Coronavirus

Covid-19 vaccines: What are the risks and benefits for each age group?

EL PAÍS compares the likelihood of developing a blood clot from the Janssen and AstraZeneca injections with the probability of suffering a serious case of the disease, according to each demographic

All licensed vaccines are extremely effective in preventing severe cases of Covid-19, a disease that has killed around three million people worldwide in the space of one year. However, there are still individuals who are more afraid of the vaccine than of the virus. The focus now is on the Oxford-AstraZeneca and Janssen vaccines, after dozens of those who were vaccinated ­– about 1 in 100,000 – developed a very rare form of potentially lethal blood clots.

Below is a sample of risk-versus-benefit calculations. While approximate, they are useful for understanding the kind of figures used by drug regulatory agencies. We start by looking at a general scenario of 100,000 adults of all ages over a period of four months with an incidence of Covid-19 that reflects the rates experienced in 2021 in Spain:

Without vaccination

With vaccination

100,000

people

Without vaccination

With vaccination

If 100,000 people don’t get vaccinated, around 3,800 people will become infected

If 100,000 people do get vaccinated, around 950 people will become infected

Fewer infections

If 100,000 people get vaccinated, around 2,850 people will avoid catching Covid-19 in this example. And that’s without taking into account the fact that getting vaccinated will also put the brakes on transmission.

If 100,000 people don’t get vaccinated, around 250 will end up in hospital

If 100,000 people do get vaccinated, around 13 will end up in hospital

Far fewer hospital admissions: the vaccine will prevent 237 hospitaliza- tions (19 out of 20 or more)

If 100,000 people do get vaccinated, around 2 will be admitted to an ICU

If 100,000 people don’t get vaccinated, around 25 will be admitted to an ICU

The admissions to ICUs will be one or two, if not fewer, instead of 25.

If 100,000 people do get vaccinated, around 3 will die

If 100,000 people don’t get vaccinated, around 60 will die

57 people

(out of 100,000)

will avoid death

One serious

case of

thrombosis

From a sample of 100,000 people vaccinated, over the space of four months one person will develop thrombosis

But around 20 ICU cases and between 50 and 60 deaths will have been avoided

Note: In this scenario, we have assumed a 3.8% infection rate in Spain – which is the rate registered between January and now; also that the vaccine is 75% effective.

100,000

people

Without vaccination

With vaccination

If 100,000 people don’t get vaccinated, around 3,800 people will become infected

If 100,000 people do get vaccinated, around 950 people will become infected

Fewer infections

If 100,000 people get vaccinated, around 2,850 people will avoid catching Covid-19 in this example. And that’s without taking into account the fact that getting vaccinated will also put the brakes on transmission.

If 100,000 people don’t get vaccinated, around 250 will end up in hospital

If 100,000 people do get vaccinated, around 13 will end up in hospital

Far fewer hospital admissions: the vaccine will prevent 237 hospitalizations (19 out of 20 or more)

If 100,000 people do get vaccinated, around 2 will be admitted to an ICU

If 100,000 people don’t get vaccinated, around 25 will be admitted to an ICU

The admissions to ICUs will be one or two, if not fewer, instead of 25.

If 100,000 people don’t get vaccinated, around 60 will die

If 100,000 people do get vaccinated, around 3 will die

57 people

(out of 100,000)

will avoid death

One serious

case of

thrombosis

From a sample of 100,000 people vaccinated, over the space of four months one person will develop thrombosis

But around 20 ICU cases and between 50 and 60 deaths will have been avoided

Note: In this scenario, we have assumed a 3.8% infection rate in Spain – which is the rate registered between January and now; also that the vaccine is 75% effective.

100,000

people

Without vaccination

With vaccination

If 100,000 people don’t get vaccinated, around 3,800 people will become infected

If 100,000 people do get vaccinated, around 950 people will become infected

Fewer infections

If 100,000 people get vaccinated, around 2,850 people will avoid catching Covid-19 in this example. And that’s without taking into account the fact that getting vaccinated will also put the brakes on transmission.

If 100,000 people don’t get vaccinated, around 250 will end up in hospital

If 100,000 people do get vaccinated, around 13 will end up in hospital

Far fewer hospital admissions: the vaccine will prevent 237 hospitalizations (19 out of 20 or more)

If 100,000 people do get vaccinated, around 2 will be admitted to an ICU

If 100,000 people don’t get vaccinated, around 25 will be admitted to an ICU

The admissions to ICUs will be one or two, if not fewer, instead of 25.

