Is a woman more likely than a man to be seriously injured in a car crash of similar severity?

Question

This article makes a number of interesting points about the potential dangers faced by women in a world "designed for men", but I want to ask specifically about this claim:

But when a woman is involved in a car crash, she is 47% more likely to be seriously injured, and 71% more likely to be moderately injured, even when researchers control for factors such as height, weight, seatbelt usage, and crash intensity. She is also 17% more likely to die. And it’s all to do with how the car is designed – and for whom.

The links in this paragraph are either dead or go to similar articles with dead links.

Is the statistic credible and does it mean that car design is to blame?

Answer 1

I remember reading this article at the time; these kinds of articles tend to be riddled with hyperbole and exaggeration.

The article says 'But when a woman is involved in a car crash, she is 47% more likely to be seriously injured,'

They omit the key words https://genderedinnovations.stanford.edu/Gendered%20Innovations.pdf

Studies of crash outcomes show that women drivers are approximately 47% more likely than men drivers to sustain severe injuries in automotive crashes when researchers control for factors such as height, weight, seatbelt usage, and crash intensity; that is to say, a seatbelt-wearing woman driver involved in a crash is more likely to be injured than a seatbelt-wearing man driver of identical height, weight, and age involved in an identical crash

and in fact the source gives three possible reasons, all with sources:

1. Injury threshold: Women have a lower average injury threshold than men for some mechanisms of injury, such as whiplash, but young men have a lower velocity injury threshold than young women (Talmor et al., 2010; Stemper et al., 2004).
2. Design: Women may have excess risk because “effectiveness of occupant safety devices is biased toward the male occupants” (Dipan et al., 2011).
3. Type of vehicle driven: In the US, where data is available, women tend to drive cars with higher safety ratings than do men (Ryb et al., 2010).

This latter point does not ring true, and no citation is given for the horribly vague reference. My experience, which is supported by multiple studies, tells you that women driver smaller cars (https://pubmed.ncbi.nlm.nih.gov/12214367/). Smaller cars might score well on crash tests, but these crash tests are NOT comparable across sizes: https://www.edmunds.com/car-safety/are-smaller-cars-as-safe-as-large-cars.html In fact, there are five times more deaths per million vehicles for small cars compared to large SUVs, and it's rather obvious that you would rather be in the 3 tonne car hitting the 1 tonne car, than vice versa, assuming both are built for safety. Perhaps women choose safer cars on average within each category than men, but the male preference for larger cars will provide a safety benefit to men.

There are numerous sources which support the idea that men reduce their risk by driving larger vehicles: https://www.iihs.org/api/datastoredocument/bibliography/2219

As such, the claim in the Guardian article "And it’s all to do with how the car is designed – and for whom." is best categorized as 'untrue', since 'it's all to do with' isn't supported by the source. The source gives multiple possible explanations, without a firm conclusion.

The 47% claim is derived here: https://www.plan-norge.no/sites/default/files/body_attachments/Bose2011_Vulnerability%20of%20Female%20Drivers%20Involved%20in%20Motor%20Vehicle.pdf (it's freely viewable on many sites), however the study fails to adjust for vehicle weight.

The 2021 study "Injury risks and crashworthiness benefits for females and males: Which differences are physiological?" by Matthew L. Brumbelow and Jessica S. Jermakian found:

After controlling for some of these differences, the current study found that the odds of serious (AIS >= 3) non-extremity injury were essentially equal for males and females (female OR: 0.98). Differences did remain for moderate severity injuries (AIS >= 2), especially to the extremities. As male drivers tended to be in heavier vehicles, even within the controls of this study, it is unknown how much of the remaining differences in injury odds may still be due to non-physiological factors.

The study notes that females have an increased risk of lower extremity injury, and suggests this could be due to "bone and ligament properties (Forman et al. 2019), stature (Dischinger et al. 1995; Crandall et al. 1996), or footwear (Crandall et al. 1996)."

It is further noted that

Previous work has shown serious and fatal injury risk has decreased more for females than males when comparing newer and older model year vehicles (Kahane 2013; Kullgren et al. 2020). The data available for this study indicate current testing programs have promoted countermeasures that are effective at reducing the risk to both sexes, perhaps to females more. Seat belts and modern front and side airbags have been shown to be equally effective at reducing fatal injuries for males and females (McCartt and Kyrychenko 2007; Braver et al. 2010; Kahane 2015), suggesting that some of the differential benefit to females may be due to nonrestraint crashworthiness improvements such as stronger structures and improved energy management.

