It makes me very happy that you're open to this! I'll write a proper mini-essay tomorrow (when I don't need to get some sleep), because many women have worked hard to improve safety outcomes for everyone regardless of body shape, and future is looking bright too.

What I will say for now is that I think I've seen at least one of your referenced books before, and it quoted statistics that were well out of date.

Disclaimer: I've still been snowed under in work, so I haven't found the research I wanted to, and a lot of this is going to be from memory.

tldr; mortality and serious injury outcomes from road traffic accidents for people in cars made in the last 10 years have nearly reached gender parity, the newer the vehicle the more equal (and in general better over all) it gets. That includes people of different body shapes and ages, too.

Now for the long part ... a lot of this is general history of crash testing for context.

After "unsafe at any speed" in 1965 vehicle safety began to be taken seriously in the richest country in the world, which also dragged the rest of the "first world" with it. However it was generally something only of interest to boffins wearing white coats, undertaken behind closed doors, and oversight was entrusted to government departments that didn't interact directly with the public.

Slow and steady progress was made, in 1959 Volvo came up with the modern 3 point seatbelt design and chose to share it with the world for free (they then went on to become obsessed with safety, but that's another story). Every improvement in safety was resisted by car manufacturers, bitterly, because it cost them money. Progress was slow, and research was basic. It lead to some really cool and weird technology, but I'm not going to get started on that here.

Fast-forward to the mid-nineties and the outcomes from accidents hadn't got that much better. In 1996 the first properly dedicated female crash test dummy was released, the Hybrid III fifth percentile (fifth percentile means that it's the size of the statistically smallest 5% of women, ie. petite). It had been developed because of a specific problem that women face - "submarining" - where they basically slide out under a seatbelt because it was designed for men.

This was a big deal because developing a new crash test dummy takes a really long time and costs an insane amount of money. This is something else that I find cool but won't go into here, however because female test dummies come after the male ones they are, and continue to be, more advanced.

In 1997 the giant gorilla in the room appeared and shook things up - EuroNCAP. Made to be advocates for car safety for the public they were the first organisation to openly promote the results of crash testing, which they did spectacularly with the Rover Metro (you can see one being tested on youtube for the 20th anniversary of EuroNCAP). Sales of the car plummeted after it was shown how graphically dangerous it was; production ceased, and the attention of car manufacturers was focused - they were now on notice.

The public also noticed, and when the Renault Laguna became the first car to pass a EuroNCAP test with five stars it became a central part of the car's advertising campaign.

EuroNCAP has now spawned GlobalNCAP and many regional organisations (the most interesting one to watch at the moment is LatinNCAP, where many car makers try and pawn off shitty old designs with a makeover onto the Central and South American populace).

This had an interesting side-effect of seeing a lot of money being poured into both research and design of safety systems in cars (some of which has also trickled down into commercial vehicles, another subject for another day), and it's dragged research for the safety of women and children along with it.

After the Hybrid III 5% dummy came the SID IIs, which I just spent the last 2 days working on because my most senior technician is on paternity leave for the next couple of months. It's a fantastic design, again a fifth percentile female in size, with many more sensors than it's male equivalent, the EuroSID.

Between these two dummies the vast majority of crash scenarios that see women hurt where men would not be are covered (and thus poorly performing cars can be forced out of production), however they're not perfect. In crash test dummy design the thing always being chased is "biofidelity", or "how like a real person this thing reacts". The Thor and WorldSID dummies are currently the most advanced standard dummies available, and it has been noticeable that development of the more advanced female versions began as soon as the male ones entered use.

The female versions of both Thor and WorldSID are currently available, however because of the speed of regulations and the sheer cost, they won't be used for standards testing until a couple of years time. To give perspective, design of Thor started in 1992 and it first got used in widespread testing in 2020. A new one costs about a million euros. Standards testing with male dummies still uses the older designs for most things, so the continuing use of the older female dummies isn't an issue of sexism.

Special dummies: so, this is where I would put the info about the research if I had it, suffice it to say some great work has been done with a "pregnant" test dummy which used abdominal pressure sensors. IIRC the result of the research affected the programming of seatbelt load-limiters, whose effects can be monitored using the standard dummies because they have sensors for the sacro iliac. This is from memory though, so I could be wrong!

The world's most expensive crash test dummy is modelled on an average 70 year old woman, and includes sensors for the major internal organs. It was funded by the EU as part of their research into improving road safety, I've been involved in some other aspects of the research and it's been well thought out and forward looking. They have specifically checked for gender parity and diversity in the research, for instance accounting for internal camera data for women who wear hijabs.

The future: In a presentation last year the head of dummy research at Humanetics, the largest dummy manufacturer and main researcher into testing technologies, stated that he sees there's a gap in dummy body shapes between the average male and fifth percentile female, and wants to find out if any bad effects are slipping through the net. He did seem to genuinely make sure that road safety for women is taken seriously,

In summary, testing has been sexist, but over the last 30 years a mountain of work has been done to improve it, along with testing in general. It's now headed on a healthy trajectory, and car manufacturers can't get away with short-changing people on safety like they used to.

This is an article that talks about statistics taken from 1998 to 2015 which covers some of the points, but also brings up other things which may affect crash outcomes for women: https://www.iihs.org/news/detail/vehicle-choice-crash-differences-help-explain-greater-injury-risks-for-women For me it's interesting because here in Europe we have (statistically) different crashes and the vehicle size differential is less pronounced. Things get a bit more worrying when you see the results split by socioeconomic background, but again that's a subject for another day.

Hope this helps clarify things!