[mlyle]

E.g. in Israel, B.1.351 represented ~5% of dual dose vaccine breakthrough events when it was at a prevalence of less than 1% in the population. This means that B.1.351 is better at causing infection of people with the vaccine than the previously prevalent variants.

This is despite people with the vaccine having higher neutralizing antibody titers against B.1.351 than people naturally infected with previous strains.

Therefore it seems very likely to expect that B.1.351 will also cause reinfection at higher rates than we've been used to, since it appears that people with the vaccine are more protected than people naturally infected. This is especially true if people who were infected earliest have their immune responses start to wane.

Again: I don't think this is a huge absolute risk. I just think it's significant enough to be a significant confound for those of us trying to guess what Rt would be with a return to baseline behavior, and enough for people who've "had it" to be less protected than they might expect.