There is no diagnosis in the group that doesn't have dementia, there is no "after flu vaccine" in people who dont get the flu vaccine. Its called selection bia

There is no diagnosis in the group that doesn't have dementia, there is no "after flu vaccine" in people who dont get the flu vaccine. Its called selection bias. The root cause of the measurement issue is that people are less likely to get the flu vaccine if they have dementia. People who had the flu vaccine are already much less likely to suffer cognitive decline even if the flu vaccine was a placebo. This sort of stuff is why double blind RCTs exist and have millions spent on them. Also why all the dementia treatments to date failed to work, because these issues are so often ignored. Also perversely - mental assessments are typically carried out on people who doctors "dont" think have dementia (to rule it out as a cause of symptoms) Consider this "back of the envelope" calculation. 10% of the population will develop dementia. having early dementia symptoms makes you 50% less likely to get the flu vaccine. 70% of people get the flu vaccine. If you randomly selected 200 who didnt get the vaccine and 200 who did from this group, you would expect to see only 13 of the 200 develop dementia when vaccinated, but 53/200 in the unvaccinated group.