[erdo]

These stats are only useful if we know how many people get dementia in the first place.

What are the chances of a random control group of 14 over 85 year olds also not getting dementia?

If it's very common, and you would usually expect 7 of them to develop dementia, then 0 cases in the test group is potentially impressive. If it's quite rare and you would usually expect only one of the control group to get dementia, then it's not that impressive that the test group had 0, and easily down to chance

[Epenthesis]

That's what the first image in the article is purporting to show (~33% of people in the general population over 85 have dementia)

[erdo]

Ha of course it does, thanks :) I don't know why I didn't see that

so we could expect about 5 people with dementia in that control group, and the test group got 0, so not bad