Invisible Women


22 July 2019

The popular sleeping pill Ambien doesn’t work for women.

Or more precisely, it doesn’t work as designed. Zolpidem, the active ingredient that forms the basis of Ambien, is metabolised more slowly by women than by men, meaning that women are usually still impaired by the drug after eight hours of sleep. As it turns out, Ambien was developed by testing on male mice in the lab and on men in clinical trials. This practice is remarkably common in medical research, as a recent Science article by Northeastern University’s Rebecca Shansky shows. Even when animals of both sexes are tested, researchers often work things out in males first and then repeat the experiment on females. “It perpetuates the dated, sexist and scientifically inaccurate idea that male brains are a standard from which female brains deviate,” Shansky told The Guardian.

Shansky’s article came out as I was finishing Invisible Women: Exposing Data Bias in a World Designed for Men, a new book by the journalist and campaigner Caroline Criado-Perez. To say that the Science findings resonated with Criado-Perez’s argument is an understatement. Towards the end of Invisible Women, a chapter titled ‘The Drugs Don’t Work’ is a gruelling litany of all the ways in which the pharmaceutical industry is failing those of us who happen to inhabit female bodies. It gets a lot worse than Ambien. Antipsychotics, antihistamines, antibiotics, heart medications, antidepressants: all, Criado-Perez explains, have been shown to exhibit “menstrual-cycle impacts”. This means that depending on where one is in one’s cycle, the prescribed dosage can be either too low, and therefore ineffective, or too high, and therefore sometimes downright dangerous.

There are physiological differences between the sexes related to the heart, lungs, neurobiology and autoimmune system – in fact, to most tissues and organ systems in the human body. Although it is important not to overstate such differences – lest they get “weaponised by misogynists or used to justify and promote inequality,” as Shansky warns – the failure to acknowledge such differences has adverse effects on women’s health. Women are consistently underrepresented in clinical trials, even for treatments of conditions that are female-prevalent, such as depression. This is, in part, because women are seen to have “atypical” hormones (rather than simply different hormones to men), rendering them, writes Criado-Perez, “too complex, too variable, too costly to be tested on” in the eyes of a medical community conditioned to view the male body as the norm.1 And if you’re pregnant and fall ill? Forget it. The absence of pregnant women in clinical trials means we don’t have much data on how to treat you for, in Criado-Perez’s words, “pretty much anything”.2

How did we get here? Invisible Women mounts a persuasive argument about the existence of an almighty “gender data gap” – an absence of information about women’s bodily experience that characterises everything from medical research and urban infrastructure to automotive design and pension schemes. It shows, again and again, that even where data about women’s lives could be available, it is rarely sex-disaggregated. Criado-Perez takes pains to stress that this data gap “is not generally malicious, or even deliberate”. Often, she observes on a recent broadcast of BBC Radio 4’s Woman’s Hour, it’s just that “we are so used to thinking ‘man’ when we think of a human that we kind of forget to think about women unless we specifically mention them.” This bias is deeply embedded in language. Think of “man” in English and how it is taken for “humankind”. Or how, in most Romance languages, the generic masculine is used to describe groups of people whose genders are unknown, or which include a mix of genders. Thus, 100 female teachers would be “las profesoras” in Spanish, but “los profesores” as soon as you add one male teacher. “Such,” writes Criado-Perez, “is the power of the default male.”

These things matter because representation in language has a direct impact on representation in life. For instance, Criado-Perez cites studies in which children are asked to draw a scientist. While the proportion of drawings of female scientists has gone up considerably since the same experiment was conducted in the early 1960s (it is up from 1 per cent in 1960 to 28 per cent today), the meta-data nevertheless shows an alarming trend: “When children start school they draw roughly equal percentages of male and female scientists, averaged out across boys and girls,” Criado-Perez writes. “By the time children are seven or eight, male scientists significantly outnumber female scientists. By the age of fourteen, children are drawing four times as many male scientists as female scientists.” These are biases that are learnt and internalised from a very young age, through schooling and socialisation. Is the woeful lack of representation later on in life – in certain careers and in positions of power – so surprising?

In Invisible Women, Criado-Perez synthesises a staggering number of studies and interviews to show that, when representation is lacking in decision-making positions – be it in urban planning, medical research, or tech startups – we get urban plans, medicines and technology that fail to fully account or even to properly work for women. These include: mobile-phone handsets (too big), public toilets (too few), the standard piano keyboard (too wide), public transport (too unsafe), voice-recognition software (won’t register women’s voices as efficiently as men’s), pensions (won’t account for many women’s unpaid care burden), bullet- proof vests (don’t account for breasts) and cars (too deadly).

Car safety is a particularly shocking example of gender bias in product development. Criado-Perez shows how, although men are more likely to be involved in a car crash, women are 47 per cent more likely to be seriously injured and 71 per cent more likely to be moderately injured than men if caught up in one. This is because the design of crash-test dummies dates from the 1950s, and is based on the 50th-percentile male: a 1.77m, 76kg dummy that has “male muscle-mass proportions and a male spinal column”. It wasn’t until the early 2000s that “female” crash-test dummies started being used in the US, but even then, they were simply the standard male ones scaled down to match the female 50th-percentile height. They do not mimic female muscle-mass and spines. They do not have breasts. They are currently only used in the front passenger seat.

Of course, it gets worse.3 Although a pregnant dummy was created in 1996, its use is still not mandated in either the US or EU automotive industries, despite the fact that car crashes are the main cause of foetal death related to maternal trauma. Criado-Perez also observes that we haven’t yet developed a seatbelt that works for pregnant women. In fact, even the standard three-point seatbelt isn’t great if you have breasts: “In an effort to accommodate our breasts many of us are wearing seat belts ‘improperly’,” she explains. By this stage, I’m sure you’re beginning to get the point. The gender data gap does not just create products and systems that are inconvenient for women. It renders the world deadlier for us.

