Hacker News Reader

That's not more accurately at all; you're still insinuating a causal relationship.

"More accurately, people with yellow-stained fingertips are more likely to get lung cancer."

The actual study's own title puts it more accurately:

"Association of sleep duration in middle and old age with incidence of dementia"

I don't think it insinuates a causal relationship.

I interpret "30% more likely" as the same as "associated with a 30% increase," which is what the article claims.

I would say that "People who do X are 30% more likely to get X" is commonly understood to imply causation.

How would you phrase it? Because I did not read that as meaning that less sleep caused dementia.

"is associated with" is a much more neutral phrasing. The phrasing did not denote causality but does imply it to a casual reader.