[somewhereoutth]

My understanding is that disease induced immunity is a bit more unreliable compared to vaccine induced immunity - perhaps antibodies may be generated that target a part of the virus that is not highly conserved or important for infection, so leaving the individual vulnerable to slight variations, whereas vaccines produce highly targeted antibodies. Also, the virus contains components that interfere with the immune response, perhaps degrading the immune memory?

Vaccination after infection does appear to provide excellent protection though.

We are still doing the science on all this of course.

[esyir]

I'd expect the opposite, and from what I understand, this is particularly so for delta. Consistency is an issue though, as natural immunity is more variable in it's response.

> https://www.science.org/content/article/having-sars-cov-2-on...

This report on an Israeli study seems to imply that natural immunity was superior to the vaccine for protection against delta, though a combination of vaccination and infection provided the best response. This, to me is intuitive.

Note: 1, educated conjecture ahead

It seems intuitive to me that the natural immune response would provide greater protection against variants like delta, stemming from the nature of their targets. The vaccine is highly tuned for a specific target: the spike protein. Conversely, natural immunity performs multiple "training" runs in parallel, targeting a wider variety of antigens. If you'd take a ML perspective, this is somewhat analogous to an overfit model vs a more generalized model.

Note 2: Alas, you still have to get Covid to begin with to get natural immunity, so you probably don't want to go out of your way to get it if you haven't already.

[ceejayoz]

CDC currently says otherwise.

https://www.cdc.gov/media/releases/2021/s0806-vaccination-pr...

> In today’s MMWR, a study of COVID-19 infections in Kentucky among people who were previously infected with SAR-CoV-2 shows that unvaccinated individuals are more than twice as likely to be reinfected with COVID-19 than those who were fully vaccinated after initially contracting the virus. These data further indicate that COVID-19 vaccines offer better protection than natural immunity alone and that vaccines, even after prior infection, help prevent reinfections.

[esyir]

That CDC article is likely outdated. In particular, refer to this line in the cited paper:

>Second, the study period for this analysis occurred before the predominance of the B.1.617.2 (Delta) variant;

In the Israeli study, they specifically were observing efficacy against the delta variant, which is the dominant (or will be dominant everywhere eventually). This is notable, because Delta, while not fully escaped, has shown some degree of resistance to the current vaccines.

As such, what I think is the most correct interpretation, given the current information (and is in line with my intuition) is that yes, the vaccines provided greater protection against initial strains, but the relatively narrow target means that they provide a less robust response against later divergence as compared to the more robust natural immunity.