QUOTE(Malleus @ Tue 10th November 2009, 7:01pm)
QUOTE(Milton Roe @ Wed 11th November 2009, 12:59am)
QUOTE(Malleus @ Tue 10th November 2009, 5:36pm)
I suggest Milton that you take some time out to at least peruse Thomas Szatz's book, because you're quite clearly completely unaware of the anti-psychiatry movement. You could always start at
The Myth of Mental Illness.
I'm well aware of the book and read it years ago. Szasz reminds me of those people who refuse to believe that HIV causes AIDS or that cigarettes cause cancer. How do you PROVE it to them? Turns out you can't. It's almost impossible to prove a causal relation to somebody who refuses to believe it, if you can't do the demo where you deliberately cause the effect.
A causal relationship can be demonstrated to be statistically significant rather easily, as I'm sure you know. What's rather more difficult to prove to any individual is that their own smoking will shorten their life, as it's only a probability after all, not a certainty; "My mother lived to be 98, and she smoked 20 cigarettes every day of her life."
Actually, it's worse than that. Forget the grandmother and her n = 1 contribution. How does one demonstrate that smoking causes cancer, with mere statistical evaluation of people who chose to smoke vs. people who don't? Turns out that you can't do it. Even if one group has more disease than the other, past all reasonable possibility of chance association, this only suggests some kind of causality, but not necessarily the kind you're interested in. For example, if A correlates with B, it may be that A and B are caused by C. In the case of smoking, people who choose to smoke are NOT THE SAME POPULATION as people who don't. So they're self-selected, and different to begin with. You can't control for such differences unless you design a randomized controlled trial in which people are randomized to smoke or not, and then this is enforced to make sure that those assigned to smoke do so, and those who don't, don't. You'll never see such a study. The closest that thas ever been done is several smoking intervention trials in which people were assigned at random to heavier or lighter pressure to stop smoking. Alas, no intervention trial is perfect and many people will be unable to quit. And again there's every reason to suppose that those who try to quit but cannot, are not the same people as those who try and succeed. So they undoubtedly differ in other ways from the smokers, than just smoking.
If you think such factors aren't important, you should take a look at the epidemiology for vitamin E supplementation and the statistics behind hormone replacement for post-menopausal women. In each case, the evidence that vitamin E supplementation and hormone replace were beneficial, was massive. But this was all post-hoc analysis, which could not control for the fact that pill-popping women tend to take care of themselves better than woman who don't bother, IN GENERAL. So they're not the same groups to begin with, even disregarding the hormone/nutrient effects.
Finally, the biomed establishment was forced to run some double-blind prospective randomized placebo controlled trials of vitamin E. And also (different trials) hormone replacement. These resulted in massive failures. Very few of the good effects suggested by the epidemiology were seen. So the differences seen in retrospective statistics were NOT due to the molecules, but were INDEED self-selection bias. Wups. Now-- prove that's not happening for cigarettes, but in the other direction.
Don't expect any help from animal models-- if you're looking for lung cancer or heart disease produced by smoking in animals, forget it. It's never been seen. (IMG:
smilys0b23ax56/default/tongue.gif)