Schneier on Security
A blog covering security and security technology.
« The Onion on the Hudson River Plane Crash |
| 1801 Cipher Solved »
March 17, 2009
Leaving Infants in the Car
It happens; sometimes they die.
"Death by hyperthermia" is the official designation. When it happens to young children, the facts are often the same: An otherwise loving and attentive parent one day gets busy, or distracted, or upset, or confused by a change in his or her daily routine, and just... forgets a child is in the car. It happens that way somewhere in the United States 15 to 25 times a year, parceled out through the spring, summer and early fall.
It's a fascinating piece of reporting, with some interesting security aspects. We protect against a common risk, and increase the chances of a rare risk:
Two decades ago, this was relatively rare. But in the early 1990s, car-safety experts declared that passenger-side front airbags could kill children, and they recommended that child seats be moved to the back of the car; then, for even more safety for the very young, that the baby seats be pivoted to face the rear.
There is a theory of why we forget something so important: dropping off the baby is routine:
The human brain, he says, is a magnificent but jury-rigged device in which newer and more sophisticated structures sit atop a junk heap of prototype brains still used by lower species. At the top of the device are the smartest and most nimble parts: the prefrontal cortex, which thinks and analyzes, and the hippocampus, which makes and holds on to our immediate memories. At the bottom is the basal ganglia, nearly identical to the brains of lizards, controlling voluntary but barely conscious actions.
Diamond says that in situations involving familiar, routine motor skills, the human animal presses the basal ganglia into service as a sort of auxiliary autopilot. When our prefrontal cortex and hippocampus are planning our day on the way to work, the ignorant but efficient basal ganglia is operating the car; that's why you'll sometimes find yourself having driven from point A to point B without a clear recollection of the route you took, the turns you made or the scenery you saw.
There are technical solutions:
In 2000, Chris Edwards, Terry Mack and Edward Modlin began to work on just such a product after one of their colleagues, Kevin Shelton, accidentally left his 9-month-old son to die in the parking lot of NASA Langley Research Center in Hampton, Va. The inventors patented a device with weight sensors and a keychain alarm. Based on aerospace technology, it was easy to use; it was relatively cheap, and it worked.
Janette Fennell had high hopes for this product: The dramatic narrative behind it, she felt, and the fact that it came from NASA, created a likelihood of widespread publicity and public acceptance.
That was five years ago. The device still isn't on the shelves. The inventors could not find a commercial partner willing to manufacture it. One big problem was liability. If you made it, you could face enormous lawsuits if it malfunctioned and a child died. But another big problem was psychological: Marketing studies suggested it wouldn't sell well.
The problem is this simple: People think this could never happen to them.
There's talk of making this a mandatory safety feature, but nothing about the cost per lives saved. (In general, a regulatory goal is between $1 million and $10 million per life saved.)
And there's the question of whether someone who accidentally leaves a baby in the car, resulting in the baby's death, should be prosecuted as a criminal.
EDITED TO ADD (4/14): Tips to prevent this kind of tragedy.
Posted on March 17, 2009 at 1:10 PM
• 152 Comments
To receive these entries once a month by e-mail, sign up for the Crypto-Gram Newsletter.
Of course, without questioning that a child's life is worth a million dollars (of course it is), one must wonder how many lives that money could save if thrown at a more common risk.
Prosecuting as a criminal? To deter other parents that already believe it would never happen to them?
At $1 million per live saved, we'd be much better off investing in urban planning that doesn't rely on cars.
We wouldn't prosecute the parent if they accidentally rammed a bridge support and killed their child that way, right? Why should accidental abandonment be any different?
When one of these cases occurs, the demonization of the responsible parent is amazing. I think that part of the reason is that if other parents can label the guilty party a monster, then no one has to think about the possibility that it could happen to them. The parent who forgot the child is a monster, not someone like me, so that can't happen to *my* child.
A similar (though much less severe) problem: My wife and I have slammed our children's hands in our doors because we didn't see them grab the support between the rear and front doors as they were climbing into the back seat--we just don't see them and it's usually not a problem.
I can only barely imagine living with the guilt these parents feel. How can a marriage survive such an act? How can one look at one's surviving children without both of you being reminded of the child you killed?
It seems like the job of the law to separate those for whom this is an accident, from the small minority for whom this might be an intentional way of killing their children. (Though I suspect even in those cases severe stress and mental illness usually play a role and complicate the matter of culpability.)
At work, if I promise someone a ride home, I give them my keys so I *can't* leave without them. I've wracked my brain trying to think of some way to prevent these errors for children in cars. For the car seats, maybe a mirror that let's you see your baby's face while you're driving? For the fingers in the door, maybe a convex mirror at each front seat door? Maybe people could give their access cards or wallets to their babys? None of these solutions is really satisfying.
I'm an atheist, but something about these stories makes me think: "God bless these fathers and mothers and have mercy on their souls."
Don't prosecute them as criminals, but prosecute them as inept parents. The person who hits a bridge support is prosecuted as a bad driver.
That has happened in the southern Ohio (USA) area twice within two years. Neither person was charged with a crime. Which I believe was the proper path. What can the state do to punish a parent more? And like Bruce said, this is caused by part of our hard-wired brain.
A politician is considering changing the law to make it more likely to convict these "monsters". I hope I get to be on one of those juries. I would never convict.
How about stuffing, say, $2,000 in cash into your baby's pocket? Maybe do it just occasionally, so it doesn't become routine. Might focus the mind a bit....
Interesting idea with handing the keys to the guy you're giving a ride to. I'll have to see where else I can apply that kind of thinking.
Re: give cow-orkers your keys.
That's a brilliantly simple and effective idea.
Randy -- whosechildrenthinkhesabaddad
We need smarter cars, so they can turn up the air conditioning and call their owner if they're stuck with 'extra' passengers.
Solving the problem at the human end is probably impossible.
"When our prefrontal cortex and hippocampus are planning our day on the way to work, the ignorant but efficient basal ganglia is operating the car" ...
New bumper sticker: "WARNING: basal ganglia in use"
@Baddad: We can slam the doors of our Subaru Outback closed on our fingers, at least up in the window part. It surprised me when the salesman did it on the lot. The window-part is frameless and flexes against a soft rubber surround.
When I read the article last week, I wondered if it would be possible to hook a weight sensor into the door-open circuit, then you'd see the door-open lights on when your kid was in the back, and remote-lock signal might (on some cars like my Subaru) beep differently with an open-door signal.
This is a silly statement in the piece:
"The human brain, he says, is a magnificent but jury-rigged device in which newer and more sophisticated structures sit atop a junk heap of prototype brains still used by lower species. "
The "older" parts are the most efficient, most developed parts. It's the newer pieces that are the least developed and least sophisticated. Just think of the effort you need to do mathematics with your prefrontal cortex, versus the ease with which you drive a car.
The problem is when the frontal cortex changes something (an interruption in routine), but forgets to tell the rest of the system, because the frontal cortex is a jury-rigged recent add-on that isn't properly integrated.
Don't blame the lizard brain -- it's had 300 million years to evolve. Blame the frontal cortex, which was a nub for 100 million years and only really took off 750 thousand years ago; barely a blink of an eye in evolutionary time.
Between $1 million and $10 million per life saved?
Hell, with that kind of money to spend, we could "save" enormous numbers of lives. We could start by buying every homeless person in America a house, free and clear. Buy every person in America who has a mechanically unsound car a new car with the latest safety features. Hire a personal driver for every geriatric driver who should have turned in his or her driving license the day he or she couldn't see past the end of the hood (or over the steering wheel) any more. Buy every woman who lives alone a gun and a couple of personal-protection classes. Install a free state-of-the-art pool alarm system in every swimming pool in America. Buy straitjackets for the whole of Congress. Just for starters. And think of all the jobs we'd create!
No, I'm not serious. Not completely. But it's absurd that we're spending that kind of money per accident on rare events when there's so many far more common causes of death — exposure, auto accidents, drowning — that we could be addressing far more cheaply. How many pool alarm systems does ten million dollars buy?
Simpler and cheaper solution:
Secure a string to the carseat and yourself (belt, wrist, etc.). Leave plenty of slack so it doesn't interfere with driving, but if you try to exit the car and forget, the string will tug. Having to unclip the string will provide the reminder.
"The human brain, he says, is a magnificent but jury-rigged device in which newer and more sophisticated structures sit atop a junk heap of prototype brains still used by lower species. "
So, our brain is analogous to Microsoft Windows.
Wow, talk about priorities. If your child doesn't have your priority at all times, then I am sorry for you.
Go to work and forget your child in the car? What is more important? Apparently not your child in that case.
Slam your child's fingers in the car door? What's more important? Apparently not ensuring your child is safetly in the car before closing your door in that case. Even today, with my teenage children, I still ensure they have safely entered (properly buckled in) or exited the car/van before closing my door.
