Date published: 2008/02/22
The BBC says:
Regular unintentional daytime dozing may be an early warning sign of stroke in elderly people, say US researchers.
For those who had a habit of nodding off, the risk of stroke was two to four times higher than for those who never fell asleep in the day, a study found.
Speaking at the International Stroke Conference, the team advised doctors to check out older people who found they were dropping off in front of the TV.
The study asked 2,000 people how often they dozed off in different situations.
These included while watching TV, sitting and talking to someone, sitting quietly after a lunch without alcohol and stopping briefly in traffic while driving.
The risk of stroke over the next two years was 2.6 times greater for people who reported "some dozing" compared to those with no dozing.
Among those who reported "significant dozing" the risk was 4.5 times higher.
The researchers also found the risk of heart attack or death from vascular disease was increased.
Surprise, people who are extremely tired are not as healthy as those who are not. Who would have thought it. But what is most interesting about this article is not this rather trivial observation (but at least it comes with some quantification) but rather that for once we are not presented with a confusion of correlation and causation.
So normally, in these kinds of cases, we would then be told that it is the sleep that is the implied cause of the ill health, and that if only these people slept less then they would not be so badly off. So it's curious why that incorrect argument is not made here.
Well, usually when this kind of incorrect deduction is made, it is because the action under consideration is considered wicked by the academic middle class. So things like smoking, drinking, eating too much, etc. Thus the natural inclination is for the academic middle class (chief cheerleader, the BBC) to say that the bad consequence is all just the fault of the people involved, if only they would change their wicked ways.
Here, nobody is going to want to blame the old dears for their "unintentional" dozing off. So we are not presented with any of this nonsense.
Although causation is a much more powerful conclusion than correlation, this specific article illustrates that correlations can indeed be useful, because here it might lead to early diagnosis of possible strokes (or whatever). One downside is that any time some poor old dear nods off, some over-anxious relative might whisk them off to hospital just in case.
Another way of looking at this is that the article does not give any details to allow one to estimate how many false positives (versus actual positives) there would be in the situation being discussed, and that really determines how useful this correlation will be in practise. There is no real point worrying about someone dozing off if (say) 99 times out of 100 this is a false positive, rather than an actual positive, indication of a future stroke.
_________________________________________________________
All material not included from other sources is copyright cambridge2000.com.
For further information or questions email: info [at] cambridge2000 [dot] com
(replace "[at]" with "@" and "[dot]" with ".").