I didn't find out she was 78 years old until I had already pronounced her dead. We didn't get an age when we got the call about a high-speed motor vehicle crash. Or when we heard more details—ejected driver with an arm amputation, hypotension, head injury, and insecure airway. Even when she showed up, her age wasn't obvious; I would have guessed late 50s, but her face was too bloody to tell.

Not that her age really mattered during the brief resuscitation. It ran as smoothly as we could have hoped: the resident secured a difficult airway, blood was hung before EMS finished the report, and we had all the help and resources we needed.

But the battle was lost before it started—with an initial systolic blood pressure of 40, a GCS of 3, and major on-scene hemorrhage, her chance of survival was extremely low. But we gave it every effort, down to an emergency department (ED) thoracotomy when she arrested.

Afterward, I experienced my usual mix of emotions after an ED death: concern we didn't do enough fast enough, tinged with discomfort at how our massive resuscitation efforts can leave a patient looking like a war casualty.

But I was more troubled than usual, and it was because of her age. On paper, the case is just another bad car crash, but for me it was also a collision between my clinical practice and my research interest in older driver safety.

As I wrote my note and began looking through her medical records, I discovered her age and medical problems, all of them common with advancing age: severe arthritis, progressive dementia, and macular degeneration. I also heard from the on-scene police officer that earlier that day they had received a call about her erratic driving.

My reaction—as I suspect yours has been—was to think that this woman, given her medical history, should not have been driving a car. But my reaction is accompanied—as I hope yours will be—by frustration at how difficult it is to accurately identify which older drivers should be required to stop driving without penalizing those who are still safe. Because mobility is critical for independence, which in turn is a key to prolonged health and well-being, it's not sensible to restrict driving privileges based on age alone.

But this patient had significant problems all previously identified as crash risk factors—so why was she still driving? The patient's excellent internist knows the warning signs for increased crash risk. And we have a robust electronic medical record system that should have facilitated communication among her providers. But the patient had missed appointments, and none of the many provider notes mention driving. It's not clear whether anyone even asked the patient whether she was still driving. Or perhaps the patient had already stopped driving, but on this day for some reason she got back in the driver's seat.

I'll never know why she was behind the wheel that day. I'm thankful that no one else was hurt in the crash, and I'm relieved there won't be another “elderly driver kills innocent bystander” headline to fuel prejudice against older drivers (who generally pose a greater risk to themselves than to those around them). And I am reenergized in my efforts to find better ways to identify unsafe older drivers and to help them prepare to “retire” from driving. An emergency medicine colleague once told me it's easy to identify which older drivers are unsafe: they're the ones who roll into the trauma bay after a crash. But I believe we can, and must, do better than that.