Pregnancy and child

'Male hormones' in womb linked to autism

“Boys who develop autism may be exposed to higher levels of hormones…in the womb,” The Daily Telegraph reports.

Autistic spectrum disorder (ASD), commonly known as autism, is more common in males, but the reason why is still unknown. One hypothesis is that there may be factors in the development of male infants that increase their risk of ASD, such as exposure to certain hormones.

This study measured sex hormone levels in the amniotic fluid (which surrounds and supports the baby) of over 300 boys. Hormone levels were found to be higher in boys who developed ASD. 

However, this does not necessarily mean this is the cause of ASD. The average levels were higher, but many of the boys who developed ASD had normal and low levels, meaning there has to be other factors that play a role in the development of these conditions.

These findings have no immediate implications.

The researchers argue that even if a higher level is proven in larger studies, and is found to be a causative factor, any attempt to block the effects of hormones using drugs would be unwise, due to the risk of side effects.

This research did not show that the higher levels of the hormones caused ASD, which is still believed to be due to a combination of genetic and environmental factors.

Where did the story come from?

The study was carried out by researchers from the University of Cambridge and the Statens Serum Institute Copenhagen. It was funded by the Medical Research Council UK, alongside project grants from Nancy Lurie Marks Family Foundation, Wellcome Trust and the Autism Research Trust. The study was published in the peer-reviewed medical journal Molecular Psychiatry. It has been published on an open-access basis, meaning is free to read online.

The UK media reported the story accurately, apart from the headlines describing the hormones as “male”. All five hormones studied are present in both sexes, and it is only the level of testosterone that is higher in men.

The media acted responsibly by including important commentary from the lead author of the study, Professor Baron-Cohen, clarifying that the study’s results do not mean it would be possible to carry out prenatal tests for ASD or treat the condition in the womb using hormonal treatments.

What kind of research was this?

This was a retrospective cohort study, using data from the Danish Historic Birth Cohort.

It aimed to see if there was a link between the sex hormone levels in the amniotic fluid (which surrounds the baby in the womb) and the development of autistic spectrum disorders (ASDs).

This is because ASD is diagnosed much more frequently in males, and previous research has suggested that sex hormones and cortisol affect the developing brain. 

A retrospective cohort study is useful when a randomised controlled trial (RCT) is not possible.

Attempts can be made to account for confounders (other factors that may be causing any effects seen), but there could still be other explanations for the results, so it is difficult to imply direct causation.

What did the research involve?

Researchers measured hormone levels in the amniotic fluid samples of boys born between 1993 and 1999. They then compared the levels of hormones in 128 that later developed an autistic spectrum disorder (ASD) with 217 that didn’t (the controls).

The researchers used amniotic fluid samples that had been frozen and stored at -20°C. They used mass spectrometry techniques (where devices are used to analysis the molecular composition of a substance) to measure the level of hormones.

Specifically, cortisol (the stress hormone that is essential for life) and four sex hormones – progesterone, 17α-hydroxy-progesterone, androstenedione and testosterone – were analysed.

The researchers recorded the following potential confounding factors:

  • maternal age
  • paternal age
  • birth weight
  • gestational age (weeks of pregnancy) when the amniotic fluid sample was taken
  • apgar score (indicating the physical health of the newborn immediately after birth)
  • sample storage time

The Danish Psychiatric Register was searched over 2009 and 2010 to identify all diagnosed cases of ASD, according to the WHO-recognised International Classification of Diseases (ICD-10), which were linked with the amniotic samples.

They decided to limit the study to males for two reasons. The first was that when they looked at all samples, there were only 24 girls who developed an ASD, which was deemed to be too small a sample to produce meaningful results. Secondly, there was significant variation in two of the confounding factors between girls who developed ASD and those that didn’t – fathers were significantly older and birth weight was lower, compared to the controls.

Statistical analysis was performed to measure any difference between the two groups and also to account for the confounding factors listed above.

What were the basic results?

The average level of each hormone was higher in boys who developed ASD, but the exact measurements were not provided in the study.

All five hormones were of a similar level across the control group. They were also of a similar level in the ASD group, but the mean average was higher than in the control group. The researchers say that this indicates a similar mechanism for the production of these hormones, as there wasn’t one that was substantially higher or lower than any other.

They interpreted that one overall “steroidogenic factor” (a protein that stimulates sex hormone and cortisol production) was significantly higher in the ASD group compared to the control group, and was causing the difference.

There was no statistically significant difference between the groups for any of the confounding factors, and no change to the results when their analysis was adjusted for these factors.

How did the researchers interpret the results?

The researchers say this is “the first direct evidence that steroidogenic activity is elevated in fetal development of those who later receive diagnoses on the autism spectrum”.

The researchers are quick to point out that the “source of elevated steroidogenic activity in the fetal development of autism was not tested in the current study, and more research will be needed to understand how different sources – such as the foetus, mother, placenta or other environmental factors – might contribute to such elevations”.

They are also tentative about the elevated level of cortisol seen, stating that: “the current results may suggest a link between [pre-birth] stress and autism via heightened fetal cortisol; however, it is unclear if the association here is due to heightened [pre-birth] stress or is driven by a more primary fetal sex steroid influence that has a side effect of boosting fetal cortisol levels”.


This study found that the levels of four sex hormones and cortisol were slightly higher in the amniotic fluid of boys who developed ASD, compared to those who didn’t.

However, this does not necessarily mean it is the cause of ASD. The study accounted for some confounding factors, and the average levels were higher, but many of the boys who developed ASD had normal or low levels. This means that other factors must play a role in the development of the condition.

Although it was a well-designed study, limitations reported by its authors include analysing samples that were over 10 years old, which may have changed over time, despite being frozen at -20°C.

There are no immediate implications for these findings; this is just one part in the long process of figuring out the cause of ASD. It is still believed to be due to a combination of genetic and environmental factors.

NHS Attribution