Last week we learned that male factory workers exposed to large amounts of BPA, a chemical in some plastics, had abnormally high rates of erectile dysfunction and other sexual performance problems. This week the news is about phthalates (pronounced THAL-ates). Researchers reported in the International Journal of Andrology that this family of chemicals, used in manufacturing polyvinyl chloride plastics, seems to make little boys behave a bit more like little girls. This small study isn't as worrisome as the headlines suggest. Its main public-health value may be in spurring more pregnant women to avoid processed foods—a worthwhile choice anyway, for other reasons.
The finding hinges on the credibility of a questionnaire—the "Pre-School Activities Inventory" (PSAI), which mothers fill out in describing their child's behavior. It is a psychometric tool, developed in 1993 and considered the most scientific approach available for parsing out masculine boys from feminine boys. Shanna Swan, an epidemiologist at the University of Rochester, led the study, which is part of a series she's conducting on phthalates. Phthalates are plastic softeners found in food-processing plants (hence they're in your food) and in hospitals—and in your carpeting, your wallpaper, and until recently many of your children's toys. Toy companies have already started removing phthalates now that a new federal law goes into effect in February.
While toddlers may be mouthing less of phthalates from their toys these days, no regulations limit exposure in pregnant women, some of whom absorb phthalates from cosmetics as well as from the usual sources. Swan and her colleagues had the amount of phthalates measured in the urine of pregnant women they were studying. Then they examined these women's children to determine if there is any effect from prenatal phthalate exposure. In one of these studies, Swan reported that some phthalates were associated with a slightly shorter distance from the anus to the end of the penis in boys (a research measurement that until her study had been used only in rodents). She also found a small difference in the width (but not the length) of the penis with one phthalate.
Some industry-associated critics lambaste these results as irrelevant to human health and ignoring basic markers of genetic differences, including the parent's anatomy. Swan counters that the slight statistical anomalies she has found may portend more serious problems later on. After all, not so long ago, perfectly normal-looking girls were born to mothers who took DES (diethylstilbestrol) in pregnancy to prevent miscarriage, and the girls didn't start to develop rare vaginal cancers until years later.
While you can fault logic that implies we should outlaw phthalates now because of what we can't prove they might do years later, we are talking about our children. If there is any uncertainty, it seems moral to err on the side of safety even if the known effects at this point are completely benign. But industry has its own moral arguments: Phthalates have already been around for 75 years without any obvious detriment to human health, and their replacements would be compounds far less tested. Plus, there's the refrain that rings so true from 1950s advertising the industry still uses it today: Economical, functional plastics allow more people to live better lives.
Despite phthalates' usefulness, there's little getting around the fact that they interact with the human body, specifically by interfering with the sex hormone testosterone. Of course, humans are not isolated systems: We are riddled with microscopic organisms that produce all sorts of biologically active compounds; some of the all-natural and healthful plants we eat contain compounds that are "biosimilar" to human hormones. The question is, just how much biological give-and-take are you willing to have with your carpeting?
Besides its well-known effects on genitalia, testosterone concentration physically alters the size and shape of some brain regions before boys are born (girls do not produce significant estrogen prior to birth). We see the importance of this mental masculinization in a disorder called congenital adrenal hyperplasia (CAH). When a baby girl has CAH, a rare genetic disorder where the adrenal glands sitting atop the kidneys produce high levels of testosterone precursors, she may be born with ambiguous genitalia. Once the baby is determined to be genetically female, it's common to raise her that way (sometimes, surgery is necessary). While not a universal experience, CAH girls are more likely to be unsatisfied being girls and more likely to choose homosexuality or bisexuality. The testosterone their bodies produced before they were born, before doctors intervened to stop it, had already shaped their brain with a masculine imprint. Boys normally get this imprinting from months two through six in the womb. The fear is that phthalates may disrupt this process.
Now, back to that questionnaire. Many parents discover that preschool children are not too different from animals, and scientists can back up that observation with hard data. "Play behavior is different in male and female rodents, and in two species of monkeys there are studies where male monkeys chose wheeled car toys more than females," Swan says. "There appears to be a biological basis for these choices. If this was not measuring something objective, I don't think you'd see these differences over and over again."
Swan turned to the preschool activities inventory to measure these preferences in 74 boys and 71 girls, most of whom were around 5 years old, of the mothers whose urine had been tested for phthalates during pregnancy. The questionnaire asks mothers to rank their child as similar or dissimilar to a series of statements shown in large validation surveys to demonstrate degrees of masculinity or femininity. A mother encounters questions on the form like these: How often did the child play with jewelry in the past month? How often did the child fight in the past month? How often did the child show interest in snakes, spiders, and insects? (I find the questions about objects fascinating. Did you know that guns are slightly more boyish than swords, and that both are more masculine than trains, planes, and automobiles? Beyond those, "pretending to be a female character," as you might expect, can tell scientists a lot about your kid.)
While the study trumpets statistically significant results, the health relevance remains unproven, even if, as Swan says, the questionnaire "reflects changes in the brain." Even if a mother tested in the top 10 percent on phthalate concentration, the study projects that at most, her son will score 8 percent lower, or less masculine, than a boy whose mother wasn't exposed at all—and in some cases might not be affected at all.
Can these modest changes in play behavior, quantifiable only on such a survey, lead to any long-term changes in gender identity? "There are rare syndromes in which males are exposed prenatally to feminizing hormones, but gender identity is nevertheless profoundly affected by rearing and culture," says John Constantino, head of child psychiatry at Washington University School of Medicine in St. Louis. There are no studies demonstrating that outcomes on the preschool activities inventory carry on into adulthood.
There is good evidence that one's gender identity and sexual orientation are separate. Though CAH girls, exposed to large doses of testosterone in the womb, are more likely to be homosexual than other girls, the slight changes Swan and her colleagues found in this study are unlikely to survive all the later brain changes experiences, environment, and hormones imprint on the way to adulthood. It's disturbing that industrial chemicals influence our behavior at any age, and that alone may be enough cause to regulate them out of existence, but proving that phthalates cause lasting changes and ill health requires evidence that so far hasn't met the test.