Also it is very likely that the people with most comorbidities where vaccinated earlier, so have had the longest period of post-vaccination. So we are likely l

Also it is very likely that the people with most comorbidities where vaccinated earlier, so have had the longest period of post-vaccination. So we are likely looking at a sub-group that both has multiple comorbidities _and_ has had the longest period of post-vacination. Even assuming that the decay of effectivenesss is small over time it may be a contributor.