Sex bias has far-reaching consequences
Males aren’t females
‘I remember when I did one of my first experiments in my neurobiology studies. I was working with male rats, so I asked my supervisor why we didn’t use females as well. He said something like, “Oh I don’t know, this is how we’ve always done it.”’
Even though she didn’t get a proper answer, assistant professor of neurobiology Nicole Gervais never forgot the question. Now, her own research focuses on the female brain. That’s because she knows there’s a difference between using male and female test subjects.
Using males has been the standard in the scientific community for years; it’s called sex bias. According to Gervais, it’s usually just a matter of convenience – people tend to test their hypotheses using designs that were used in earlier studies, because it’s easier to interpret results if the only things that differ are the variables they want to look at. That means that if everyone else used male test subjects, so will the researchers after them. ‘People assumed this wouldn’t impact the results of the research, since the difference didn’t seem very big.’
Incorrect treatment
However, the opposite turned out to be true. Not only are males and females not the same, the preference for male test subjects has led to a dearth of knowledge on the female body. ‘That can lead to incorrect treatment in female patients’, says Gervais.
Heart medication Digoxin worked for men, but was dangerous for women
Heart medication Digoxin was manufactured in the late twentieth century, and it had only been tested on male subjects. ‘While it worked for men, it turned out to be dangerous for women.’ Later, it was confirmed that this was caused by a small variation in the heartbeat between men and women.
The solution might seem obvious: researchers should have as many male as female test subjects in their studies. But, says Gervais, it’s not that simple.
Oestrogen
There is a test that measures a person’s spatial skills using a three-dimensional cube in various positions. ‘Men tended to do better in this test than women. This led to the conclusion that men have better spatial skills than women. But it later came to light that the women’s performance depended on their menstrual phase’, says Gervais.
The women whose oestrogen was low at the time of the test did just as well as the male test subjects. But women with a medium or high level of oestrogen generally performed worse. ‘And because women have elevated levels of oestrogen throughout most of the month, it seemed as though they did worse on the whole.’
The women’s performance depended on their menstrual phase
In other words: if you want to involve female test subjects in your research, you need a decent set-up to factor in these kinds of differences. ‘It’s a little more complicated than just saying you’ll add females and see what happens.’
Researchers should use an equal number of male and female test subjects, apply the proper statistical comparative analysis and – depending on the research question – take into account the test subject’s hormonal state, emphasises Gervais.
Accurate tracking
Especially that stipulation can lead to practical challenges. ‘The hormonal cycle for most test subjects like mice and rats lasts only a few days.’ In order to analyse the effects of that cycle, researchers have to be able to accurately track its stages. ‘We can estimate which stadium female test subjects are in by taking samples from their vagina.’
The menstrual cycle of test subjects can also present a challenge. To establish where in the cycle a test subject is, ovulation tests can be used instead of vaginal samples.
People aren’t making an effort, they’re just maintaining the status quo
Finally, researchers have to ask themselves if the test subjects’ hormonal cycle matches those of the target. ‘If you’re making medication for women with a menstrual cycle, your test subjects need to have a hormonal cycle as well’, Gervais explains.
Conversely, if someone is developing medication for Alzheimer’s disease, they have to remember that most of those female patients no longer menstruate, which means the female test subjects must be post-menopausal as well. After all, the ageing process between men and women is dissimilar as well.
And if the human target for a particular drug uses hormonal contraceptives, the test subjects will also need to be treated with a form of hormonal contraceptive.
Chromosomes
Gervais takes it even further in her own research. Not only does she look at the sex hormones present in the test subjects’ hormone cycle, she also looks at their chromosomes. For instance, she’s studying the female brain to find out why women have a greater risk of developing Alzheimer’s disease.
For this, she uses mice whose gene that determines their sex is no longer located in the Y chromosome. This causes the female animals to develop male sexual characteristics and hormones, and vice versa. ‘This allows us to test whether the differences in results between male and females are caused by the sex hormones or these genes’, Gervais says.
Stronger impulse
In spite of all the efforts to use more female test subjects in research, the sex bias remains a problem. Gervais thinks it’s mainly a habit that needs changing. ‘People aren’t making an effort, they’re just maintaining the status quo. We need a stronger impulse, but I don’t think the scientists will provide that.’
She argues in favour of outside intervention, for instance by the European Research Council, which is responsible for handing out research subsidies. ‘They could state a preference for studies that prevent sex bias in their designs.’
But regular people could also join the fight. ‘Say your doctor prescribes you certain medication, you can ask them if the drug has been tested on women’, Gervais suggests. ‘This creates more awareness, which forms the basis for change.’