths from such events rarely exceed zero. (Here is one classic paper on risk perception.) Fear of long-term side effects is as likely to be due to the human tendency to exaggerate unfamiliar risks as it is to be due to any actual risk from the vaccine. In other words, even apart from the scientific evidence for the safety of this vaccine, there is a fair amount of scientific evidence that how safe you think something is may not be a good predictor of how safe it actually is.

Answer 2

Since there are a number of vaccine technologies being used for Covid vaccines, a complete answer is complex.

Perhaps the simplest situation is that of the mRNA vaccines (e.g. Pfizer and Moderna). With these, a short piece of mRNA (the genetic "working instructions" that direct the production of one or more proteins) is injected into a person, enters some of their cells, and directs them to make a slightly modified version of the Covid "spike" protein. Only the mRNA and some fatty molecules that coat the mRNA and protect it until it gets into the cell are injected.

The fatty molecules are small and simple and quite safe. Although a very few people may have an allergic reaction to them, that shows up very quickly and is the reason for the post-shot 15-30 minute waiting period. We are confident that there will be no long-term effects from the tiny amount of fatty molecules.

The mRNA is essentially a small snippet of the virus's mRNA. The spike protein it codes for is used by a full-blown intact virus to attach to a cell and "inject" a full virus mRNA into that cell as the first stage of infection. However, the vaccine mRNA snippet contains no instructions for the other viral proteins, so no infection results and no working viruses are produced. The spike protein that is made appears on the cell's surface, both whole and in pieces, and the body's immune system recognizes these as foreign molecules appearing on the body's own cells, triggering an immune response. What happens over the next few weeks is that out of the body's many antibody-producing cells (B cells), each recognizing a single semi-random shape, those that match the shape of part of the spike protein get activated, divide over and over, and produce lots of antibodies matching that shape. At the same time, T cells, also each recognizing a single semi-random shape, get activated and divide if they match the shape of part of the spike protein. Some of these T cells directly kill cells showing part of the spike protein on their surface (presumably, infected cells). Some of the T cells activate B cells matching the spike protein so they produce more antibodies, which circulate in the blood.

Over a time span of months, the antibody levels slowly drop if no further spike protein is present, but a small number of the B cells and T cells continue to circulate for months or years. If they are again activated by the presence of spike protein, they quickly jump-start the reaction process to produce more B cells, antibodies and T cells in only a day or two instead of the few-week process from a first exposure. The 2nd shot of 2-dose vaccines triggers this second phase, which helps the "memory" effect last even longer and be more potent at the next exposure (i.e. when actually exposed to the real virus).

The whole process triggered by the vaccine mRNA is a subset of that in the real viral infection -- the actual viral infection also involves more viral genes and additional attacks on immune system and other cells. Thus, any reaction that is due to the vaccine should also happen in an actual viral infection, so in theory there is nothing to be gained and much to lose by forgoing an mRNA vaccine and instead waiting to get an actual Covid virus infection.

There is an additional danger in waiting to get vaccinated: the live Covid virus is mutating, and selection seems to be favoring "variants" that infect people much more easily and/or make people sicker once they are infected. Thus, the longer one waits to get a vaccine, the greater the chance of being infected by Covid once exposed, and the greater the chance of more serious illness once infected.

The above analysis does not apply for other types of vaccines, such as those that use modified or crippled helper viruses to introduce parts of the Covid virus genetic material into cells, as the additional material introduced into cells is substantially more complex than just some fatty molecules.

PS - In general, a vaccine just gives the body's immune response a "preview" of a likely future infectious agent, so that system is able to respond within hours or days instead of weeks once that agent infects the body, and clear it with little or no obvious illness symptoms. We are constantly exposed to infectious agents, triggering the specific immune system at some level probably daily. This immune response is so important to continued life that the system has been maintained and extended by evolution over hundreds of millions of years in vertebrate animals. That would likely not have been the case if it were easy to induce negative long-term consequences by triggering it (e.g. from vaccines). Thus, you can think of "vaccines" as essentially having a many-hundred-million year good safety record.