4 ms·

http://www.usatoday.com/news/nation/2008-03-10-drugs-tap-water_N.htm http://www.usatoday.com/news/nation/2008-03-10-drugs-tap-wat... """ Troubled by drugs disc

http://www.usatoday.com/news/nation/2008-03-10-drugs-tap-water_N.htm http://www.usatoday.com/news/nation/2008-03-10-drugs-tap-wat... """ Troubled by drugs discovered in European waters, poisons expert and biologist Francesco Pomati set up an experiment: He exposed developing human kidney cells to a mixture of 13 drugs at levels mimicking those found in Italian rivers. There were drugs to fight high cholesterol and blood pressure, seizures and depression, pain and infection, and cancer, all in tiny amounts. The result: The pharmaceutical blend slowed cell growth by up to a third suggesting that scant amounts may exert powerful effects, said Pomati, who works at the University of New South Wales in Sydney, Australia. """

jerf 15y ago
Unfortunately (and I mean that, I wish this style worked better than it does because it is so easy compared to real experiments), that style of experiment is well known for producing useless results. Think about it; if the water was that poisonous, such that 1/3 of our celluar growth was being eliminated, we wouldn't be speculating about whether our water was killing us, we'd know and long since have taken some sort of action. The strength of the real signal is bounded by the fact that we have millions upon millions upon millions of real people consuming these things every day. Reality always trumps both theory and experiment, and don't ever listen to anybody who forgets that. An experiment that shows an enormous signal, which we can observe in the real world can not possibly be occurring, is far more likely to be a flawed-beyond-usefulness experiment than a real reason for concern. It is, however, a great way to get in the news. You get the same problem when some people discover that X causes cancer, except that if you take their results seriously you end up with (metaphorically) 245% of the population dying of cancer X by the time they are 30. Except we don't. Hidden dangers can only be so dangerous.

jerf 15y ago

Unfortunately (and I mean that, I wish this style worked better than it does because it is so easy compared to real experiments), that style of experiment is well known for producing useless results. Think about it; if the water was that poisonous, such that 1/3 of our celluar growth was being eliminated, we wouldn't be speculating about whether our water was killing us, we'd know and long since have taken some sort of action. The strength of the real signal is bounded by the fact that we have millions upon millions upon millions of real people consuming these things every day. Reality always trumps both theory and experiment, and don't ever listen to anybody who forgets that. An experiment that shows an enormous signal, which we can observe in the real world can not possibly be occurring, is far more likely to be a flawed-beyond-usefulness experiment than a real reason for concern. It is, however, a great way to get in the news. You get the same problem when some people discover that X causes cancer, except that if you take their results seriously you end up with (metaphorically) 245% of the population dying of cancer X by the time they are 30. Except we don't. Hidden dangers can only be so dangerous.