Summary: Researchers found that long-term driving behavior can reveal early signs of cognitive decline years before clinical diagnosis. Older adults who later developed impairment showed gradual reductions in trip frequency, night driving, and route variety compared to cognitively healthy peers.
Machine learning models using GPS data predicted cognitive impairment more accurately than age, genetics, or cognitive tests alone. This low-burden monitoring approach could enable earlier interventions while preserving independence and safety.
Key Facts
- Passive Detection: GPS driving patterns predicted cognitive decline with up to 87% accuracy.
- Early Behavioral Shifts: Reduced night driving, shorter trips, and less route variation signaled risk.
- Real-World Monitoring: Daily driving outperformed traditional screening methods alone.
Source: AAN
Using in-vehicle driving data may be a new way to identify people who are at risk of cognitive decline, according to a study published on November 26, 2025, in Neurology.
“Early identification of older drivers who are at risk for accidents is a public health priority, but identifying people who are unsafe is challenging and time-consuming,” said study author Ganesh M. Babulal, PhD, OTD, of Washington University School of Medicine in St. Louis, Missouri.
“We found that using a GPS data tracking device, we could more accurately determine who had developed cognitive issues than looking at just factors such as age, cognitive test scores and whether they had a genetic risk factor related to Alzheimer’s disease.”
The study involved 56 people with mild cognitive impairment, which is a precursor to Alzheimer’s disease, and 242 cognitively healthy people with an average age of 75. All participants were driving at least once a week at the start of the study.
Participants agreed to take tests of thinking skills and to have the data tracking device installed on their vehicles. They were then followed for more than three years.
While the driving patterns of the two groups were similar at the start of the study, over time older adults with mild cognitive impairment had greater reductions in how many times they drove each month, how often they drove at night and how much they varied their routine in where they drove.
The researchers used driving factors such as medium and maximum trip distance, how often people went above the speed limit and how much they varied their routine to predict whether a person had developed mild cognitive impairment with 82% accuracy.
Once they added in the factors of age and other demographics, cognitive test scores and whether people had a gene associated with Alzheimer’s, the accuracy improved to 87%. In comparison, using all of those factors without any driving information resulted in 76% accuracy.
“Looking at people’s daily driving behavior is a relatively low-burden, unobtrusive way to monitor people’s cognitive skills and ability to function,” Babulal said.
“This could help identify drivers who are at risk earlier for early intervention, before they have a crash or near miss, which is often what happens now. Of course, we also need to respect people’s autonomy, privacy and informed decision-making and ensure ethical standards are met.”
A limitation of the study is that most participants were highly educated, white people, so the results may not be generalizable to the overall population.
Funding: The study was supported by the National Institutes of Health and the National Institute on Aging.
Key Questions Answered:
A: Yes. Subtle changes in routine, distance, and night driving predicted cognitive impairment with high accuracy.
A: Driving data predicted impairment with up to 87% accuracy, outperforming traditional screening alone.
A: Yes. Continuous, passive monitoring could identify risk before dangerous driving or major symptoms appear.
Editorial Notes:
- This article was edited by a Neuroscience News editor.
- Journal paper reviewed in full.
- Additional context added by our staff.
About this cognitive decline and neurology research news
Author: Renee Tessman
Source: AAN
Contact: Renee Tessman – AAN
Image: The image is credited to Neuroscience News
Original Research: Open access.
“Association of Daily Driving Behaviors With Mild Cognitive Impairment in Older Adults Followed Over 10 Years” by Ganesh M. Babulal et al. Neurology
Abstract
Association of Daily Driving Behaviors With Mild Cognitive Impairment in Older Adults Followed Over 10 Years
Background and Objectives
Driving integrates multiple cognitive, sensory, and motor systems and may serve as a real-world indicator of functional decline in aging. Older adults with mild cognitive impairment (MCI) often experience subtle driving changes before formal dementia diagnosis, but longitudinal, real-world evidence is limited.
This study examined whether naturalistic driving data can differentiate older adults with MCI from those with normal cognition (NC) over time and evaluated the discriminative ability of driving features compared with conventional risk factors.
Methods
We conducted a prospective, observational cohort study of community-dwelling older drivers enrolled in the Driving Real-World In-Vehicle Evaluation System Project at Washington University. Participants underwent annual Clinical Dementia Rating assessment, neuropsychological testing, and apolipoprotein ε4 (APOE ε4) genotyping.
Driving behaviors were captured daily for up to 40 months using global positioning system-enabled in-vehicle dataloggers, recording trip frequency, duration, distance, time of day, speeding, hard braking, and spatial mobility (entropy, maximum distance, radius of gyration).
Longitudinal changes were analyzed using linear mixed-effect models, adjusting for age, sex, race, education, and APOE ε4. Logistic regression with reciever operator curve analysis evaluated discrimination between older adults with MCI and those with NC, compared with conventional sociodemographic and genetic markers.
Results
Among 298 participants (MCI, n = 56; NC, n = 242; mean age 75.1 years; 45.6% female), the groups were similar in age, sex, race, and APOE ε4 status at baseline, as well as in most driving behaviors. Over time, drivers with MCI showed greater reductions in monthly trip count (MCI: −0.501, standard error [SE]: 0.21, 95% CI [−0.923 to −0.083] vs NC: −0.523, SE: 0.09, 95% CI [−0.709 to −0.337]; p < 0.001), nightly trips (MCI:−0.334, SE: 0.17, 95% CI [−0.675 to 0.001] vs NC:−0.339, SE: 0.07, 95% CI [−0.480 to −0.197]; p < 0.001), and random entropy (MCI:−0.008, SE: 0.004, 95% CI [−0.016 to −0.001]; NC:−0.014, SE: 0.002, 95% CI [−0.017 to −0.011]; p < 0.001).
Key features such as medium trip distance, speeding events, entropy, and maximum distance distinguished drivers with MCI from those with NC (area under the curve [AUC] 0.82, 95% CI 0.75–0.89). Adding demographics, APOE ε4, and cognitive composite improved AUC to 0.87 (95% CI 0.81–0.93).
Discussion
MCI was associated with progressive declines in driving frequency, complexity, and spatial range, supporting naturalistic driving data as a potential unobtrusive digital biomarker for early cognitive decline. Limitations of the study include a predominantly White, highly educated sample and a lack of external validation, warranting cautious interpretation. Continuous monitoring could augment clinical assessments, inform driving safety decisions, and guide interventions to preserve mobility in aging.