One particular key phrase is B-pillar intrusion, which relates to side impacts. In other words more recent improvements in car safety might have disproportionately benefited women who do appear to differently vulnerable to men.

There seems to be a tendency in the article (which is not that old) to ignore the work that has been done to specifically focus on injuries to women and to draw conclusions that are not necessarily supported by data. Recent improvements to vehicle strength might have disproportionately improved safety for women, but whether that was achieved by focusing on women specifically, or it was simply a function of the nature of the improvements, is not possible to say.

It is worth noting that the 17% fatality claim, in the Guardian article, is largely debunked in its own source:

Since the 1990s, females’ risk relative to males has shrunk substantially, perhaps to half its original level or less; Figure 6 shows that the trend after 1990 is strong

In fact, for post-2000 cars (the study was published in 2013), the risk was given as around 8%, which is much lower than 17%.

And an explanation is given, which does not involve crash test dummies:

The combination of air bags and belt use, especially belts with pretensioners and load limiters has greatly reduced the gender gap, at least for fatality risk.

Further:

However the increase in rollovers is primarily among unbelted females. They are at 33.3 ± 9.2 higher risk than unbelted males. Unbelted females are more likely to be ejected from the vehicles than unbelted males, presumably because females, being smaller on the average than males, can pass more easily through ejection portals such as the sidewindow area.

Physiological justification is given there for female's higher risk of neck injury, but none for abdominal

In other words, women are exceptionally more vulnerable to neck and abdominal injuries. The high risk of neck injury appears to be related to the anatomy of a typical female: a male’s neck has greater spinal-column strength than a female’s, yet a female’s neck is called upon to support and control the motion of a head that is almost as large and heavy as a male’s. The high risk of abdominal injury is harder to explain.

Therefore I would conclude:

1. the article is outdated in its sources, including data based on pre-2000s cars, which no-one who cares about car safety would be driving
2. it is a British article but it is unclear to what extent its conclusions, based on the specifics of the US car market, are valid elsewhere
3. it tends to conclude that everything is based on 'failure to care about women', which is false in that some of the differences appear to be based on real physiological differences, which cannot necessarily be solved with car safety
4. it makes the false claim (e.g. in the caption to the image at the top) that 'It wasn’t until 2011 that the US started using a female crash-test dummy', and it tends to blame everything on the failure to use female-sized crash test dummies (in fact, the sources indicate that the size of the crash dummies became much more important after mandatory seatbelt laws (effective in most US states between 1984 and 1991), and mandatory airbags (from 1998)). The 'Hybrid III female' crash test dummy was first developed in 1988, and used by the car industry shortly after, was mandated in the US from 2000 in several tests. What happened in 2011 was that the 5th percentile female was included in passenger seat testing, a change which appears to have resulted in improvements for subsequent model years for certain vehicles. Obviously caution is needed in evaluating the validity of tests based on dummies, as opposed to actual real-world humans: the average American woman is closer in weight to the '50th percentile male' dummy than the 5th percentile female, the 5th percentile female dummy is essentially a teenage girl (although obviously some adult women are petite), and there exist also even smaller child dummies; a 50th percentile female dummy might be insufficiently different from the 50th percentile male. Nobody is clamouring for a 95th percentile male, however...
5. Vehicle safety standards become much more rigorous over time. NHTSA crash testing started in 1978, with side-impact testing added in 2003 (tests which were substantially improved in 2021. The argument that 'the 1978 tests should have used a 4'11" (5th percentile female), or 5'4" (50th percentile female) dummy instead/as well as 5'9" dummy' (50th percentile female) doesn't seem supported by evidence - safety improvements come from mandatory technology (such as airbags, which were resisted by the US motor industry), and more comprehensive testing
6. it ignores other factors besides 'poor testing', such as vehicle size and physiological differences

Nonetheless, it is true that women are more likely in a similar crash to suffer certain injuries, although none of the 47%, 71%, or 17% claims appear to be accurate for modern cars in comparable crashes.