Incidence and costs of accidents among attention-deficit/hyperactivity disorder patients.

Swensen A, Birnbaum HG, Ben Hamadi R, Greenberg P, Cremieux PY, Secnik K.

Eli Lilly and Company, Indianapolis, Indiana, USA (A.S., K.S.).

PURPOSE: The purpose is to analyze the incidence and costs of accidents among Attention-Deficit/Hyperactivity Disorder (ADHD) patients. METHODS: The analysis relied on administrative medical, pharmaceutical, and disability claims for a national manufacturer's employees, spouses, dependents, and retirees (n > 100,000). Accidental injuries were identified using ICD-9 codes for injuries or poisoning treatment. ADHD sample consists of individuals with at least one claim for ADHD during 1996-98 (N(ADHD) = 1308), which was compared with a matched control sample. In addition to descriptive statistics, multivariate analysis involving logistic regression was used to model the probability of having an accident claim in 1998. This probability was estimated for the whole population, for adults alone, for children (under age 12 years), and for adolescents (age 12-18 years). We also estimated a generalized estimation equation (GEE) model to account for the possibility of multiple accident claims for a single patient. RESULTS: ADHD patients had a greater probability of having at least one accident claim than their controls for children (28% vs. 18%), adolescents (32% vs. 23%), and adults (38% vs. 18%). Although ADHD patients' costs were greater than their controls for adults ($483 vs. $146), there was no difference for children or adolescents. However, among patients with accident claims, the average number of accident claims was similar for both groups (3.6 vs. 3.5) and costs were not statistically different. The multivariate analysis confirms this utilization pattern: the odds of having an accident for ADHD patients were 1.7 times greater than for controls. CONCLUSIONS: ADHD was a significant predictor of having an accident claim. However, for people with an accident claim, ADHD patients and controls had a similar number of accident claims and costs.

PMID: 15450551 [PubMed - in process]

Cool! A quick round of Who can spot the flaws?

First and most obvious to me: they haven't accounted for differences in reporting accidents between ADDers and normals.

I was taught that you have to account first for the bias brought about by the partitioning principle. Once you divide the population, you have to determine if the two groups have different response rates to whatever is being compared.

If you break down the population into blind and sighted groups, and then study the relative preference for pastels vs. bold wall paint, the statistics won't mean a thing.

Not the greatest example, but the first that came to mind.

Next?

i've had an accident every year since i've learned how to drive..the good news? i only started driving when i was 21 or 22...and that means that it hasn't been eons..
but all my accidents were minor and lo and behold ! some of these accidents weren't even my fault! but since i had an accident two wks ago and had to say a goodbye to my car
my husband stated blandly that:
you've had an accident every year that you've driven....
what am i supposed to say?

We know something about this, and we know how to fix it. I did it for myself, and then years later for Kay, and my sons Chris and Bryan.

The hard thing is accepting that you can prevent accidents that aren't your fault. When I say it's hard, I mean it is really amazingly difficult; only Bryan was able to do it without relying on that desperate feeling that you have right after a crash.

I totaled six or seven cars over a bunch of years before I figured it out. None of them were my fault. But I could have prevented all of them, and I sure wish I had.

It doesn't have anything to do with those 'statistics' in the original post, though.

I should add this: the fix works, but driving at night without any other traffic will make you crazy. But it's well worth it.

yeah i know
but that post reminded me of the accident list ...
whats this fix you're talking about?
and what kind of engineering are we talking about?

Simple head engineering, that you do to yourself.

You have to do two things. First, you develop the ability to 'read' what other drivers are thinking. You look at every other automobile and see where it's going, how it's going to accelerate, when it's going to change lanes or turn.

This is simple enough, but it does require looking at everyone, even at their eyes in their own rear view mirrors – exactly what every state has warned drivers not to do, because of 'road rage'. It doesn't happen, so don't worry about it.

This part is like a game, and you'll quickly develop the ability to predict what almost every other car on the road is going to do, well before they do it.

We can pull up to a light and tell you what the other cars already there are going to do when the light changes. We're right about 80% of the time. If we pull up to the light with traffic, so we see everybody driving before they stop, we’re close to 100%.

You can do it, too. It's really no different than listening to the conversations in a restaurant. We told our boys to think of the drivers as talking to you with every nuance of how they control their vehicle. You just listen to what they have to say.

That is all it takes to see problems before they develop, and avoid them. But it's not near enough, because there are two potential holes in the information you get this way.

The first one is stuff you don’t see. If you come up to a blind intersection, there's no way that you can know ahead of time what a car careening at you is about to do.

The second is yourself. We get tired, and forget to look. Our attention wanes for a minute, and we miss the guy ahead changing lanes. Nobody's perfect, and everybody misses stuff sometimes.

So this is where the second piece of self-engineering comes in, and it's a little more esoteric. You learn to set alarms, little things that make you tingle funny when a certain situation arises.

