Sunday, June 05, 2005

Ayn Rand

I've read a few of her books. I'm not so sure she was the bastion of rationality she claimed, but I do think that the individual should be favored over the collective, if only because I don't want the collective telling me what to do. On the other hand, I don't want Ayn Rand telling me what to do either.

[ It's Rational! ]

Really, though, I am just farting around with HTML code, and wanted to try this out. I don't know diddly about web programming, but I am starting to fiddle around. I program in C++ or LabView when I have some project I need to do. I do a little C programming for embedded controlers. But there are a lot of cool things one can do with web pages. I am especially interested in Java script, which will let you do all kinds of stuff. But I find it difficult, as yet, to get things to work.

Wednesday, June 01, 2005

Will phthalates shrivel your nuggets?

New Scientist has a story with the shocking headline " 'Gender-bending' chemicals found to 'feminise' boys ". In it, the implication is made that a new study shows that chemicals called phthalates are harmful to the sexual development of infant boys. In contrast to the sensational headline, the article body uses words that are slightly more weasely: it says the results "suggest" adverse effects of phthalates on genital development in baby boys. This sounds serious. Let's look at it.

Phthalates are common plasticizers used in cosmetics,
plastics, food containers (including pop bottles), infant toys etc. This is a chemical to which people have been and are routinely being exposed . Despite years of concerns being raised, though, these substances have never been shown to be a hazard.

So, do phthalates shrink testicles, or more accurately, cause genital defects? Well, if they do, you can't tell it from the study to which the New Scientist article refers. Stats.org at George Mason University points out that the study did not even consider this hypothesis.

Are phthalates "gender bending" like the headline says? High doses cause problems in rats, known as incomplete virilization. In extreme cases, the genitalia can be ambiguous, not clearly male or female. So far as I know, no data exists for humans at high exposures. Phthalates haven't been shown to have effects in human until this study, says the New Scientist article. So even if this study could be counted as evidence for this, which it can not, it's only one study.

I'm not an expert in this field, but I can apply some general scientific knowledge to evaluate the article . If the study didn't test a hypothesis, then it isn't appropriate to interpret it as lending evidence to that hypothesis. So if the study shows anything at all, it is not showing that phthalates affect genital genital development in humans. This wasn't even considered.

And if you read the Stats.org description of the study results and their statistical significance, you'll see that there was only weak correlation between the phthalate metabolite concentration in mother's urine and the calculated anogenital index, a ratio defined by the distance between the anus and the base of the penis divided by the weight of the child at the time of measurement. Other ratios, like the anoscrotal index, didn't show a correlation. Whatever the study did show, it showed poorly. So poorly that no conclusion can be drawn; none of the phthalate/ anogenital index correlations passed the test for statistical significance. One can't conclude that any of the phthalates had any significant effect based on the meaurements made. The study results don't suggest that phthalates cause genital problems or otherwise feminize boys, and it is incorrect and misleading to suggest otherwise.

I am completely sympathetic with anyone who wants to avoid exposure to chemicals, and understand why they would want to protect their children from them. The term "chemical" is intimidating, especially in light of what the term can mean. "Chemical" can mean things that are corrosive, caustic, foul-smelling and toxic. Some things chemical are clearly dangerous. Some are dangerous while looking benign. And some, a good many useful things, are innocuous when they seem threatening. We really only know by doing the science. We should not ignore data that shows that something is harmful, but we also need to recognize when the data says we have something that is useful and not harmful. It isn't helpful to trumpet non-results in a frightening and confusing manner. It is deceitful to conflate detecting something, and that something being automatically dangerous or even significant.

Is it useful to continue studying phthalate saftety? I think that it is possible that someone might have a genetic predisposition to sensitivity to phthalates, and I would like to see this studied. I don't know of a way that one could prove unequivocally that a substance is not harmful to anyone under every reasonable set of conditions. But I think that lack of proof of harm, in spite of vigorous efforts to find it by people completely free of chemical industry influence, means we can go on using things containing phthalates without fear.