The weight sensor idea is good, but you couldn't hook it into the door open circuit, because your car would complain continually when driving that a door is ajar. It could also be real problematic because a baby in a car seat weighs in the same range as a heavy bag, so you'd get alarms when you left a bag on the seat. (My car does this for the front passenger seat, it's a pain...)
Uh oh, I see a line about "marketing studies".
Remember how the Herman Miller Aeron chair was unpopular during the metrics of consumer research? And yet someone decided to sell it anyway.
Perhaps all it needs is a little creative viral marketing and this Child Presence Sensor would get a better name and off the ground.
My guess would be something related to the bureaucracy around a NASA research center product is really what has complicated and stalled interest; not marketing studies, not liability, but the process of managing an inexpensive consumer product out of NASA and into the market.
If you can't, even accidentally, concentrate on something other than your child for even a moment, then I'm sorry for you! Must be an insanely stressful life...
There is no way to really know for sure, but I wouldn't be surprised if sensor devices and measures to prevent this may actually cause an increase in incidents (perhaps even incidents of another kind). People may depend on such devices and be more careless.
Sort of like how banning peanuts makes people less aware of how to react when they or someone else who is allergic eats one, or how AutoCorrect on word processors make people worse spellers because they seldom proof read or learn what they consistently misspell.
May happen, may not. Just a consideration.
@ HJohn at March 17, 2009 2:46
Also wanted to add that Uncle Murphy will hitch a ride in the car on a day when the sensor quits working. The person will not double check because they truest their handy-dandy device. Then, of course, they'll sue the manufacturer because the batteries went dead or they didn't know it would quit working after they dropped it in a water puddle.
> if I promise someone a ride home, I give them my keys
Great idea. I connect my keys to my usb flash drive so I won't leave home without it.
Maybe an effective abandoment-prevention habit could be to string something from the carseat to the ignition key or driver's door handle. Or maybe position a small mirror in the rear window so that when driver checks his rear-view, he can see the infant.
@Wow: This isn't about priorities, it's about distraction. No matter how important something is, you can be distracted by something urgent but less important. Your brain simply won't let you consider something urgent if you're dealing with it every hour of every day. It happens on battlefields when your own life is at stake, and it happens to even the best parents. Indeed, parenting itself is an exercise in maintaining one's composure in the face of fatigue.
If anything, your attitude would make you more susceptible since you figure you're immune. The best parents I know are the ones who are aware of their limitations and find tricks to circumvent them.
I was thinking OSI when I read that.
A few years back (I suspect it still exists) Volvo introduced a car with a car alarm system that could detect a heartbeat.
The purpose was to alert you in case you were walking to your car and someone was inside.
Perhaps a far-fetched scenario, but using that technology to alert you if there's a heartbeat in the car after you lock the doors may be quite wise.
"Wow, talk about priorities. If your child doesn't have your priority at all times, then I am sorry for you. Go to work and forget your child in the car? What is more important? Apparently not your child in that case."
Read the article before you comment. It has nothing to do with priorities; it's about distraction.
One of the tricks for use of a device like this will be figuring out a way to disable the alarm for those times when it makes sense to leave your infant in the car, while still preserving its function when needed. (For example, when pumping and paying for gas.)
Rework it as a temperature sensing device and sell it in pet stores, people would buy it.
Frankie B: The patent referenced in the article uses a weight sensor in the car seat, radio-coupled with a key fob, and some logic to send an armed(child present)/disarm(child not present) signal. The fob alerts when you go out of range without getting a disarm signal. The nice thing is that it is mostly passive.
Certainly a weight sensor could be fooled by a bag, but responding to false alarms of a bag left on the seat might be a small price to pay compared to hypertherming your kid.
If the radio-fob device will never be produced because of liability concerns, I was pondering how a person could hack their car to get an adequate result, and hi-jacking the open-door signal seemed easiest for my car: you get an idiot light telling you you have a baby on board, and the same old unusual beep (not secure chirp!) when you try to lock up.
I wonder if a proximity sensor would work for this?
Just have a keychain ping the carseat to see if it's still closeby. If not, your keychain starts yelling at you. Remembering to turn it on could be an issue, as would battery life. Maybe it could self-activate when the carseat it buckled?
If what you say is true (and it might be), it is commendable, but I do have trouble believing it. Do you keep your door open for your teenagers in a driving rain? What if you are getting into the car and your spouse is helping your child get out of the car on a busy street? Do you drive a mini-van that has a handle next to the rear door so it's unnecessary for your child to grab for the roof support?
Again, I'm not calling you a liar, (If you've never slammed your children's fingers in a door--good for you!), I'm just making the case that not all accidents are due to gross negligence, and both vigilance and circumstances are involved in accidents.
I remain a very,
That is one of the most intelligent ideas I've heard. It would not only help avert this type of situation, it would also reduce traffic accidents, environmental side-effects of millions of cars on the road, and decrease the stress factor of "rush hour". While not all of these are necessarily safety/security related issues, they touch all our lives in one way or another.
What people fail to realize is _no action/thing_ is an island. Everything touches everything else through ripples(side effects) it makes in existence.
Will it cure the underlying cause of forgetfulness, rushing, whatever; no. Such things can only be cured through a cultural/sociological reprogramming scheme whereby people are conditioned to align their priorities, and to constantly use critical thinking and environmental assessment to ensure all is well.
This is why pilots have to use checklists and why, when they don't bother, a minor distraction can result in a very expensive gear-up landing or similar mishap.
It ought to be possible to hook a "your child is still in your car" warning system into OnStar and similar systems, although I suppose GM have bigger things to worry about right now.
Speaking of silly little systems, my wife laughs at me because I never close a hotel room door behind me when I leave unless I am holding the key in my other hand. Mind you, I've never locked myself out of a hotel room.
Lo-tech...Take all the padding out of the carseat so that our little boo-boo-monkey can't fall asleep.
Hi-tech...Add an annoying noise activation module to the OnStar..."You seem to be crying, would you like me to call your negligent parent on his cell or just roll 911?"
The problem with both implementations is complacency due to false positives. If you get alarms continuously (idiot light, continuous beeping while driving), you learn to ignore them.
Trust me, it _sucks_ to never be able to leave a light bag on the seat. I've learned not to, but others might just as easily learn to live with the beeping, meaning that they wouldn't notice when they exchange the bag for a kid.
The idea of binding it to a key fob is better, but if it's self-arming, it would have the same issues. If it's not self-arming, then people would forget to turn it on. There's no easy solution for this, the brain can learn to ignore anything. Remember, most of these accidents are caused by a change in routine. People who drop their kid off every day don't suffer from this...
This is a topic that inspires a lot of anger -- for reasons the article itself describes. I'm asking in advance that everyone be respectful of the parents involved, and refrain from knee-jerk condemnations. If after reading the *entire article* you still think the parents involved are more culpable than it suggests, then you can argue that, but do keep it respectful, and make sure you are actually engaging with the arguments the article makes. Thank you.
@Unix Ronin: "But it's absurd that we're spending that kind of money per accident on rare events when there's so many far more common causes of death — exposure, auto accidents, drowning — that we could be addressing far more cheaply."
That isn't how it works. When actuaries say "per life saved", they're already figuring in the frequency of the life-threatening event. So an auto-accident prevention mechanism that cost $1M per life saved would turn out to be much more expensive overall than, e.g., a toaster-electrocution prevention mechanism that cost $1M PLS. From some perspectives they would be equally valuable mechanisms.
"Even today, with my teenage children, I still ensure they have safely entered (properly buckled in) or exited the car/van before closing my door."
Which is one of the reasons that people need systems to remind them. If your teenager is not capable of learning to keep his hand out of the door frame while you open and close it for him, what is going to protect him when he's a grownup? Will you be there to point out he forgot his child in the car as well?
I guess you don't need such a device. Obviously, it could never happen to you.
I think proximity would be the way to go for this situation, but instead of attaching the sensor to the seat, why not just keep it attached to the child?
OK, I know there would be issues with this, but it seems like the simplest thing, the disadvantages and pitfalls of which would be most obvious and understandable to distracted parents.
ps. anon at 3:31 was me too.
pss. @wow: for shame, madam
I like the string idea, simple, cheap and effective.
The onstar quote is quite funny.
On another note, I wonder how hard it would be to DIY the sensor those guys made. I mean, reading the make blog (www.makezine.com), these guys use that arduino (sp?) to build a ton of stuff. If they open sourced their design, could folks build their own easily and cheaply? Could DIY kits be built? Would a DIY kit lower and/or remove liability concerns?
I don't know how the patents and legalities would work, but just wondering....
@mcb, that OnStar thing is classic.
Please don't pile on "Wow," either.
RE slamming fingers in doors; it's possible to always pay attention. I grew up on a farm in an era when there weren't many safety devices. I learned at a very early age that there were mechanical things that would kill me in an instant if I didn't pay attention. It focuses the mind wonderfully.