If 100,000 people do get vaccinated, around 3 will die

If 100,000 people don’t get vaccinated, around 60 will die

57 people (out of 100,000)

will avoid death

One serious

case of

thrombosis

From a sample of 100,000 people vaccinated, over the space of four months one person will develop thrombosis

But around 20 ICU cases and between 50 and 60 deaths will have been avoided

Note: In this scenario, we have assumed a 3.8% infection rate in Spain – which is the rate registered between January and now; also that the vaccine is 75% effective.

100,000

people

Without vaccination

With vaccination

If 100,000 people don’t get vaccinated, around 3,800 people will become infected

If 100,000 people do get vaccinated, around 950 people will become infected

2,850 (out of 100,000)

will avoid becoming

infected

Fewer infections

If 100,000 people get vaccinated, around 2,850 people will avoid catching Covid-19 in this example. And that’s without taking into account the fact that getting vaccinated will also put the brakes on transmission.

Note: In this scenario, we have assumed a 3.8% infection rate in Spain – which is the rate registered between January and now; also that the vaccine is 75% effective.

If 100,000 people do get vaccinated, around 13 will end up in hospital

If 100,000 people don’t get vaccinated, around 250 will end up in hospital

237 (out of 100,000) will

avoid hospitalization

Far fewer hospital admissions: the vaccine will prevent 237 hospitalizations (19 out of 20 or more)

If 100,000 people don’t get vaccinated, around 25 will be admitted to an ICU

If 100,000 people do get vaccinated, around 2 will be admitted to an ICU

The admissions to ICUs will be one or two, if not fewer, instead of 25.

If 100,000 people do get vaccinated, around 3 will die

If 100,000 people don’t get vaccinated, around 60 will die

57 people (out of 100,000) will avoid death

One serious

case of

thrombosis

From a sample of 100,000 people vaccinated, over the space of four months one person will develop thrombosis

But around 20 ICU cases and between 50 and 60 deaths will have been avoided

These approximate calculations are also similar to estimates from the British regulatory agency MHRA. They make it very clear that, irrespective of age, the benefits of the AstraZeneca and Janssen vaccines far outweigh their known risks.

In a scenario such as the above, in which the infection rate is 3.8% over several months, vaccination prevents 50 to 60 deaths and 200 hospital admissions per 100,000 people while triggering just one serious and potentially lethal thrombosis, according to the European Medicines Agency (EMA).

However, the analysis should be carried out according to age as Covid-19 is not as dangerous at 20 as it is at 70. This is why the vaccination campaign was initiated in nursing homes, because the disease is 15 times more deadly at age 55 than at 25.

The following graphic flags up these differences, showing the number of people hospitalized, admitted to intensive care or killed by Covid-19 per 100,000 in each age bracket, assuming, again, that the probability of becoming infected is 3.8% over a given period:

Benefits and risks of the vaccine

Assumption: a 3.8% probability of becoming infected with Covid-19 at any given moment (without a vaccine).

Without vaccination

AgeAdmissions
(per 100.000)
ICU
(per 100.000)
Deaths
(per 100.000)
20-293120.4
30-396140.9
40-4910392
50-591902410
60-693706245
70-7964886156
80+92816465

With vaccination

AgeAdmissions
(per 100.000)
ICU
(per 100.000)
Deaths
(per 100.000)
Thrombosis
(per 100.000)
20-291.60.10,02
30-3930.20,01.6
40-4950.50.11
50-5991.20.50.8
60-6919320.4
70-7932480.4
80+460.8230.4

For the elderly, the difference between being vaccinated and not being vaccinated is massive. Among the over-80s, deaths with the vaccine are reduced from 500 to 25 per 100,000, with no cases of thrombosis expected. These differences are slightly reduced but remain significant in the 70s, 60s and 50s age brackets. For those in their 40s, the vaccine would prevent two or three deaths per 100,000 people, nine intensive care admissions and 100 hospitalizations, but would trigger thrombosis in one person, according to EMA estimates.

“The benefits of AstraZeneca’s Covid vaccine continue to far outweigh the risks for most people,” said the head of the British regulatory body, June Raine, referring to the estimates.

