Recent research published on 26 November 2025 in Neurology®, the medical journal of the American Academy of Neurology, suggests that patterns captured through in-vehicle driving data may help identify individuals at elevated risk of cognitive decline. Rather than relying solely on clinic-based tests or demographic factors, the study explored the informative potential of everyday driving behaviour, recorded passively through a GPS tracking device. This approach reflects a broader movement toward real-world, unobtrusive monitoring tools that can reveal early cognitive changes long before they become clinically apparent.
Lead author Ganesh M. Babulal, PhD, OTD, of Washington University School of Medicine in St. Louis, explained that spotting older drivers who may be unsafe on the road has long proved challenging. Traditional assessments are time-consuming, episodic, and often unable to capture the gradual behavioural shifts associated with early impairment. In the study, however, the driving data provided a more sensitive indicator of emerging issues. Babulal noted that GPS-derived patterns were more accurate at detecting cognitive decline than models drawing only on age, cognitive test results, or genetic risk factors for Alzheimer’s disease, offering a compelling rationale for incorporating behavioural monitoring into future assessment strategies.
The research followed 56 older adults with mild cognitive impairment—widely understood as a precursor to Alzheimer’s—alongside 242 cognitively healthy peers, all with an average age of 75. At the outset, everyone was driving at least weekly and agreed to complete cognitive tests and have a monitoring device installed in their car. Over more than three years of follow-up, the long observation window enabled the team to identify subtle trajectories of behavioural change, capturing patterns that short-term studies or one-off evaluations might easily overlook.
Although both groups displayed similar driving profiles at the start, differences emerged over time. Participants with mild cognitive impairment gradually reduced how often they drove each month, curtailed night-time driving, and showed less variation in the places they travelled. These shifts, while not immediately dramatic, signalled the gradual tightening of routine that often accompanies early cognitive difficulties. By examining characteristics such as trip distance, speeding frequency, and route variability, the researchers developed models that detected mild cognitive impairment with 82 per cent accuracy, rising to 87 per cent when demographic data, cognitive test scores, and genetic information were added. Without driving data, accuracy fell to 76 per cent, demonstrating the unique value of real-world behavioural indicators.
While the findings offer an encouraging step toward earlier identification and intervention, the authors acknowledge essential limitations. Most participants were highly educated and white, raising questions about whether the results can be generalised to broader, more diverse populations. The promise of driving-behaviour monitoring also raises ethical considerations around privacy, autonomy, and data governance. As Babulal pointed out, although such monitoring is low-burden and unobtrusive, its use must be balanced with safeguards that preserve individuals’ rights and dignity. Further research across more varied communities will be essential to determine how widely these behavioural patterns apply and how best to implement such tools responsibly.
More information: Ganesh M. Babulal et al, Association of Daily Driving Behaviors With Mild Cognitive Impairment in Older Adults Followed Over 10 Years, Neurology. DOI: 10.1212/WNL.0000000000214440
Journal information: Neurology Provided by American Academy of Neurology