Whenever I'm operating equipment, whether closing a door or starting a chainsaw, I do a check of what's in the way, and what's in the area. It's just a habit that you have to develop. I've actually grabbed doors that people were trying to close and kept them from closing them on other people's fingers, etc.
Now thinking about it, any attempt at a solution is pretty idiotic. It's going to add expense for a very rare event -- and you will continue with problems that are almost exactly alike. Plus you'll get the constant beeping because either: you immediately beep on some sensors that you're leaving the car, or you risk in a hurry not getting the signal. Now, anyone who has children knows that the last thing you do is unstrap the kids -- and you don't want the child awakened by beeping!
So you have a weight sensor -- what about forgetting the kid in the harness atop the car? Has happened! What about what about what about...
Why don't we just think like mature human beings that some risks are unavoidable, and move on with our lives?
At the same time, I can imagine that I could conceivably accidentally leave a child in a car. I imagine it's a horrific death, and I have to say that I imagine that death, and how it would feel if I'd done it, every time I read about these kinds of things. Reading such stories sometimes gives me nightmares.
Franky B, my particular car isn't as obnoxious as yours (or my father's car) My Outback only beeps on startup with open door, and chirps once for locking with an open door. Those seem OK, in my particular situation. I would perceive the mostly silent idiot light as a feature, not a bug.
My dad's car whines frequently if the passenger fails to buckle the seat belt, which isn't something that wouldn't help. The error those alarms have is that they do not have reasonable overrides -- If you could just push some button once (or buckle the belt) to clear the seat alarm and acknowlege that you have a bag in your seat, you might think your car _sucks_ less.
I do agree with you about the false alarms and passive arming. I'm saying that some car's existing behavior could accommodate much of the required functionality.
@Jess -- A portable, plain proximity alarm would be good: The "child is close" message would remind you the whole system is working, and the "child is no longer close" message would remind you in the odd cases that need reminders.
Code it up with some bluetooth fob and a smart cellphone, and it's done.
My low-tech solution, our day care provider gives us a call if we are more than an hour or so late.
True. There are plenty of much cheaper low tech solutions:
1. Put your office keys in the car seat. You can't enter your office without checking the car seat.
2. Have child care providers call if the child is too late. "Will Billy be in today?" "Gasp--I better check my car."
Granted, it's probably not fair to place responsibility onto the day care workers, but it doesn't hurt for them to know whether or not to expect the child. I grew up in the country, and a call from my school (in the days before cell phones) is how my dad knew to go look for my mom and found us with the car broke down.
That's actually a really good solution! I don't remember the daycare provider being mentioned in the article, but maybe I missed it.
May be better to make a common practice of putting your wallet in the carseat... You may not always be heading to the office but you usually don't go anywhere without your wallet.
@Kevin D. S.: "May be better to make a common practice of putting your wallet in the carseat... You may not always be heading to the office but you usually don't go anywhere without your wallet."
True. Then again, the most common reason for daycare is work and someone could work for hours without using their wallet.
Maybe the best practice would be to put something in the car seat that you will need at the planned destination. Office keys, wallet, grocery list, etc.
Simple solution. Put your wallet in your kids car seat. If you are like me, I always check to make sure I didn't lose it. In addition, they will get used to taking money from you at an early age.
Or perhaps a very simple chit system..
1) when carseat is empty - a trinket is in carseat
2) when child in carseat - place trinket on dash
Trinket on dash could provide visual clue to check the carseat...
I suspect that the list of things that kill more than 25 infants a year is a very, very long one.
It may well be possible to leverage the addition sensors and automation to do something more generally useful than prevent a vanishingly small fraction of the year's infant fatalities. Weight sensors in seats could trigger seatbelt reminders (and often do in current vehicles). Internal temperature sensors could drive the climate control system (and often do in current vehicles). An internal microphone (to pick up the screaming child, which is our species's build in child-in-danger notification) could be used as a part of a hands-free phone system (and often is in current vehicles).
The key fob notifier seems like the weakest link. Actual remediation seems like a better solution: either drop the windows to control the temperature (which has security implications), or start the engine and run the AC (could empty the gas tank, but it would work for several hours).
@JRR: Yup -- I have a child who's just turned two, and reading that story made me feel physically ill in a way few things have done for a long time.
The closest I've come to this was driving home with him safely sitting in his seat but not buckled in, due to distractions to do with his bag of stuff while putting him in the car. No bad side effects other than retroactively terrifying myself, and now I have another habit I need to get into to _consciously_ check every time, rather than being on autopilot.
(as for the solutions involving string et al, I don't think they'd work -- it's easy to unconsciously do a surprising amount of things, and unclipping something is probably one of those. )
Everybody's lives are in a truly sad state by that standard.
There are only two kinds of people in this world: those who have made a potentially life-threatening mistake, and those who will.
People are fallible, and that's a fact. Blaming people for being fallible and simply saying "well, don't do that!" does nothing to save lives.
If your phone loses contact with the paired binkie, it starts vibrating/beeping madly.
I am curious whether that article helped any prosecutors see what monsters they have been.
Regarding "cost per life saved": it is worth considering from time to time that there's really no such thing as saving a life, only postponing a death. In the long run, the death rate answered to only one factor, and that's the birth rate.
In many cases, the number of people "saved" by a measure is a good heuristic, but we should ultimately be looking at changes to life expectancy and quality of life, not "lives saved."
Myself and a few people I know (and I suspect many of you) follow a "pat down" routine. When I step outside my apartment, or depart from a bus, or out of a car, I do the pat down - I pat the pockets of my jeans to ensure my wallet, phone and keys are there. It's subconcious and it's stopped me losing or leaving those things behind many times.
Perhaps when putting the baby in the car, put your wallet in the back seat too. Then when you do the pat down you'll remember the wallet and remember the child.
Instead of a wallet, why not put your cell phone in the car seat? That way, in addition to helping remember your kids, you won't be tempted to talk on the phone while driving.
The problem with a lot of these solutions is that they're self-selecting, and those who believe it will never happen to them won't buy/use them. That leaves us with a regulatory solution -- and I'd like to see a cost per life saved before formulating an opinion about that.
My car got stolen after I got a steering wheel lock -- because I once forgot to put the thing on.
I solved that problem by a simple policy: when I take the lock off the wheel, I put it in my lap. There is no possibility of getting out of the car later with that thing in my lap without noticing it.
For the kid in the back seat, put an article of the child's clothing in your lap.
(I've got ways of never forgetting keys, and never forgetting to mail letters.)
The article comments on legal liability for failure. Something is terribly wrong with the legal system when it prevents you from producing something that will save many lives because it won't save all lives.
@BadDad: "For the car seats, maybe a mirror that let's you see your baby's face while you're driving?"
When my kids were still in baby carriers, we actually *had* those on both cars. They were very helpful for being able to check on the baby by just looking in the rear view mirror - a godsend when you're travelling at interstate highway speeds and a distressing noise comes from the back seat.
There's plenty available - just google for "baby car mirror".
@Ted: that's horrible! You may save their lives every now and then, but by giving them a cellphone before they even know their numbers, you doom them to a life tied to their cell phone!
@Filias Cupio: agreed, there's a classic flaw in our thinking - CYA always wins. No one is willing to think through the risk assesment, they'd rather assume a system is foolproof and sue if it wasn't
What we need is a complete deregulation of safety features that protect the occupants of a car. Let the owner decide if they prefer airbags and their kid in the back seat where they might forget them, or no airbags and no risk of forgetting their kid.
Safety regulation can protect those outside of the car (bumpers rather than pointy spikes etc).
A car seat with cheap sensors and a 9v battery (and/or car-lighter plugin) should be able to
1) detect the presense of a child
2) monitor the child's temperature along her back
A simple pattern of a child's presense , coupled with a steady rising (or falling) of her temperature to unhealthy levels should be easy to detect.
The question then becomes: what can the seat do when danger is detected?
Perhaps a 1-minute warning chirp (with a voice-activated "reset" command and/or a reset button to reduce false-alerts), followed by some sort of escalation, until reset button is pushed:
1) A shrill smoke-detector alert.
2) A call to the parents' cell phone.
3) A signal to trigger the owner's car's alarm (if possible)
4) A 911 call to local authorities.
If such an option were built into model A of a child seat and not model B; all else being equal, I'd spend $20 - $30 more on it.
Megan McArdle at The Atlantic posted on this last week. I commented with a simple device:
For someone even slightly handy with electrical stuff, attach a radio or mp3 player to the car seat and run a wire along the harness so that the player plays anytime, and only when, the harness is fastened. As long as you don't let the batteries run down (or start ignoring it on a long trip) you will always have a reminder that the seat is buckled up.
(If you are a little handier, you can wire it into the car's electrical system and don't have to worry about batteries. Don't just run a control wire from your car radio though, it's too easy just to turn that off.)
