Context: Around the world, slightly more boys are born than girls. This pattern is remarkably universal, with 51 to 54 boys born for each 50 girls born. Demographers use these data to calculate a population sex ratio, calculated as the number of male births divided by the total number of births.
Some demographers have reported small but significant changes in sex ratio in some countries, for example, in Canada and the US. Fewer boys are being born than expected. Several studies have reported striking local changes in sex ratio associated with chemical exposures, for example in Russian pesticide workers, babies born to men who were exposed to dioxin in the 1976 Seveso (Italy) accident, and babies born to men in Taiwan who were exposed to PCBs while less than 20-years old. All these cases involve fewer boys. Most recently, epidemiologists have observed a dramatic departure from normal in the sex ratio at birth of a First Nations community in SW Ontario. For the 5-year period beginning in 1999, fewer than 35% of live births were boys.
PCBs are a now-banned family of industrial chemical with many different applications, including use as coolants and insulating fluids in electronic equipment like transformers, plasticizers, flame retardants, caulking, wood floor finishes, etc. They are very stable and, through volatilization in warm air, become universally distributed around the world. Peak manufacturing occurred in the 1960s. They were banned beginning in 1973, for use in 'open' sources.' Production was banned in the US in 1977, although uses in capacitors and transformers continued. In 2001 their production was banned globally under the Stockholm Convention on Persistant Organic Pollutants.
PCB toxicity at relatively high doses are overtly toxic. At levels experienced by the general public they are associated with neurological and reproductive harm. Several public agencies, including the US EPA, consider them to be carcinogenic to humans. |
|
What did they do? The research team took advantage of the fact that a study of pregnant women and their babies (the Child Health and Development Study) had been carried out in the 1960s in which serum of was stored for later analysis. Conducted by San Francisco Bay area scientists, the study obtained data on about 20,000 pregnancies and has been a rich source of information about how conditions during pregnancy, including contamination, can affect health later in life (e.g., risk of breast cancer). People in the study were representative of a broad cross-section of Bay Area residents, although there were relatively few very poor or very wealthy people.
For this new analysis, the scientists focused on mothers who had stayed with the research program through an extensive examination when their child reached the age of five. There also had to be enough serum left to carry out chemical analyses for PCBs. Mothers were also left out of this assessment if they birth involved more than one baby (twins, etc.), the baby had severe birth defects, and on the basis of several other factors designed to make sure the assessment focused on normal births and babies. If mothers had had more than one child eligible on the basis of these criteria, only one was used in this study.
The serum had been frozen since obtained in the late 1960s at a facility of the National Institute of Health. Serum concentrations of PCBs were obtained for 399 women.
The sex of the babies had been recorded at birth. To explore the relationship between PCB levels and the likelihood of a boy vs. a girl baby, the scientists carried out a series of analysis, beginning with identification of factors that could confound, or confuse, the statistics. These included detailed information about drugs taken during pregnancy and information about how the samples had been stored.
In their analysis, they examined how the chance of giving birth to a boy changed at higher PCB levels compared to lower levels. In a crucial calculation they compared the sex ratio of babies born to women who's PCB levels fell in the lowest (10th percentile) group vs. that of women in the highest (90th percentile).
What did they find? The PCB levels they observed in their samples were similar to what scientists were seeing at the time (1960s) across the US. They were significantly higher than those observed today, following the bans that were instituted in the 1970s. The average level of PCBs across all women was 5.4 μg/L (micrograms per liter, or parts per billion). The 10th percentile was 4.7 μg/L and the 90th percentile was 8.7 μg/L.
In their overall assessment of the relationship between PCB level and sex of baby, they found that as PCB levels went up, the likelihood of a boy went down. This overall trend was statistically significant (p < 0.02). For every additional 1 μg/L PCB in serum, the odds of a boy dropped by 7%.
Comparing the women at or below the 10th percentile of PCB levels with those at or above the 90th, they found women with the relatively high levels the likelihood of a boy baby decreased by 33%. |