However, their calculations, like those made by EL PAÍS, do pose a dilemma for younger people. On account of this, the British agency proposes vaccinating this demographic with the Pfizer-BioNTech and Moderna vaccines, which have shown no similar side effects so far. The virus is much less dangerous for very young people and, if its circulation becomes minimized, the risk posed by Covid-19 for this group is equal to the risk, also low, posed by the thrombosis associated with the AstraZeneca and Janssen vaccines.

According to EL PAÍS’ approximate calculations regarding people in the 20-to-29 age bracket, the vaccine prevents a couple of admissions to intensive care and half a death per 100,000 people, but it is possible that it could cause one or two cases of lethal blood clotting, according to estimates being used by the different regulators.

It is difficult to know at this point the precise risk of thrombosis associated with the vaccines. The estimates are currently under investigation with the possibility that they will be different based on age, gender or other characteristics. In the United Kingdom, authorities estimate a thrombosis risk of approximately one in 250,000 with the AstraZeneca vaccine. They also believe that the risk is twice as high in younger people as in older people, and, although they have found more cases in women, they have not yet published estimates according to gender. The EMA estimates the risk of thrombosis to be one in 100,000, without yet linking that risk to gender or age. In Denmark, they estimate the risk to be one in 40,000.

This is in connection with the AstraZeneca vaccine. However, the US Federal Drug Administration (FDA) has just released a report on the Janssen vaccine, which uses the same delivery mechanism as AstraZeneca. The FDA has found six cases of thrombosis, almost all in young women, among the millions who have been vaccinated, so the incidence in that group could be one in 100,000. But, as US epidemiologist Caitlin Rivers explains, more cases could come to light. This uncertainty explains why regulators are doing more studies: it is logical to be vigilant.

What is the risk of contracting Covid-19?

The figures above outline a scenario in which the rate of future infections mirrors the rate recorded over the past four months: 3.8% of the population, or one in 25. But this is only a point of reference. Obviously, the benefit of being vaccinated increases the more at risk you are of infection; if, say, the pandemic lasts for another year.

It is useful to use the 100% case as an example as it shows how the risks and benefits compare if we assume that you will definitely end up getting infected at some point in the future. This graphic charts the risks and benefits in a 100% scenario according to age:

Analysis of risks and benefits

Assumption: a 100% probability of becoming infected with Covid-19 at any given moment (without a vaccine).

Without vaccination

AgeAdmissions
(per 100.000)
ICU
(per 100.000)
Deaths
(per 100.000)
20-298244310
30-391,60911022
40-492,72124965
50-594,989643276
60-699,7481,6251,175
70-7917,0452,2604,107
80+24,41242012,230

With vaccination

AgeAdmissions
(per 100.000)
ICU
(per 100.000)
Deaths
(per 100.000)
Thrombosis
(per 100.000)
20-294120.52
30-398051.11.6
40-491361231
50-5924932140.8
60-6948781590.4
70-798521132050.4
80+1,221216110.4

In a 100% scenario, the advantages of the vaccine are multiplied. For those over 80, 10,000 deaths would be avoided. And the benefit would be even clearer for younger people: for a young infected person, the risk of ending up in intensive care would be 40 in 100,000, while the risk of suffering a thrombosis would be more like two in 100,000.

The probability of becoming infected in the next few months is unknown. On the one hand, this is because it depends on the individual. But it is also because it is a future prediction. How can we know what the incidence of cases will be in Spain or in any other country next month or in six months’ time? Or in two years’ time, for that matter? To explore these possibilities, we have prepared a small interactive graphic of risk and benefit estimates according to different age brackets and the probability of infection.

The biotechnologist Lluís Montoliu finds it concerning that the focus has been placed on the exceptional adverse effects of the vaccines while hundreds have been dying every day from Covid-19. “We lack the understanding that drugs are not harmless, that they all carry a certain risk,” says Montoliu, president of the Spanish National Research Council’s Ethics Committee.

The oral polio vaccine, containing a weakened virus, is a case in point. After giving it to three billion children worldwide, the World Health Organization (WHO) recorded 760 cases of polio caused by the vaccine. But the institution estimates that the vaccination campaign as a whole prevented 13 million cases. The disease is now on the point of being eradicated from the planet.