Posted by billswift | March 10, 2009 7:31 PM
It was very hard for me to read that article. I almost didn't, thinking it was going to be harrowing but pointless. The explanation was enlightening although I wonder if I was willing to grab onto any explanation that was vaguely plausible. (I remember when the pediatrician did this and couldn't figure out how. Children were ze's job, etc., etc., etc.) Even if the explanation isn't accurate a description of the common circumstances will help me be more on guard.
We have the mirror setup that BadDad suggests. They're available mail order from BuyBuyBaby, among other places. Useful in a number of different ways.
I like the office keys/trinket/stuffed animal idea. Nothing's perfect but this is practical, easy to do, and inexpensive.
PS - I, too, never close a door unless I'm touching the key. Works for me.
"When one of these cases occurs, the demonization of the responsible parent is amazing."
Part of it's dispositionism-- the assumption that everything people do is an pure expression of their true intentions. The human brain comes with this built in, though it easily accepts mitigating circumstances for people you know and/or sympathize with. (E.g., that cashier was rude because she's a nasty, loveless person, but your friend snapped at someone because she's having a bad day. Some random person who fires a gun in anger is a reckless or criminal danger to society who should be locked away for life, but your favorite football player who does it is really a nice guy who fell in with a bad crowd.)
To this day I still don't understand why we bother with airbags. Five-point safety harnesses are safer, and unlike airbags don't move your children to the back seat where it's easier to forget them.
Surely the easiest solution would be to eliminate airbags. Or, as a compromise, make it easy to turn the front passenger airbag off so children can sit in the front seat without worries about airbag-induced injuries.
Do you have a link on five-point harnesses being safer than airbags? I'd like to read more.
I imagine the major difficulty with five-point harnesses is just that they're too inconvenient for most people and it would be impossible to enforce due to mass disobedience. As for disabling passenger airbags, that seems to be happening now. I've seen many newer cars with a button for turning them off.
It seems to me that the best thing to do about this risk (and some other related risks such as leaving children too close to air-conditioners in summer) is to have a monitor for their vital signs connected to a GSM modem and a GPS. I've written more about this at the above URL.
In regard to the comments about disabling airbags. The 1999 model VW Passat that I used to drive allowed me to turn off the front passenger airbag. However that would do no good in this regard as there were no attachment points for a baby carrier in the front, due to engineering issues related to baby carriers there is often great difficulty in putting baby seats where you might want them without a significant redesign of the vehicle.
Turning off the front passenger airbag seems most suited to having a child in the 6-12 age range in the front passenger seat. So if you have two children who are less than 12 one can be in the front to avoid fights.
Anyone who thinks that this couldn't possibly happen to them probably has an unrealistic belief in their infallibility. Keeping the child in front is likely not a good tradeoff: the back seats really are safer, since high-energy impacts are more likely to come from the front of the car. I really like the idea of placing some essential item near the child seat (wallet, etc.). Yes, you might forget to do that, but it's still an extra "layer" of security. Concerning a sensor system that could detect when the child is left in the car and take appropriate action: liability concerns needn't rule out such systems. Consider automotive airbags, which can cause all sorts of injuries under the right (wrong?) circumstances. Yet, they've obviously been quite successful.
While we're all tossing around possible engineering solutions, here's my two cents:
I like mithrandir's point about leveraging systems in the car anyway, that makes the economics much better.
So I would:
1. Make the air-conditioning system totally electrical instead of driven by the motor.
2. Add photo-voltaic cells to the roof of the car.
3. Forget about trying to sense that anyone's in the car. Just *always* keep the car from getting *too* hot.
The system doesn't need to have enough power to run the AC full blast and maintain perfect 72 degrees for hours. It just needs to be able to keep the temperature in the nominally-safe range. On days that are hot enough to kill your kids, there is a good chance that there is enough solar energy to do the job.
Even without kids in the car this would be a convenience/luxury feature. The car might not be the perfect temp when you get in, but it will be a lot more comfortable and quicker to cool to ideal temp.
Looking through my files and thinking back, I have no evidence to back my claim.
I was assuming they're safer, but upon further research neglected to take the overall picture into account when forming that opinion: While five-point harnesses are good enough for race cars and baby seats, race cars also have roll cages and drivers wear helmets. So, perhaps not.
With between $1 million and $10 million per life saved, the USA could fund a paid maternity leave program that actually lets mothers stay home to care for their newborns for longer than 6 weeks. Expecting the parent (and especially the mother) of a newborn to return to full-time work that soon is just insane. By 6 weeks postpartum, you are lucky to be back at the level of a fully-functioning human being. I know I personally was not anywhere close to functioning by 6 weeks postpartum, and not for quite some time after that, either.
Apart from a few very emotionally significant events, I remember almost nothing from the first year of my daughter's life. I pretty much sleepwalked through it on instinct. I went back to part-time study when she was 7 months old, but I had completely forgotten until recently that I had gone to school during that time. I have no clear memories of being at school or of what I studied. The only way I know I studied then is because it says so on my transcripts. I was not mentally ready to return to full-time study and part-time work until she was about 14 months old.
Thankfully, my daughter was a high-needs baby, which made it almost impossible to forget she was there and have an accident like this, but I can easily understand how it could happen.
>> Just *always* keep the car from getting *too* hot
This will not work when temperature is about +35C outside. In some places in world it's usual condition during the day.
You have to detect presence.
How about opening the car window a little bit.
You may remember to close it as it would be visible from outside and if you don´t the danger of hyperthermia inside the car would be decreased.
I don't know whether it will help with overheating enough, but the specs on the new Prius have a roof solar panel that supposedly will run a fan to bring "in outside air, ventilating the cabin close to ambient temperature". They aren't promoting this as a safety feature, but it made me think that it could potentially save lives.
I have to find it interesting that not only was there no discussion or even mention of the economics involved in the article, there hasn't been the slightest one in the 82 comments left here, either.
Also, some of the suggested solutions could almost have been entries for the last Movie Plot Threat contest.
Supposing there's an average of 17.5 deaths per year from this problem, and children need to be protected for 2.5 years (which is about what the article seems to indicate.)
That's what, 9.3 million children needing protection at any given time? - and at 5.5 million each life saved, for a perfect solution, that's $96 million. 17.5 lives for a hundred million dollars? That's completely out of whack, because of the more effective ways that money could be spent. I wonder how long you could run a whole hospital for that.
Anyway, you have $10 for each kid. For 2.5 years of protection, the batteries alone would be too much for a fire alarm like detector. So...... Opening the car window a bit sounds about right, except you will forget to close it, and the average cost of which over 2.5 years might be more than $10 as a result of theft and rain damage.
If it takes 15 seconds to put and retrieve your wallet, over 2.5 years, I dunno, twice a day, that's $30 worth of time, at minimum wage.
And arguably, now that you've read that article, it could be worth it to eliminate the stress involved.
It's called a car alarm folks.
1. Lock - mine autolocks after a few mins of inactivity
2. Alarm detect - internal and external sensors are enabled once the car is locked.
3. Alarm response - internal movement, or external rocking sets it off.
4. Known vulnerability - An override button lets me lock the the car without enabling the sensors; it's for nipping out and getting a parking ticket (say). This could provide a distraction window.
There are other threats with higher probability and similar impact. Stairgates being one, and shear infant ingenuity being another. If we think adults are bad a risk assessment, toddlers are off the scale!
What a sad article - reading it at my desk at work there were bits of it I couldn't face.
I think it is wrong to prosecute as long as it is clear that the child was left in the car unintentionally. Not just because it serves no purpose in terms of prevention or punishment, but also because it puts the parent in the strange position of somehow defending what is indefensible.
Lots good low tech solutions suggested here, but I doubt many people will use them as that means admitting to themselves that they could forget their child. Though I wonder how many 'near misses' there are each year, where a child is forgotten, but not long enough to suffer harm.
I would expect a carer to ring if a child didn't arrive, but that is no guarantee of safety.
Oh and any teenager unable to take some responsibility for their own safety in the family car is going to be at real risk out there in the big bad world.
Cut the knot. Stop driving your car. There aren't many hyperthermia deaths on buses or trains.
I am 100% sure this parents do make the best possible carrying babysitters. Nothing will happen to another child in their presence. Am I right? Were where any repeated offenses?
What about the most simple and cheap of solutions?
I will sell you a pack of numbered balls which are attached to a sucker cup. When you get into the car (and you're less likely to be on auto-pilot), pick up the one which shows you how many passenger you could be leaving in the car and stick it to the windshield.
When you get out, I would say that most people will look at it and *realise* - even if they are on autopilot.
I would be prepared to design clip-on versions for your phone, purse, belt - all for the same, low $9.99 price.
For the extremely forgetful parent, there will be a cheese-wire version which hooks around your neck and connects to the drivers door ;-)
Obviously its the child's fault for not screaming its head off.....
