Evaluation and modification of the multifactorial model of driving safety: an empirical assessment using the ACTIVE study
Open Access
- Author:
- Tian, Joanne
- Graduate Program:
- Human Development and Family Studies
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- May 10, 2021
- Committee Members:
- Lesley Ross, Thesis Advisor/Co-Advisor
Jacqueline Ann Mogle, Committee Member
Douglas Teti, Program Head/Chair - Keywords:
- older adults
driving safety
cognition
vision
physical function - Abstract:
- Previous research has demonstrated that cognitive, visual, and physical performance are associated with driving safety among older adults. However, there are few comprehensive models examining how all such factors jointly impact driving safety. The present study first evaluated the Multifactorial Model of Driving Safety developed by Anstey’s team (2005), then modified this model according to recent research. We used structural equation modeling (SEM) to analyze the impact of cognition, vision and physical function on older adult’s driving safety using the baseline data from the Advanced Cognitive Training for Independent and Vital Elderly (ACTIVE) study. Participants (n = 2391) were between 65 and 91 years old, and 73.3% of were female. While the original model had a poor fit (CFI = 0.846, TLI = 0.777), the modified model demonstrated a good fit (CFI = 0.991, TLI = 0.986). There was a negative relationship between physical function and driving avoidance, and there was a positive relationship between cognition and physical function. Visual acuity was not associated with driving safety. Additionally, in our final modified model, driving avoidance partially mediated the relationship between physical function and crashes, which indicated that only the effect of physical function on crash operated through driving avoidance. This study highlights the importance of these predictors in older adults’ safe driving. Future research should examine possible dynamic changes between these predictors and driving in a longitudinal model.