I agree with you for a small sample size but when dealing with a large sample, I think that those things (age and BMI) would tend to average out.
Here is a similar thing for Ontario as New Zealand... https://covid-19.ontario.ca/data
Scroll down to this section (near the top) “COVID-19 cases by vaccination status”.
The entry for today says that this...
- Not fully vaccinated 13.89 per 100,000
- Fully vaccinated 16.64 per 100,000
- Fully vaccinated with booster 24.28 per 100,000
If you read the full site, you’ll find that ‘Not fully vaccinated’ is essentially synonymous with ‘Not vaccinated’.... they changed the description a few weeks ago.
Regardless, what this means is that it is fair to compare the first category (not fully vaccinated) with the sum of the other two to say that “folks who are fully vaccinated including those who have had boosters are 2.96 x as likely to be a covid case as those who are unvaccinated.”
Only one way to be sure and that is to find data where it is broken out. I’d like to see other co-morbidities as well included. Because likely the people who are now dying of covid likely have more confounding factors.
But I’m done with covid. Not an expert. Happy to never discuss it again.