(well if the parent had been a big bank, a large corporation or the government then that is how it would have been handled).
btw - anybody who doesn't think "there but for the grace of God (chance) go I" when they read a story like this really worries me
@ James Woods
I was just about to post similar numbers. I think your 2.5 years of protection is actually a little low. From page 5:
"In a collage on Fennell's wall are snapshots of dozens of infants and toddlers, some proudly holding up fingers, as if saying, "I'm 2!" Or "I'm 3!" The photos, typically, are from their final birthdays. "
@ Unix Ronin, yt
That's $10 million for each life saved. The problem is that there are at most 25 deaths per year in the US. If we could save all those lives then we have at most $250 million which is only about $1 per registered vehicle in the US.
I'm really confused by this risk analysis.
From a parents perspective: $20-$40 more to ensure that this potential fallable aspect is prevented without having to rely on other mechanisms that I might forget to do. I'd gladly pay it.
From a manufactures perspective: Of the average number of deaths per year, only a percentage of those are in a specific brand of car seat (not sure how many exist). My margin per sale is increased as a result (most likely -- I could probably charge 200% per cost of implementation). And (IANAL), contributory negligence should lower any potential lawsuit payout (how many car seat manufacture lawsuits are there if a baby is killed while in a car seat?). So it would seem that I should be able to make money even IF there is a chance of this happening.
This whole dollar value per life saved only has relevance if that money were actually going to be doing something. For instnace: if the Government were spending $10M per life saved then you could say the value could hold true. But the government wouldn't.
Can someone help me understand this risk analysis. Might help me do a better job at work!
I also wanted to suggest that regulation that saves the life of a child also saves the parents of that child as well. So I would count a measure as saving 3 lives for every child saved. So a regulatory solution with a focus on children, as opposed to all registered vehicles, would have a budget of around $30 for 3.5 years of protection.
The reason no one wants a safeguard against accidentally leaving children in the car on a hot day is because it is not accidental. Who is going to admit to deliberately leaving an infant in the car on a sweltering day? I'm sure that there are a few oversights, but I would bet most of the people who 'forgot' are covering themselves with a convenient story. it's the difference between no charges and negligent homicide.
That would be my expectation as well: that five-point harnesses are safer in race cars being driven at high speeds by professional drivers, but that airbags win in cars driven on public roads by morons.
We don't have children yet, but we've gotten ourselves in the habit of never carrying anything in the front seat -- that way we always have to open the back door of the car *every day* to get the laptop, the lunch bag or other articles. We've trained ourselves well enough that we open the back door of the car now even when we don't have anything with us -- autopilot.
"There are only two kinds of people in this world: those who have made a potentially life-threatening mistake, and those who will."
Very true. My dad has always been anal retentive about safety, especially in his workshop. It annoyed the hell out of me growing up, and I still make fun of him for it occasionally (jokingly, of course.)
Even with that, he almost cut my hand off with a chainsaw last fall. I was helping him clear some land, and we had some small trees to cut into firewood. We were very careful about placement, making sure nothing was near the chainsaw, and making sure our legs were out of the way if it swung around after going through the wood. And chainsaw usage isn't even a habit for either of us; he had to borrow one for the job because neither of us own one. Still, we made one cut with me holding the log less than a foot away from the cut. After the cut end fell away, we looked at each other wide eyed realizing what we had just done. Needless to say, we were both VERY anal retentive for the rest of the job.
Do you jump to conclusions and fail to read articles? Or do you just enjoy trolling with inflammatory nonsense?
@bill at March 18, 2009 6:03 AM,
Did you read the article? Here's an excerpt:
"Then there is the Chattanooga, Tenn., business executive who must live with this: His motion-detector car alarm went off, three separate times, out there in the broiling sun. But when he looked out, he couldn't see anyone tampering with the car. So he remotely deactivated the alarm and went calmly back to work. "
When I first read this (on a parenting blog over at the Houston Chronicle), I *knew* the"hang 'em all" crowd would weigh in. Sure enough, they showed up, in full vitriolic regalia. Much worse there than here, actually. Though some of the comments seem eerily familiar.
As I said there, what probably saved my kid (who can fall asleep taking a drive down the block) was a combination of two things:
1) I am *extremely* OCD, and tend to check things... then check them AGAIN... and AGAIN, AND AGAIN!!!! Then check one more time, to be certain. Some days, it takes like 30 minutes to get out of the garage.
2) He grew *extremely* fast, meaning he was bigger than any infant car seat made well before he turned 4.
Would I have ever knowingly left him in the car? Hell, no - where I live, that is pretty much a guaranteed multi-year all-expenses paid trip to the state pen. But, by accident? Because something intruded on my concious thoughts at the wrong moment? Thank goodness, I will never need to find out. But, as a person who actually managed to lock not 1, not 2, but *3* sets of car keys in the car one day (I had my mother bring the *fourth* set from home so I could open the door), it was never too far from my mind.
@Jeff - you wouldn't believe just how often a parent admits to "stepping out of the car do just buy one thing at the store" while leaving the kid in the vehicle. With the windows rolled up, and the door locked. Not to mention the engine (and therefore the A/C) turned off. In the middle of the afternoon. In August. In Houston, TX.
These folks - I can understand making an example of.
I know exactly one person whose baby who died from hyperthermia. A woman I know, whose name I will keep anonymous to protect the guilty, planned a night out dancing. Her baby sitter cancelled, so she took the baby and went out anyway. Yes, she left the baby in the car in a bar parking lot until 3 am. She drove home under the influence, went inside, forgetting the baby, and fell asleep. She woke up around noon and found the baby in the car dead.
If you ask me, she was grossly negligent. She was never charged with a crime. I think most reasonable people could agree that this is far different than someone who gets distracted and goes to work forgetting about the baby. Yet, in the linked article someone was charged with second degree murder.
I have every sympathy for an otherwise good parent who will live with the guilt of a horrible accident for the rest of their life. Gross negligence, like the case where the acquaintance's baby died, is a far different matter.
There are lots of good ideas here but I didn't see mention of adapting these existing solutions:
School buses have an alarm that goes off if you don't push a button at the back of the bus within so many seconds of stopping the engine. Maybe this could be made part of the car alarm system? The ordinary car alarm with a motion/sound sensor inside the vehicle mentioned above might be better. They work in different ways though.
K9 patrol cars have gadgets that open windows or turn on AC to keep dogs alive. I suspect these have the problems mentioned above, but since the systems are in use some kind of solutions have probably been discussed.
You're absolutely right, there's a huge difference between forgetting your baby because you're human and an incident at the drycleaner's distracted you at precisely the wrong moment, and forgetting your baby because you went out partying and were too drunk to take care of your child.
As with anything, the circumstances surrounding the ultimate action are as important as the action itself.
Hook your keys to one of those retractable keychains. Leave the keychain attached to the baby car seat.
This happened in Israel a couple of times. During the summer. It wasn't hypothermia that killed those babies. Think cooked babies.
@Assaf: "This happened in Israel a couple of times. During the summer. It wasn't hypothermia that killed those babies. Think cooked babies."
No one has said the babies died of hypothermia (freezing). We're talking about hyperthermia (baking).
Simple mistake. I misread it the first time as well.
The solution: don't sell it as a child monitor. Sell it as a pet monitor. Let people "abuse" the agreeement, which specifically calls for only using with white mice, or other such nonsense.
Get PETA to fund it. After all, people are animals too!
That is a damn good idea, Republicrat :)
No need for any /new/ sensor (like by the NASA guys). These things already exist. Usually used to let you know when the kid gets too far away from you in a crowd (at the zoo, mall, wherever) but they trigger on distance. You have a unit, the kid has a unit. Distance too far, alarm goes off. They EXIST. The trick is to a) let people know they exist, and b) convince 'em to use 'em with the non-mobile children in the car seat (they are usually used for active kids) -- this would be a slightly different application than they are currently marked for ... and from the other comments here, most people aren't even aware of 'em for what they /are/ currently marketed for!
google: child distance alarm
Amazing that the underlying condition is completely overlooked by us all. This becoming an issue in the last two decades correlates with the rise in both parents working outside the home. Not only is the false economy that this has created crumbling, but it is having these nasty effects. Time to go back to basics?
To me the issue is that there is liability attached to trying to solve a problem.
Why should the makers of the child alarm have liability? Shouldn't there be a special category of immunity for products like the child alert? Not blanket immunity - if the device is defective, its not immune. If you don't change the batteries, or you forget to use it, no liability.
IF seats can sense the weight of a passenger and turn off, or adjust airbags, based on that weight, why not a sensor for a baby seat? One that measures weight with the baby seat empty, one with it full.
If you close the doors without inserting the key in the ignition (with a 15 second delay maybe) with the seat at "full" weight, the car alarm goes off. If you don 't respond in 1 minute, it turns on the AC or heat, as applicable, and "on star" or whatever, calls emergency services.
This becoming an issue in the last two decades *also* correlates with the rise of the Safety Culture. All kinds of problems which existed for ages are only now being raised in the public consciousness, because before the Safety Culture they were accepted as simply being part of life.