Everybody already does this; it's what we usually call a sense of the situation. You have to think through the correct sense of the situation of being in a car driving, so you can get that funny feeling when something isn't right.

That's an alarm, and it's pretty easy to do. All you need is the correct mental picture, and when the scene doesn't match, Ding!

Here's the picture you need: practice seeing the scene around you when you drive as a seething cauldron of information, all being fed to you by the actions of the other vehicles, their drivers all shouting what their intentions are. You want to feel immersed in it, constantly, because we need that for the second half of the alarm part of the engineering.

When you see someone announcing that they're going to do something, you automatically adjust to avoid problems. In fact, you learn to drive in the cracks, between the groups when traffic's not so bad.

You'll learn the difference between freeway traffic at seven AM and noon. At seven AM, everyone knows exactly what everyone else is going to do, because if you don't, you can't tailgate. Everybody watches the third or fourth car ahead, because if you don't you can't stop fast enough to avoid the person in front of you.

Anyone that ventures out into seven AM freeway traffic without understanding the rules sticks out like a sore thumb, and everyone automatically leaves more room, because the driver is telling everyone that s/he can't be trusted to do the right thing.

Traffic at noon is normal by comparison. Nobody pays too much attention to anyone else, and people get ticked if you follow too close. Other situations are less structured, but they have the same character, a kind of babble that you have to sort out as you go.

Once you get accustomed to always being immersed when you drive, you'll find yourself noticing an odd thing: a driver that you can't read. The first time you see it, you'll know you've got the alarms working right, because this is the toughest one to see.

Your instinct will be to slow down and leave the person room, and that's exactly what you have to do. Often, people that aren't telling you what they're going to do are drunk, but you'll learn to read that soon enough, too. The real problems are caused by people so scared, or so nutty that you can't tell what they're going to do no matter how closely you look.

Once you begin to be alarmed by stuff that you can't read, you can start to learn to look at the scene, instead of just the other vehicles. Then you'll be able to get that antsy feeling about those blind corners and driveways that run between buildings, and like that.

Cars parked close together on a residential street is another good example, because of the danger of kids running out between them. What you want to cultivate is an alarm for the situation in which you don't know what to expect from some part of the general scene, and then pay particular attention to that part.

Our experience with this is a kind of feeling of dread when there's the potential for a surprise, almost like watching a horror film when you know you're being set up for something. It can be so physical that it forces you to slow down, even if you can't exactly put your finger on what's wrong.

The last bit of engineering the alarms has to do with you. So far, the alarms go off when you see something alarming. (Duh!) No alarms is what you want when you drive, but how do you tell the difference between Everything's OK and I'm daydreaming and don't see a thing?

This is an interesting problem, which causes something called the ratchet effect. The ratchet effect is responsible for the persistent idea that patients in institutions like nursing homes or hospitals are more restless during a full moon.

What you do to defeat it is set a second, different alarm. You concentrate on the feeling of being constantly immersed in the sea of information around you when you drive, and cultivate the fear that you're not safe if you're not having that immersive experience.

So this is a negative alarm, one that is easy to set, but (surprise) only ADDers can do well.

What you want is to have silence scream at you.

Kind of like flying along around ten or eleven at night on the deserted back roads of North Carolina,

on your way to the Outer Banks for a two week vacation that you needed to take six months ago to keep from burning out,

bugs streaking at you in your headlights, making it impossible to get a feel for just how fast you're really going,

and not another car in sight for at least the last half hour, no towns, barely any buildings, the lines on the road are useless,

you've been on the road for nine hours, the last five straight, and you don't know if the realtor's left the key in the mailbox,

and you're not getting any input at all about how safe the situation outside the car, on the road, really is,

and you're reminded about that every single second of every single moment that the road slides by, until you want to

!!!SCREAM!!!!

then you've got it. Barring the stuff you really can't anticipate, like a forklift dropping off a building onto your roof while you're parked, you won't have any more accidents.

Really.

* * * *

BTW, when I said it was hard, amazingly hard, it's the part about getting past the idea that other people should look out for themselves. We're all more childlike when driving than anyone realizes, and it's a lot like growing up all over again, and learning to have a mature attitude.

If you can remember the strange distance that you felt when you first realized that you actually thought that your little brother should take the last piece of cake, that's what it feels like.

Kay had to have a guy jump out from between two cars in a line of stopped traffic, smack her in the side, and take off.

What she said to me was this: "What could I have done? Is there anything I could do other than just giving up driving? I don't care what it is, I just don't ever want to feel like this again if I can help it."

But it was still hard for her to accept the responsibility for keeping the other guy from hitting her. Once she did, the whole immersing herself in the sea of information and setting alarms thing only took a couple of weeks.

It's way more complicated to tell than show someone, but any ADDer can do it. And since most normals are closet ADDers in one way or another, they can probably do it, too. But it will get derailed if someone is stubborn enough, either about taking responsibility for the other guy's mistakes or about being completely normally normal.