Iowa's population of older adults is one of the highest in the country. As Iowans age, the number of cases of dementia also increases. Doctors may soon be able to test older drivers with dementia to determine if they can safely continue to drive, thanks to research done at The University of Iowa. Jeff Dawson, Sc D, associate professor of biostatics in the UI College of Public Health, talks about the study:
Who participated in the study?
We had 40 elderly drivers who had mild Alzheimer's disease and 115 elderly drivers that did not have Alzheimer's disease, and we used those as controls. All 155 drivers in the study had valid driver's licenses and were still driving on a regular basis, so we weren't forcing or encouraging anybody to drive in our study that wasn't already driving in the first place.
How were driving skills tested in both groups?
As far as the driving skills, we have a vehicle that has special instrumentation to record electronic data, as well as to record video and audio of what the driver is doing and what's out on the road. Each participant drove a standard 35 mile route. There was a research assistant right next to them, acting kind of like a driver's ed instructor would act, that could put on the brakes if there was any particular problem, to ensure safety.
How were results evaluated?
This is lot of data. This is 155 video tapes, each about 45 minutes long. But these were very carefully reviewed by a professional driving instructor who, frame by frame, looked to see what the drivers were doing. And he counted how many safety errors and what type of safety errors they were making. So if they were crossing the line, or if they ran a stop sign, or if they ran a red light, he would mark these sorts of things down.
What were the outcomes of the study?
We found that those with mild Alzheimer's disease committed about 25 percent more safety errors than healthy drivers. The most common safety error category was lane deviation, such as crossing the center line, or straddling the center line, or going off on the shoulder a little bit. These occurred about 50 percent more often in the Alzheimer group. But also, and in a sense maybe more importantly, we found that poor performance on cognitive tests (one example of cognitive tests involves completing a dot to dot diagram), those that did poorly in that within the Alzheimer's group did even worse than those that did OK. On the other hand, those that did a little better than average on those types of cognitive tests, really did just as well as those that did not have Alzheimer's disease—as far as driving goes.
If one of the drivers was coming at me, was there a set amount of time—with the Alzheimer's drivers—where they finally did eventually realize they were over the center line or did run a stop sign? Did it have to be brought to their attention? Was it continual?
That's a good question. We marked down safety errors, and some were marked as more serious than others, but there weren't any near crashes or any problems like that where safety was a huge issue to either the driver or to the other drivers on the road during the study.
Ultimately, what was the goal of the study?
We sometimes hear about tragic car accidents involving people with Alzheimer's disease and some of these are sensationalized, perhaps, by the media. It's not real clear how often this sort of thing really occurs, especially with those with early stages of the disease. We wanted to see whether cognitive tests could be used to help to decide whether a person should be driving or not if they have Alzheimer's disease.
The study basically gave support and evidence that these tests are useful in making this determination. But that's not the only thing that should be used and driving behavior is very complex; so we haven't really been able to propose precise cut-off scores for these tests to say, 'OK, if you score above this value or below this value, then you should or should not be driving.' We're not there yet.
What are the next steps in determining driving safety in adults with dementia?
One thing I think is very promising is going back to our electronic data from the vehicle. We have 10 data points per second and we want to do some complex analyses to see if certain electronic patterns are predictive of driving safety errors.
If someone has a lateral acceleration of too many G forces, then maybe that's predictive of these driving safety errors whether or not they have Alzheimer's disease—and this sort of thing is happening with teen drivers, too.
We want to try to monitor their data over time and automatically send warnings to family members or care givers if a patient's driving safety is deteriorating, so that we could hopefully avoid an accident or a tragedy. |