A couple of dozen children dying each year in a country the size of the US is a non-event, and I see no evidence that this number has grown out of proportion to the number of children in the country. Until and unless such evidence is presented, my opinion is that the most reasonable assumption is simply people blowing things out of proportion because they have no sense of scale.
Now, as to why the Safety Culture is so prevalent these days, that's a more interesting question and a harder one to answer, but I doubt it has much with the rise of women in the workplace.
I don't think I saw this specific solution mentioned yet. The feature of keyless ignition is starting to trickle down into non-luxury brands (e.g. infinitis have had it for while, this year some nissans are getting it).
So instead of putting your office keys (which you may not have if you don't regularly open the office building, or may not apply if you are going shopping and just forgot that you packed the kid instead of leaving it with a sitter) in the carseat with the kid, put your CAR keys in the carseat with the kid. You can still start the car as long as the keys are in car's cabin. But if you get out of the driver's seat without the keys the car will beep at you, and you will have a tough time locking it up (my car has a major hissy fit if I try to lock my keys in the cabin).
@Posted by: Wow at March 17, 2009 2:38 PM:"Wow, talk about priorities. If your child doesn't have your priority at all times, then I am sorry for you."
While I would not have used those words, I do agree with the sentiment.
There are a lot of posts on this thread where the issue has become somewhat objectified and abstracted, I think there is a danger of losing sight of the moral and emotional imperative here. A parent has responsbility to care for their child: even in a dispassionate analysis, these examples are not accidents, nor were they a split second decision like what may occur in a traffic accident - they were easily in the parent's control to have avoided. So it does fall under the header of negligence.
@Posted by: Franky B. at March 17, 2009 2:46 PM:"If you can't, even accidentally, concentrate on something other than your child for even a moment, then I'm sorry for you! Must be an insanely stressful life..."
That wasn't what "Wow" had said. I also don't see how having your children as your first priority at all times naturally means an "insanely stressful life" - is having your children as your priority not part of the definition of a good parent?
@Posted by: Bruce Schneier at March 17, 2009 3:05 PM:"Read the article before you comment. It has nothing to do with priorities; it's about distraction."
That's not really an argument, I've read the article and I disagree with you - it has everything to do with priorities. Yes, you can get distracted but your priorities dictate what your attention is focused on after any immediate distraction is dealt with.
i) On reading the article, one of the first things that caught my eye was the references to parents using mobile/cellphones while driving. In the UK that in itself is illegal, and for very good reason. For example, only last week I was almost hit at a pedestrian crossing when someone talking on a mobile phone drove through a red light - luckly the car was bright pink and I'd noticed in time to jump back. So the parents who are driving while talking on mobile phones are already putting themselves, their children and others at an increased risk. Which is a perfact example of selecting convenient communication at a higher priority than the safety of others.
ii) Going back to the issue of distractions... say you are walking down the street to work and you meet someone you know. You have a brief conversation and you continue to go to work. Why? During the conversation you're probably concentrating on what you are saying, but as soon as the conversation ends your brain re-engages with what your previous priority was, i.e. going to work. If a parent's priority was their child, then they would naturally return to thinking about their child after a brief distraction.
iii) All this is ignoring the external cues like "why is my left hand empty, ah, not carrying the baby seat", or "has my baby settled in ok at the nursery/daycare before I leave for work".
Despite my views on the subject, I do acknowledge that it is a problem that needs a definite and focused solution.
In the UK, the government is constantly enacting new legistlation to regulate driving, from speed cameras to recent changes in childrens car seats. The exams to get a driving license have also been made more stringent over the last 20 years, and there are constant advertising campaigns on driving safety. To the best of my knowledge, this does have an appreciable effect on reducing the road traffic deaths.
I see no reason why similar measures cannot be enacted to protect against children dying of hyperthermia: whether that be a technical solution, awareness campaigns, or whatever. In my view, the problem has the potential to be dramatically reduced if there was enough of a will to do so.
"Despite my views on the subject, I do acknowledge that it is a problem that needs a definite and focused solution."
Why, exactly? It's literally a one-in-a-million problem, on the same order as lightning strikes and shark attacks. Surely our energies could be better used elsewhere, no?
I think the simplest and cheapest solution would be to require daycare centers to call when parents fail to drop off their children.
They already have home, work and mobile phone numbers for parents. (And if they don't, they should).
No extra cost and it would prevent cases like that of Miles Harrison and others.
Also: New cars are often fitted with a mobile module, that transmits the car location, to prevent car jacking (lojack). How much extra would it cost to add a sensor, that detects that someone (child or pet) is in the car and phones 911 when temperatures get dangerous?
(It could also lower the windows and/or turn on the air conditioning if possible).
It wouldn't just prevent the death of children, but also non-fatal neglect, and that of pets, over the lifetime of the car. Bundled with a lojack device, it would also prevent car jacking.
Taken together, the cost/benefit ratio becomes much more sane.
I was left in a parking lot like this when I was a child. It left me hypersensative to heat and I never trusted my parents ever sense. Afterwards i've always called them by their first names.
Restraining kids in car seats is absolutely horrible.
that is why this thing always needs governemtn backing. Just like airbags, it should be extensively tested by the government, and if found effective mandated as required equipment. Safety equipment rarely succeeds if not mandated, sadly. This also gets rid of the liability.
Sadly I don't see this type of things happening, anything like seat belt and airbag testing etc. Even though they undoubtably saves countless lives, these basic things would have never been implemented if it wasn't for Nader and the lucky publicity of General Motors' harassment of Nader
Someone alluded to pilots using checklists earlier, I think its a good example. Pilots have a saying "there are 2 kinds of pilots - those who have landed gear up and those who will" - the implication being that sooner or later everyone will make a very costly error.
A system to prevent this type of death would not be to prevent 25 deaths a year. It would be to prevent 25 per year times 2-5 family members suffering times the number of years left in their lives.
It seems to me that a child seat that:
1) had a weight sensor in it OR a sensor on the buckle
2) came with a fob
3) caused the fob to beep when it got more than 20' away from the seat while the seat detected occupancy
would not be expensive to mandate. And due to the "it wont happen to ME" nature of the event it will take a couple of times for well-publicized "my child seat had that thing but I never bothered putting the fob on my keyring" deaths before people carry it with them. But after that it would probably eliminate 2/3 of the 25 annual events.
Or we could admit that airbags arent that dangerous (especially now that they are smarter) and let people put the kid in the front seat again. That would probably be more lives saved in the long run, but would require government to reverse course and government never admits it is wrong. (although with a major regime change it can blame the previous administration)
The safest place for a child in a car is in the front passenger seat, facing backward, with airbag deactivated. There are however many uninformed/old recommendations for other alternatives that are hard to get rid of.
Car manufacturers know this and have recommendations for safe airbag deactivation (/removal, depending on model and make) for all newer cars.
The primary reason for placing young children in the front seat is that there is more legroom there for them to sit facing backwards, thus making it possible to use such a children's chair a longer time. Most 4-5 year olds can sit backwards in the front seat, while it is getting cramped in the backseat in many cars before that. Young children have a bad head weight to neck strength ratio and handle the common case of head on collisions very badly. They should therefore sit backwards as long as possible!
The added benefit of making it much harder to accidently forget your kid is a bonus.
@to all people who think automatically turning on heat/AC is a good idea: Bad idea - it would require starting the engine for either one (except some hybrids which have a limited battery A/C capability) and have the engine start when no one is around raises as many safety issues as it solves. Starting a fire. Accidentally going into gear (or being in gear if its a stickshift) and running over someone else or going under water. CO poisoning.
A few kids dying each year after their parents screw up in a country the size of the US? People, get a grip. The world has bigger problems than that.
If you really want to save the children, maybe you could stop sending your military to places halfway across the world. You could have saved half a million children right there by not invading Iraq for example.
Use the money you save by sending AIDS medicine and condoms to Africa. Should be good for another couple million lives.
I sent this article to my father (who raised the three of us kids to adulthood), and he told me that it actually happened to him on two occasions, fortunately without harmful results: once he forgot to pick me up from preschool after the morning class, and arrived hours later, terrified, to find me happily playing with the afternoon class. And once, eight or nine years later, he forgot to pick up my sister from elementary school (remembering about an hour after he was supposed to be there). My father is not a forgetful person, but I can even recall seeing him spontaneously remember something he had forgotten to do, and how shocked he looked that he had forgotten to do it. Not many times, maybe a handful of times over a 15+ year period. Memory can be fallible, and in these very infrequent but statistically possible cases, awful tragedy can result.
Bottom line: If you think the terrible scenario described in this article could never happen to you, think again.
@Posted by: Michael Ash at March 18, 2009 11:58 PM:"Why, exactly? It's literally a one-in-a-million problem, on the same order as lightning strikes and shark attacks. Surely our energies could be better used elsewhere, no?"
Yet there's lightening rods and copper conduction strips put on most even moderately high buildings... go figure. Your example would have worked better if you'd picked something that isn't already taken seriously.
