Are the covid-19 vaccines more protective against severe disease and death than against simple infection?

Question

I have frequently heard variations of the following claim: "The vaccination reduces your chances of getting covid, and even if you do get it, it is much less likely to be severe." I can't find the source, but I recall a quote of someone saying that, if they tested positive after being vaccinated, they wouldn't be worried because the vaccine would reduce their chances of getting severe disease.

But, as I read the statistics, they seem to call this claim into question. Of course, different vaccines have different efficacies, but in general, the values seem to be in the same ballpark for the merely symptomatic and the severe categories.

I want to make two important points: First all the evidence indicates that the vaccines are very effective against all levels of severity. The statistics from Michigan below are hugely positive for the vaccine. There were less than three hundred cases out of 1.7 million vaccinated people, at a time when you would have expected many thousands if they weren't vaccinated. Being vaccinated makes the risk low enough that you can basically ignore it for most purposes. This question is about differences in effectiveness for more severe outcomes.

Second, the efficacy indicates how much the vaccine reduces a risk. So, if a vaccine has an efficacy of 95% against both documented infection and death, then it makes the chances of both 20 times smaller than they were, which is great.

All of that said, here are examples of what I mean:

• The Clalit study in Israel reports 92% efficacy for the Pfizer/BioNTech vaccine against both documented and severe infection. There were no deaths in the vaccinated group, but there were so few deaths in the control group that no real conclusions can be drawn.
• Another study in Israel found 92% efficacy against documented infection and approximately 97% efficacy against more serious outcomes. So, based on this study, the efficacy is almost the same against any symptomatic illness as it is against death.
• More informally, a report from Michigan indicates that there were 246 fully vaccinated residents who got covid, of whom 3 died. Both of these numbers are tiny compared to the vaccinated population, but of those rare people who did get sick, more than 1% died, similar to the general pattern. Also the accepted answer to this question suggests a roughly similar fatality rate for the country as a whole.

There are somewhat countervailing studies:

• The original Moderna study had no severe cases in the treatment group (but the number in the control group was only 30).
• Different sources report efficacy values for the Janssen/J&J vaccine of 60-75% for infection and values in the 80's for severe infection, a significant difference.
• This study from Qatar is maybe the most supportive of this idea. They find efficacies of 75-90% (depending on the variant) against documented infection, but 97.4% against severe disease.

Taking all the data together, can a fully vaccinated person with a symptomatic case of covid be reassured that their chances of death (given that they have a symptomatic case) are much less than if they had not been vaccinated? For instance, is there a strong argument for supporting the Qatar study over the Israel studies? Or are there reasons for thinking that the statistics on deaths among fully vaccinated people are misleading? (If most or all of the deaths are from cases that were acquired before full vaccination, that would be significant.)

I am also interested in the same question for severe illness as an outcome, which I should have made clear; this paragraph is being added in an edit.

Answer 1

Apparently, you are asking whether CFR (i.e. chance of dying given a confirmed infection) is lower in the vaccinated group. Note that chance of getting a [symptomatic] infection in vaccinated groups is much lower, so your question isn't capturing the whole effect of vaccines.

But nonetheless, Public Health England has published some such data in Feb 2021 relative to the previous month.

[CFR] estimates exhibit large falls in recent weeks, particularly in the 80+ population that was first prioritised for vaccination.

As I noted in my comments, it's also pointed out in this study that one needs to be careful with interpreting such observational studies. This one was possible because at the time there were different rates of vaccination among different age groups, the over 80 were vaccinated in [much] greater numbers, and [in the graphs above] the CFR reduction was noted among them:

While the 80+ population was prioritised for vaccination and was vaccinated in much greater proportion as a result, substantial numbers of the under 80 population have also been vaccinated. While this would tend to dampen down our ability to detect any impact, the much greater extent of vaccination among the 80+ population should still allow the detection of such an effect: as of 7 January, just over one month ago, official Public Health England figures show around 28% of the 80+ population to have been vaccinated compared to only 3% of the 75-79 population. Furthermore, smaller numbers of deaths in younger age groups may make more difficult the accurate estimation of the CFR for younger such groups.

Nonetheless, there was also other unexplained variation in CFR among other age groups [in those graphs] so one cannot be totally certain of the causality.

If, on the other hand, you want to consider the combined effect of not getting a symptomatic case plus death, vaccines (even the Chinese ones) do much better, as e.g. shown in this study/data in Indonesia:

(table not formatted as block quote as that seems to mess up the table.)

The number of Covid-19 deaths among health workers in the study

Groups	Deaths	Total	%
Not vaccinated	17	28,055	0.66
Vaccinated with the first dose	3	8,458	0.03
Vaccinated with the second dose	1	91,777	0.001

Source: Health Ministry

As for the number of cases:

The number of Covid-19 infection among health workers in the study

Groups	Infected	Total	%
Not vaccinated	2,431	28,055	8.66
Vaccinated with the first dose	657	8,458	7.76
Vaccinated with the second dose	521	91,777	0.56

Source: Health Ministry

If you want to do a quick CFR estimate here: for those vaccinated with both doses 1/521 = 0.19% vs no vaccine 17/2,431 = 0.7% doesn't seem that much of an improvement (although there is some), but deaths relative to the treatment/control group size (0.001% vs 0.66%) certainly is a much more noticeable improvement for the vaccine group.