A while back I was marveling at the 10th revision of the International Statistical Classification of Diseases and Related Health Problems (ICD-10), not so much at the name but at the number of codes available for classifying bad things that can happen to people. The occasion then was the transition from ICD-9’s puny list of 14,000 injury codes to ICD-10’s majestic 70,000-plus. My point was that 70,000 codes might be a few too many because it seemed unlikely we would ever need, for example, ICD W59.22 (Struck by Turtle), given that there have been few if any cases involving verified turtle strikes. V95.40 (Unspecified Spacecraft Accident) also seemed unlikely, but not necessarily more unlikely.

It turns out this is still not technically a W59.22, but it’s about as close as you could get without having a turtle dropped on your head by an eagle.

According to ABC News, on May 9 a woman was driving down I-4 in Florida when a turtle suddenly smashed through her windshield, bounced off the passenger seat and came to rest on the dashboard. Admitting she was “a little freaked out” by this occurrence, the woman said she pulled over to assess the damage, and another driver stopped to help her and called police. Luckily, she suffered only “minor cuts” in the incident, but that of course is an “injury” we must classify, yes?

So the questions for coding purposes I think are: (1) how did this come about, and (2) what actually caused the injury?

First, the woman claimed that the car in front of her hit the turtle while it was unwisely (and very slowly) trying to cross the highway, somehow knocking it up into the air and enabling it to intercept her windshield on the way down. The police seem to have taken her word for this, but I think it’s not so clear.

According to the report, she “said she saw the debris in the road, but did not realize it was actually a turtle” until it ended up on her dashboard. But if she was following another car, how could she have seen the turtle while it was on the road? If she was in another lane, how did the turtle get the transverse velocity necessary to cross over? (I’m not saying she’s trying to cover something up, just that she may be mistaken.) And what are the odds a car would knock a turtle up into the air rather than just crushing it or knocking it forward? I guess a car impact is the most likely explanation, it just doesn’t seem that much more likely than an eagle drop or the discovery of an unknown leaping-turtle species. That’s all I’m saying.

Second, there is no dispute that the woman was not, in fact, “struck by turtle.” The turtle’s trajectory seems pretty well established, and did not intersect her at any time. She suffered a small cut or cuts on the forehead, but she believed that was caused by the rear-view mirror, which as the photos show was knocked off by the impact (and apparently deflected the turtle to the right.) So while this was very, very close to being a W59.22, I think we have to call it a V40.5 (Car Driver Injured in Collision with Pedestrian or Animal in Traffic Accident). In other words, you should use the most specific code that actually applies, but because W59.22 doesn’t strictly apply here, we have to default back to a more general code that also applies. While there is some room for interpretation with these things, that is my conclusion here.

Is this conclusion also consistent with my earlier argument that we don’t really need a code for “Struck by Turtle”? Yes it is. Is that just a coincidence? Yes it probably is.

Oh, the turtle survived with only a small tail injury (V03.10, Pedestrian on Foot Injured in Collision with Car, Pick-Up Truck or Van in Traffic Accident), and was released into a nearby pond. This species may be too dumb to know not to cross a highway and/or too slow to do it successfully, but once you’ve evolved a shell like that I guess it doesn’t matter.