I think the example works perfectly fine. We put lightning rods on tall buildings, because otherwise they might get struck by lightning and set on fire. We don't leave our infants alone in the car, because otherwise they might die from hyperthermia. Due to these measures, deaths due to both lightning strikes and infant-car-abandonment-hyperthermia have been reduced to similar numbers, literally one-in-a-million events. Why the hysteria to reduce things even further?
"I am curious whether that article helped any prosecutors see what monsters they have been."
Probably no more than Barbara Rosenblatt's critiques in the Wall Street Journal of unconstrained pedophilia prosecutions.
"What we need is a complete deregulation of safety features that protect the occupants of a car. Let the owner decide if they prefer airbags and their kid in the back seat where they might forget them, or no airbags and no risk of forgetting their kid.
Safety regulation can protect those outside of the car (bumpers rather than pointy spikes etc)."
I agree. I recall that airbags were introduced under the rubric of "passive restraint," along with auto-deploying seat belts. I never owned a car with the latter, but rode in one my brother rented. Seemed to work OK, but I understand many people disliked them, and I think they are no longer a player. With most states (maybe all, if New Hampshire has succumbed) having adopted mandatory seat belt laws, I wonder whether air bags now qualify legally as "passive restraint." I'd just as soon do without, not being an "unarried, under-25 male driver," which I also understand is the only group for which air bags are cost-effective (again using an arbitrary standard, such as $1M/life), but my guess is that subsequent laws or the economics of automobile production militate against cars being offered with an air bag option. If I were likely to have more kids, I'd settle for an on / off switch. However, I expect that the law now mandates the child seat in the rear, without regard for the provenance, so the on / off switch would have no effect in this regard.
> ... to both lightning strikes and infant-car-abandonment-hyperthermia have been reduced to similar numbers, literally one-in-a-million events. Why the hysteria to reduce things even further?
I would hardly call the invention of a cheap alarm "hysteria." But the reason is the claim that circumstances have recently changed to significantly increase the risk of infant hyperthermia in cars; it seems to have quadrupled over the last ten years, and continues to rise.
True, it is still a fairly rare problem*, but as the effects are devastating and the potential remedy cheap, it seems a measured and logical response. It may not be optimal, but it certainly isn't hysterical.
* (The figure of 15 to 20 per year seems to be dated, by the way. Last year it seems to have been 36 with a 10 year average of 30 / year.) Whether or not it is a one in a million event depends on whether you look at the population as a whole, or only the at risk population. For the total US population the risk is about 1 in 8 million per person per year for deaths, or about 1 in 2 million per person per year if we include serious non-fatal injury. However for the life exposure (nearly all risk in ~3 years) of the at risk population, it is closer to 1 in 120,000 per person for deaths, or 1 in 30,000 for serious non-fatal injuries. If the regulatory goal is $10 million per life saved (or equivalently, thinking that your child's life is worth $10 million), it would be rational to obtain the alarm if it costs less than $80.
> ... it would be rational to obtain the alarm if it costs less than $80.
Actually, that is only the case if the alarm can only be used for one child. If a family has 2.5 children, and the alarm lasts long enough to be used to protect all of them, then it remains a rational investment at prices of up to $200.
When I refer to hysteria I refer to statements like this, found in the comments of the linked article: "It's both a tribute to life and an indictment of our materialistic culture where innocents suffer because profit trumps their safety."
An alarm may or may not be cost effective. A lot hangs on just how cheap and reliable it can be made and, more importantly, just how much money it takes to make the lawyers go away. Frankly I'm not *that* interested in this aspect of the question, although it presents at least a bit of a technical challenge.
But when I refer to hysteria, I refer to commenters like the above, who think that this rate of deaths is so utterly unacceptable that it somehow means our entire society is morally bankrupt.
What interests me is the rise in deaths, and why this should be so.
Is it simply that the child is in the back due to "air bags" or is it something else (like stress caused by changes in working practices).
Also there is the issue of the childs age. Are these "children" actually babies or older?
The reason I ask this is that in the UK we have not only a series of quite tough standards for babies but also the mandatory use of "booster seats" for children bellow a certain hight (which is reached on average around 11 years of age).
The thought occurs to me "why not put an alarm in the childs seat" such that turning the engine off or opening the doors gives of a low level alarm if the child is in the seat.
The only mandatory things would be,
1, The alarm in the seat
2, The seat is fitted by qualified personel.
In reality the level of knowledge required is such that they only need to know how to wire a plug into the cars wiring in the same way as a radio or other entertainment system so not overly difficult.
To make it easier then a "wiring loop" for each rear seat can be mandated for every new car (with rear seating)
The only other question is how the alarm should escalate should it be initialy ignored (how about flashing lights and sounding horn even possibly progressing to unlocking the doors as well). In most cases in modern cars this would simply be a software upgrade to other systems (ie for dash displays and drivers consol etc).
The factors brought up by Jeff sound repugnant but are actually important when studying things like this: they are also a reason why the SIDS phenomenon is so difficult to study (because it's politic to blame an infant's death on "SIDS" and not "accidental smothering" or even "purposeful smothering").
Secondly, the understandability of a phenomenon has nothing to do with how legally or morally condemnatory we can expect to be about. Flying into a murderous rage and killing one's unfaithful spouse is highly "understandable", but that makes it neither legal nor morally defensible. It might be understandable how someone can forget they have a child in a car; but that doesn't make it alright.
I do agree that this is not enough of a systemic problem to demand additional regulation or a systemic solution.
A couple of comments:
- I think that almost all of the alarm/reminder mechanisms mentioned aren't going to be very effective. They either involve reminders you set up yourself (something important on the kid seat/buckle) or a machine. People won't use the reminders because "I wouldn't forget something as important as my child"... until they do. The machine/monitors would need to be perfect, or else they'd be immediately turned off as annoying.
- A solution should work as well for a dog as a child.
- People are at least as likely to intentionally leave their kid in the car (because they're asleep and I'll only be gone for 2 minutes - the distraction occurs in the store...) as they are to walk away having forgot them. The solution should work as well in both cases.
- Having the car start air conditioning is impractical for power considerations. Having it take measures to keep its temperature below a critical temperature by perhaps lowering all the windows an inch, starting a fan, or somesuch could be worth exploring.
Wrt children being safer in the front seat, I think there are other factors that should be kept in mind. I noticed with my kid that when he moved from rear-facing to front-facing in the back seat, there was a real temptation to turn and see how he was doing - exactly the sort of driver distraction most likely to have us both in an accident. The same thing can be true of an infant in the front seat - it becomes possible to try and adjust a pacifier without pulling over, which might ultimately be the most dangerous thing for both of you.
> 2. Add photo-voltaic cells to the roof of the car.
this is, in general, an excellent idea. I once thought about doing this when I owned a car and used to drive for days on end, with a laptop and a camera battery charger and several other electrical devices which tended to strain the DC outlet quite a bit.
nowadays, alas, I rent cars, so I'd need a *portable* roof-mounted solar power grid.
(Sorry, that "Anonymous" was me -- seems to be a bug, as I was "logged in" as shown by the follow up comment 3 minutes later.)
> But when I refer to hysteria, I refer to commenters like the above, who think that this rate of deaths is so utterly unacceptable that it somehow means our entire society is morally bankrupt.
Oh, fair enough, I quite agree with that. The problem is that most people in our society have little or no understanding of probabilistic risk, nor indeed any aspect of probability, especially as applied to large samples.
> Is it simply that the child is in the back due to "air bags" or is it something else (like stress caused by changes in working practices).
I don't think we can do much more than speculate. If the data has been collected, it should be possible to assess this by comparing deaths using the new style seats, to those who couldn't afford to upgrade and are still using an older style seat. However, there would be obvious confounding factors with income, which would relate to car accessories, type of job etc. as well.
While we are speculating, I would like to suggest that the increasing prevalence of car air conditioning may have several effects: the noise makes it harder to hear a child in the car, the cooler ride has encouraged the fashion-driven proliferation of dark coloured cars in sunny climes where they were once considered impractical, and it becomes easy to forget how fast a closed up car in full sunlight turns into an oven, even on a cool day.
> Also there is the issue of the childs age. Are these "children" actually babies or older?
The majority of victims in the last year were under 1 year old, with some up to 3 years old. (I have difficulty seeing it affect any normally developing child older than 3; my niece already winds down the window if she is getting too hot. She is only two and half years old.)
> The thought occurs to me "why not put an alarm in the childs seat" such that turning the engine off or opening the doors gives of a low level alarm if the child is in the seat.
I agree. The risk / cost numbers do not make sense if the alarm is delivered to the whole population, but possibly do make sense if it is targeted at people who have young children, and take them in their cars. Putting the alarm in child seats targets it effectively, as well as being the simplest and most logical engineering solution for detecting the presence of the child.
> 1, The alarm in the seat
> 2, The seat is fitted by qualified personel.
