Plastic 'makes boys feminine'

Tuesday November 17 2009

"Chemicals used in plastics feminise the brains of little boys", according to the_ Daily Mail._ The newspaper claimed that boys who are exposed to high doses of phthalate chemicals in the womb are less likely to play with male toys or join in with rough games. Phthalates are a family of chemicals found in PVC shower curtains and vinyl flooring.

The research behind this news compared phthalate concentration in pregnant women’s urine to whether their children’s play habits were typically masculine or feminine at the ages of four to seven. However, the study only looked at a small number of children, and of those invited to take part, only half responded. In addition, phthalate concentration was only measured once during pregnancy. These limitations mean that the evidence from this study alone is too weak to form any definite conclusions.

Parents should not be concerned about reports of “gender-bending” phthalates affecting their children’s brains or habits.

Where did the story come from?

This research was conducted by Shanna H Swan and colleagues from the University of Rochester, New York, and other institutions in the US and UK. The study was funded by grants from the US Environmental Protection Agency, the National Institutes of Health in the US, and the State of Iowa. The study was published in the peer-reviewed International Journal of Andrology.

Although the BBC, Daily Telegraph and Daily Mail have accurately reported the sample size and main findings of the study, none of them mentions the most important limitation: that extremely few of the initial sample participated in the follow-up research sessions. When considered in isolation, the evidence gathered from this small number is unlikely to be representative of the whole sample.

What kind of research was this?

This was a small cohort study which investigated how boys’ brain function and development are affected by foetal exposure to phthalate chemicals. Animal studies have previously demonstrated that foetal exposure to such ‘antiandrogens’ in rats leads to less male-type behaviour.

A cohort study is usually a reliable form of study for assessing the relationship between cause and effect. However, this particular cohort study has several limitations, principally its small size. Only 74 boys and 71 girls were assessed, which represents 45% of those who were invited to participate. There is also some difficulty in attributing phthalate exposure to the cause of the children’s gender-type behaviour due to numerous possible confounding factors that were not taken into account.

What did the research involve?

The researchers contacted couples who took part in the earlier Study for Future Families, which recruited expectant mothers and their partners between 2002 and 2005. At the time, they completed a questionnaire and gave a urine sample, which was used in this latest study to measure phthalate concentration. For the purposes of this study, the women were contacted when their child was four to seven years old.

The researchers looked at how the mother’s phthalate concentration in mid-pregnancy related to the child’s gender-type behaviour. Child behaviour was reported by parents using a list of 24 activities, child characteristics and items (e.g. type of toys). Half of the listed entries were considered to be feminine and half masculine. A Parental Attitude Scale was also used to take into account factors that may have affected the children’s choice of play, such as the type of toys available in their household and parents’ attitude towards boys playing with “girls’” toys.

The study design had strengths in that it used a validated scale for assessing child gender-type behaviour and also took into account how parental attitudes may have influenced this. When assessing the relationship between gender-behaviour and phthalate concentration in the mother’s urine in pregnancy, the researchers considered various possible confounding factors, including mother’s kidney function, sex and age of child, parental education, number and sex of siblings, ethnicity and parental attitudes.

However, little is known about the effects of phthalates in humans, and as children’s play behaviour is likely to be complex and affected by many factors, there are probably numerous other confounding factors that were not accounted for. Also, measuring phthalates in the urine at a single point during pregnancy is not a reliable indication of exposure levels over time, which may be highly variable. 

Importantly, the researchers only received completed questionnaires from 45% of families who were sent them (150/334). This is a low follow-up rate and limits the findings of a cohort study which had a small sample size to begin with. It is also unclear exactly what proportion of the original Study for Future Families this represents.

What were the basic results?

The research found that higher concentrations of certain phthalate chemicals in expectant mothers’ urine were associated with a less masculine behavioural score in sons. No relationship was seen between concentrations of other phthalate chemicals and boys’ behaviours, or between any phthalate chemicals and girls’ behaviour.

How did the researchers interpret the results?

The researchers conclude that their data, “although based on a small sample”, suggests that exposure to phthalates in the uterus may be associated with less male-typical play behaviour in boys.


This research has attempted to address the question of how phthalate exposure may affect gender-type play behaviour. However, there are important limitations to the study and the evidence is too weak to form any definite conclusions about the relationship.

  • The cohort study was very small and represents less than half of the eligible families. As such, results should be interpreted with caution and are unlikely to be representative of the whole sample. If the whole cohort had been included, the results may have been different.
  • Little is certain about the effects of phthalates on humans, and children’s play is a complex behaviour that is likely to be affected by a wide variety of factors. Given their complexity, both areas of interest are likely to be confounded by factors that have not been accounted for.
  • A one-off measurement of the mother’s phthalate exposure gives no indication of her exposure over a longer period of time, which may be variable, or of the child’s direct exposure during their own lifetime.
  • The significant association in boys was seen for only some phthalate chemicals and not others.
  • The study carried out multiple statistical tests, which increases the possibility that significant findings will be found by chance.

The implications of this research are currently limited. Without much further research, there is no definite evidence of the influence that phthalates might have on gender play and behaviour, or which types of plastics are likely to give the highest exposure.

Analysis by Bazian
Edited by NHS Choices