In addition to the designs and systems that fail to account for women’s bodies and female hormones, Criado-Perez identifies two more areas in which the gender data gap makes itself particularly felt. One revolves around women’s unpaid care burden. “Globally,” Criado-Perez writes, citing a 2015 McKinsey Global Institute report, “75 per cent of unpaid work is done
by women, who spend between three and six hours per day on it compared to men’s average of thirty minutes to two hours.” Housework, childcare, elderly care – even in the countries that boast the highest metrics of gender equality, these tasks fall disproportionately on women, especially when we cohabit in heterosexual relationships. In effect, many women in paid work perform an extra shift of unpaid work when they come home, which takes a toll on our mental and physical health.

The invisibility of women’s unpaid care burden has long been a feminist issue. The International Wages for Housework campaign, for instance, originated in Italy in the 1970s, and sought to make domestic and care work – “reproductive labour”, in the movement’s term – visible as the very foundation of industrial work. “They say it is love. We say it is unwaged work,” wrote Silvia Federici, a proponent of the campaign, in a 1975 text about the ways in which care work is naturalised as something women just like to do. Criado-Perez shows that we haven’t come very far since then. For example, GDP, the standard measure of a country’s economy, does not account for unpaid work in the home because, as one of Criado-Perez’s interviewees, economics professor Diane Coyle, explains, “this [is considered] too big a task in terms of collecting the data.” Current estimates, however, suggest that unpaid domestic and care work could account for up to 50 per cent of GDP in high-income countries, and 80 per cent in low-income ones. If it counted as productive labour, that is.

Beyond GDP, there are the everyday systems and services designed without taking women’s unpaid work into account. Criado-Perez shows how urban planning and transportation often fails to consider the more complicated travel needs of women “encumbered” (the industry term, not mine) by care work. “A typical female travel pattern involves, for example, dropping children off at school [in London, women are three times more likely than men to take a child to school] before going to work; taking an elderly relative to the doctor and doing the grocery shopping on the way home,” writes Criado-Perez. This is known as “trip-chaining” as opposed to travelling to and from work in a single-trip commute. Invisible Women sets out how everything from the design of housing projects to snow-clearing schedules privileges the motorised single-trip commute, a travel pattern that is much more likely to be the norm for “unencumbered” men.

Lastly, the other area identified by Criado-Perez as being particularly aggravated by the gender data gap is men’s violence against women. The final two chapters of Invisible Women are dedicated to the failures of supposedly gender-neutral disaster relief and aid programmes to account for women’s safety, and make for harrowing reading. Women are more likely to be sexually assaulted and abused – both within and outside of the domestic setting – when disaster strikes. Yet this is not reflected in the designs of shelters, whose poorly lit unisex facilities often leave women vulnerable. Criado-Perez cites “lurid stories of violence, of rapes and beatings” from Louisiana’s Superdome in the wake of Hurricane Katrina in 2005. “They said things didn’t happen at the Superdome,” one witness recalls. “They happened. They happened. People were getting raped. You could hear people, women, screaming. Because there’s no lights,
it’s so dark, you know.”

When vulnerable women are put in contact with men in positions of power, abuse and sexual exploitation often result. This was evident in the scandal that shook the global UK charity Oxfam in 2018, when it emerged that a number of its staff had patronised sex workers in the aftermath of the 2011 earthquake in Haiti – something Oxfam subsequently tried to cover up. Criado Perez points to this example as well as countless other instances of aid programmes, institutions and facilities meant to help, protect, or guard women that show that sexual harassment and abuse is endemic unless it is identified as a risk and mitigated by design. “Given the steady stream of abuse reports from around the world,” she writes, “perhaps it’s time to recognise that the assumption that male staff can work in female facilities as they do in male facilities is another example of where gender neutrality turns into gender discrimination.”

So what is to be done? Criado-Perez accepts that “closing the data gender gap will not magically fix all the problems faced by women, whether or not they are displaced.” But she is a strong proponent of collecting more data and making sure it is sex-disaggregated. It’s hard not to agree, given the overwhelmingly powerful case Invisible Women makes. But there is also reason to pause and consider how this should best be done. If sex- disaggregated data about any number of aspects of women’s lives is lacking, then it is even more absent when it comes to the intersections of race, class, disability, and so on, within that data. In the instances where information is available, we see how important these intersectional metrics can be in identifying systemic injustices. In the US, for instance, African-American women have worse health outcomes, overall, than white women. But, as Criado-Perez points out, when it comes to pregnancy and childcare, the comparison becomes truly abysmal: in the US, African-American women are 243 per cent more likely than white women to die from pregnancy and childbirth-related complications. An intersectional approach to collecting data is crucial here.

I had to read Invisible Women in short chunks. The panoply of injustices presented in the book begins to feel vertiginous if consumed too quickly. As a design journalist, I also felt acutely the cruel irony of what it usually means to “design” for women. As a 2016 report in The Times revealed, products that are needlessly branded for women cost 37 per cent more than those branded for men, whether that be a pink scooter (£5 more expensive) or Levi’s 501 jeans (46 per cent pricier than the men’s version). We live in a world where we get pastel Bic pens “for her”, but no cure for endometriosis; expensive kick-scooters but no functioning car seatbelts. Let’s collect more data, yes. But let’s also make sure that identifying women’s needs is not just a shorthand for identifying more consumers.