Disagree. This would double the cost and result in greatly increased non-compliance -- reducing the very small hyperthermia risk at the cost of increasing the much higher car accident risk. There is a simpler way to make sure people plug it into the accessories socket: ship it with the internal battery already charged, and make it bleat whenever the child-detecting microswitch is depressed but there is no power on the external power line. As this will occur either because it is not plugged in OR because the key was just removed from the ignition switch, it makes normal operation and nag mode the same thing, simplifying design and increasing reliability. (In fact, now I come to think of it, it seems likely that nearly all the suggested $10 to $20 cost is either a 12 V rechargeable battery, or circuitry for a step down charger for a lower voltage battery. All the rest -- a switch, a sounder (piezo buzzer with built in oscillator), a transistor, a couple of resistors, and some connectors -- would amount to only $2 - $3.)
@Posted by: Michael Ash at March 21, 2009 8:13 AM:"I think the example works perfectly fine. We put lightning rods on tall buildings, because otherwise they might get struck by lightning and set on fire. We don't leave our infants alone in the car, because otherwise they might die from hyperthermia. "
That's my point, there's a natural risk of lightening stiking a building - a strikingly low risk (sorry, I had to do it) - but conductors and lightening rods are still installed. The natural risk inherant with parents forgetting that their child is in the car is also low - but children are still dying because of it - so why not ameliorate that risk as well?
With your argument, we'd just hope that there's no lightening storms.
The answer that is foolproof and prevents this problem (and doesn't require the parents to remember to do anything) is to take something like the wireless baby leash that others have mentioned and put one part (child part) part in the car seat. Then run a wire through the buckle system. If the seat part detects that the parent part (which goes on the car keys for example) is no longer in range and that the seat is buckled, it will signal the alarm to go off. Since its fixed to the seat, the only way it can fail (baring actual equipment faults/battery drain/etc) is if the parent somehow doesn't take the parent part with them or if the child is in the seat but is not buckled in
Requires no modifications to the car whatsoever, just the seat. Doesn't require any weight sensors either and wont fail if for some reason you have shopping in the baby seat and not the baby)
For regular seats (and bigger kids), just build a device that closes over the seatbelt buckle (they already have devices that can alert when someone in the seat unbuckles the seat belt for e.g. kids or those with dementia or something) and use that to detect if the belt is buckled for the purpose of setting off the alarm.
My system would also work for dogs or other animals (at least those which are buckled into the car set or stored in a carrier which is then buckled in)
While this discussion is pretty much dead, it's always amazed me that people don't consider the other, extremely cheap solution that would not only save 90% of the kids, but help with another stupid problem:
A temp controlled air vent on the top and bottom of the car. That opens and keeps it from become 160 degrees in the car.
No false alarms, no 'arming', just something that says 'It's more than 80 degrees in here, and it's cooler outside' and opens a tiny vent set into the rear window, and one set under the front seats, so air goes through the car and the car doesn't heat up.
Granted, kids could still freeze to death, in theory, although in practice a person can keep an enclosed car warm enough through their body heat. I don't know about babies, though.
It's completely idiotic that we build cars that it gets so hot inside, and if we solved that problem, we'd incidentally stop probably 9/10ths the deaths from kids left in cars.
I consider myself an attentive parent, but this happened to me when my second child was a newborn. Fortunately, his older sister (just 20 months old at the time) asked me where he was after I had come into the house. The combination of sleep deprivation, hormones, and a crazy busy life contributed to my forgetting he was in the car. I was, of course, horrified and it never happened again. I did start the habit of putting my purse in the backseat because, it appears, I am wired to not forget my purse. I think those that demonize the parents for whom this has happened are either not parents themselves are need to rethink that this could never happen to them.
I think solar cells on the root, coupled to the ventilation are a great idea, since they also are a comfort feature for non-parents, i.e. anyone who comes back to his car parked in the open on a sunny day.
For a specific, low(ish) cost solution to this problem, I'd think that extending the standardized "isofix" child seat mounting with a weight sensor or a belt-closed-sensor and having that wired to the car alarm, some audible alarm as used for the lights or the central lock (that could require some elaborate dance to lock as long as a child-seat occupant is detected) would be a nice idea. Actually, even as a non-parent, I think that isofix should be mandatory for new cars considering how complicated mounting child seats with conventional 3 point belts is, but I haven't checked if there are any accidents/deaths attributed to improper mounting of seats.
Any of the solutions mentioned abocve that require active participation, like leaving your wallet or office keys with your child, or even setting up a reminder of the number of occupants are sill likely to fail if the parent is distracted. The absence of something non-essential is easier to overlook than an alert that goes off (and should have a very low false positive rate).
This *almost* happened to me. When my son was about 1.5 yrs old his mom was driving with him buckled into a carseat in the back of the minivan. Upon arriving home and getting out of the car, she saw the neighbor watering her shrubs, said hello, and got into a conversation. As they talked, they moved a few paces away from the car, so that she couldn't hear the baby beginning to cry in the car as the temperature, in Virginia on a hot summer day, begin to rapidly rise. Luckily, I happened to come out of the house and saw my son crying in the car. It had only been a short time, maybe 20 minutes, but already it was very hot in the car and he was drenched in sweat and red in the face. No harm done, but if even a few more minutes had passed...
Was she negligent? I would say so. Could it have happened to me? Yes. Could it have happened to you? You decide.
All the tips about how to avoid this tragedy are great.
People have brought up the analogy of checklists that pilots go through before taking off, but it occurs to me that these measures don't really apply to less regulated situations where it's not so easy to ensure that individuals are going to follow a particular routine.
Another area where safety routines come into play is rock climbing. Unlike piloting an airplane, climbers are not licensed or regulated, and all of the techniques and procedures come from "accepted lore" and are adopted (or not) by each individual as they see fit. There are however a set of more or less well established routines that rock climbers use to avoid some common causes of accidents. For example, forgetting to tie the rope to one's harness before starting a climb. Or, more commonly, to start tying the knot, then get distracted, and forget to finish it, then start climbing, with the impression that one is tied in. In most cases, the climber realizes it, corrects the problem, then goes down in search of a clean pair of underwear. If you're not so lucky, you don't discover the problem until you weight the rope, the unfinished knot fails, and you fall to the ground. To prevent this, accepted practice is to always check your partner's knot before he/she leaves the ground. Also, I make it a rule to *never* let myself get interrupted once I've started my knot until it's complete.
We all hate getting into a car that is an oven.
Why not add a solar powered fan to force ventilate
the car. In addition it could maintain the charge on the battery. I have already reduced the number of days in a week that I drive to the point that battery discharge is a possible issue and we want to do that across the nation. Also keeping the internal temp of an auto lower would increase the life of upholstery, door seals and more.
blaming it on the brain is b------t.
(I assume you are quoting Dr. Marian
My new honda has a sensor that turns off the air bag in the passenger seat if someone under weight is sitting in it. I would not think it would take NASA to develop another sensor with a warning signal that your baby is still in the car seat??
"I would not think it would take NASA to develop another sensor with a warning signal that your baby is still in the car seat??"
The problem is that it is not just babies that get put on back seats in the average car. Any weight sensor would pick up any object and register as being on the bag seat. It would not take long befor the sensor got disabled.
The next idea would be a movment sensor but a sleeping baby will often not noticably move.
The next guess might be an IR sensor to detect body heat, but with sunlight coming through car windows some objects will easily heat up to body temp or above.
There are microwave breathing detectors but with all the fear about mobile phone masts etc how easy is it going to be to sell to a parent that you are going to put their child under a microwave source?
Like a number of things that apears simple for humans to do, it is far from easy to do technicaly in a reliable way (both from false positives and false negatives).
"That leaves us with a regulatory solution -- and I'd like to see a cost per life saved before formulating an opinion about that." Posted by: Bruce Schneier
You're kidding right? You'd let a baby die to save $?
In response to another poster, cracking windows an inch will have no affect; a device setting off an alarm and lowering the windows completely down is a better alternative. Theft isn't as likely a possiblilty as a dead child. Instead of complaining, let's call/write our reps and ask them to endorse a bill (upcoming) that will require car makers to install a simple carseat sensor warning device, such as those mentioned that already exist for seatbelts. Let's put our voices and pens to work for something truely worthwhile, a child's life.
If parents aren't likely to buy an alarm device on their own, buy it for them. It would make a great baby shower gift for those overworked families.
@C Bennett: You're kidding right? You'd let a baby die to save $?
I've been on paternity leave, and didn't catch this when posted.
Though Bruce can certainly defend himself, I hate to see this unanswered. If you have read much of his writings you should know that he isn't talking about letting someone die to save $$$, he's talking about investing our life saving $$$ wisely. Spending a ton of money in one area to save 1 life is not wise if that same money could save 1,000 somewhere else. Unfortunately, we do not life in an infinite world... i.e., we have finite resources and must prioritize. He isn't talking about letting people die, he is talking about maximizing how many are saved.
Schneier.com is a personal website. Opinions expressed are not necessarily those of BT.