The article is literally titled "Association Between Slow-Wave Sleep Loss and Incident Dementia". Association is not causation.