Vaccines to reduce the risk of dementia.
The new study linking flu vaccine to lower #Alzheimer’s risk avoids these traps. It’s a matched study, meaning they took data from nearly 1 million people vaccinated against flu and compared it with nearly 1 million other unvaccinated people, matching them by age, race, gender and where they lived. They also accounted for medication usage and comorbidities to ensure that, for example, they didn’t have a disproportionate number of people with diabetes or high blood pressure (both risk factors for dementia) in either group.
Looking back at 10 years (from 2009 to 2019) of health insurance claims data from these pairs of people aged over 65 years, they were able to study the impact of flu #vaccination on subsequent Alzheimer’s diagnosis. And the results were clear: flu #vaccines cut your risk of #dementia. In fact, the strength of the protective effect increased with the number of years the person received an annual flu vaccine.
This is not entirely shocking, but it is a very strong conclusion to a very large study. An earlier paper by scientists in Taiwan, looking at the risk of dementia among a smaller number of army veterans, also found that vaccinated people reduce their risk of cognitive decline. https://www.vaccinestoday.eu/stories/could-flu-vaccines-reduce-alzheimers-risk/
The way that study's described it seems that it showed us that people who would choose to get vaccinated against flu are less likely to suffer from Alzheimer later on. So, maybe recommending flu vaccinations would cause the same effect, or maybe it's caused by something upstream of the vaccination that wouldn't be affected by recommending flu vaccinations alone.
Sad thing about controlling for confounders is that it's not the case that adding more controlled confounders makes the result more likely to be correct: sometimes one can _introduce_ a confounder that way. Consider e.g. a hypothetical case where people who are more diligent will both be vaccinated against flu *and* will be more likely to have their ailments discovered. In that case correcting for the confounder of "ailments that have been discovered" actually creates a fake correlation between flu vaccinations and lack of ailments.