The situation could be compared to the trolley problem – the famous thought experiment devised in 1967 by British philosopher Philippa Foot: a passerby sees that a streetcar is going to run over five people, but can divert it to a side track where it will only run over one person. With the AstraZeneca vaccine, the streetcar – in this case, Covid-19 – will kill 50 or 60 people and leave another 20 in intensive care, but with vaccination, it is possible to switch the streetcar to another track and cause only one potentially lethal case of thrombosis.

But, according to Federico de Montalvo, a lawyer and president of the Spanish Bioethics Committee, “vaccination is not done directly for the good of the community, it is done for the good of the individual.” De Montalvo recalls a January ruling by a court in Santiago de Compostela, in the northwestern region of Galicia, where a judge authorized the vaccination of an elderly woman in a nursing home to avoid her individual risk, not the collective one.

“The situation is more like the issue of colonoscopies in colorectal cancer screening,” says Montalvo, who heads the Spanish government’s main advisory body in the field of scientific ethics. “A healthy person could get a perforation in the bowel and die while having a routine colonoscopy to rule out a tumor. One of these very rare adverse episodes can occur, but how many deaths from cancer do you avoid?” Vaccination, he stresses, not only prevents serious cases of Covid-19, it also reduces the asymptomatic transmission of the virus within society. “The vaccine addresses the collective interest, but we do not expose ourselves to risk for the sake of others, but first and foremost for ourselves,” he says.

The vaccine minimizes or eliminates the individual risk of dying from Covid-19, but there is another benefit: the social one. The estimates here are individual, but another advantage of vaccination is that it reduces the likelihood of infecting others. When someone gets vaccinated, they are doing their bit to protect everyone, literally.

All the figures in this article are approximate – the precise calculations will be those of the regulatory agencies. But they are an exercise in understanding the kind of dilemmas we face and the kind of analyses that health authorities do.

Sources and methodology

Lethality. Virus lethality – deaths per infected person, or Infection Fatality Ratio (IFR) – was calculated from the data on deaths and age-confirmed cases from the Carlos III Health Institute. Cases were based on an extensive period during the second wave – from October to mid-February – and deaths within that same period with a three-week variation. Cases are based on the assumption of a 75% detection rate (the resulting IFR is, for example, 1.17% for people aged 60-69). If detection were worse at 60% or 65%, the case fatality ratio would be about 10% lower. The results are similar to those obtained in this study based on serology research and the confirmed and excess deaths during the first wave.

Risk of Covid hospitalization. To calculate the risk of hospital and intensive care unit (ICU) admissions, we calculated the admission rates per death, also using data from the Carlos III Health Institute. The cases are based on the period between October and mid-February; admissions are based on the same period with a difference of one week and ICU admissions with a difference of two weeks.

Vaccine effectiveness. The calculations of deaths, hospitalizations, ICU admissions and infection rates in vaccinated people suggest an effectiveness of 75% against transmission and 95% regarding admissions and deaths. This is an estimate based on a bibliography shared with EL PAÍS by the Computational Biology group from the Polytechnic University of Catalonia (UPC). It is a conservative estimate that does not take into account the additional reduction of transmission achieved by having a vaccinated population.

Thrombosis risk. The figure for the risk of serious blood clots with AstraZeneca is based on estimates from the European Medicines Agency (EMA), which calculates one case of thrombosis in 100,000 vaccine recipients. The distribution of these cases by age was made using information from the British regulator MHRA.

Transmission probability. The central scenario used at the beginning of the article assumes that 3.8% of Spaniards will be infected in the coming months. This is an indicative figure that is useful as a reference. It is the same number of people who have been infected since January until now, according to data from the Carlos III Health Institute, again based on the assumption that 75% of real cases are detected.

For the initial calculations, we have taken 100,000 people across all age brackets based on the population structure as registered by the Spanish Institute of Statistics (INE). According to the INE structure, 20-29-year-old make up 13%s, 40-49-year-olds 21% and the over-80s 8% of the 100,000.

Acknowledgments: Miguel Ángel Martínez Beneito, statistician at the University of Valencia; Jesús Sierra, pharmacist from the Spanish Society of Hospital Pharmacy; and professionals from the National Epidemiology Center.

English version by Heather Galloway.

More information