Response to “Motor Output Variability Impairs Driving Ability in Older Adults”

We read the article by Lodha and colleagues, titled Motor Output Variability Impairs Driving Ability in Older Adults, with great interest. In their study, a small number of younger (n = 12) and older (n = 16) drivers completed a series of three laboratory tasks aimed at quantifying foot and ankle motor output force and variability, visual tracking, and reactive responses. In the reactive response task, participants sat in front of a computer monitor and were instructed to track a visual target using a gas pedal and react to a sudden visual stimulus (ie, brake lights appearing) by swiftly moving their foot to the brake pedal and exerting force. Their analysis showed that, compared to younger participants, older participants showed poorer performance on the reactive response task and showed greater motor output variability. Motor output variability was shown to be associated with performance on the reactive response task, leading the authors to conclude that their “study provides novel evidence that age-related declines in motor control but not strength impair reactive driving” (1). Identification of tasks that can accurately predict driving impairment would prove useful to health care providers who often have to make decisions regarding driving fitness. However, driving is a complex and multifactorial behavior, drawing upon multiple physical, cognitive, and sensory systems. Driving is further complicated by factors such as behavioral adaptation (eg, maintaining slower speeds), allowing drivers to compensate for deficits in one or more domains. The complexity of the driving task itself as well as the many factors that can contribute to performance may help explain why attempts to identify tasks that predict impaired driving performance among older adults with sufficient accuracy have proven unsuccessful (2,3). In our reading of the article, we noted several methodological inconsistencies. In particular, recruitment of participants relied upon self-reported health, which may not screen out cognitively impaired drivers. The inclusion of individuals with some cognitive deficits may help explain that some older drivers exhibited a greater number of brake pedal errors. In their analysis, the authors reported a statistically significant t-value (t = 1.87) when comparing positional variability of the gas pedal between groups. With 26 degrees of freedom, the t-value needed to reach statistical significance (assuming two-tailed with α = .05) is t = 2.056. The authors did not correct for family-wise error rate associated with multiple contrasts. Moreover, using four independent variables and stepwise selection with a small sample size (n = 28), will likely lead to “over-fitting” of the regression model (4). Thus, the large R-square is not particularly surprising nor is likely to be a true reflection of the association in the population. Our greatest concerns with the article, however, are related to the interpretation of the data given the paradigm used. The study did not measure a driving outcome but instead used a surrogate measure (ie, a laboratory-based reactive response task), thereby ignoring factors such as behavioral adaptation. It is clear from research evidence that brake reaction time alone is not predictive of fitness to drive for older adults; the activity of driving has much more complexity than moving a foot from an accelerator to a brake quickly. Even driving simulators do not demonstrate perfect correspondence with real-world performance. Similarly, it is not clear whether the statistically significant differences observed are clinically meaningful. Conclusions regarding driving impairment among older adults necessitate the examination of task performance relative to a driving outcome (eg, on-road test, at-fault collisions). Because the study does not assess driving, our interpretation of the data suggests that there are age-based differences in physical strength and response time, findings which have been well-documented in the literature, but which do not necessarily translate into a greater crash risk. While we commend the authors for their enthusiasm in seeking to identify physical factors that impair driving performance among older drivers (and particularly factors that may be amenable to remediation), we strongly advise that the authors and readers temper their conclusions to be consistent with the data and the nature of the dependent variable. The article does not demonstrate that motor output variability impairs driving ability, as presented in the title, but instead shows that motor output variables are associated with a computer-based task assessing response time, accuracy, visual attention, executive functions, and motor output; such a task does not account for the complexity of driving and the role of behavioral factors in promoting safe driving. Findings from this study cannot be used to inform clinical decisions regarding driving fitness nor should they be used to oversimplify the complex and dynamic nature of the driving task, especially considering the importance of driving to mobility, independence, and quality of life among older adults.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.