Development of digital driving safety screening tool for older drivers

Rhode Island Hospital, Providence RI

Investigators

Linked publications & trials

Abstract

Motor vehicle crashes (MVCs) are a leading cause of injury and mortality in older drivers, in part due to injury susceptibility and frailty. Cognitive impairment is a major risk factor for unsafe driving behavior and MVCs. However, existing assessments of driving safety are limited by only modestly predicting on-road behavior and crash risk, requiring administration by a specialty provider, high expense, need for a lengthy monitoring period, and/or only narrowly focusing on a single aspect of complex driving behavior (e.g., field of view). This means most existing assessments are not scalable for use as screening instruments. This project capitalizes on insights from our team’s extensive background in identifying risky driving behavior in older drivers and the development of a digital assessment sensitive to detect cognitive impairment in this population. We have identified four new subtests that are validated predictors of driving safety but are missing from other existing screening tools. We propose to incorporate these new subtests into our existing digital assessment to create the Digital Rhode Island Vehicular Risk Tool (DRIVER). We will use an iterative participatory design process to ensure aging-friendly design and user experience (Aim 1). We will then conduct a clinical validation study (Aim 2) examining DRIVER’s ability, compared to conventional paper-and-pencil neuropsychological tests, to predict driving safety and MVC risk in 100 older drivers (split evenly between those with and without cognitive impairment). We will also determine the feasibility and acceptability of assessing naturalistic driving behavior using a person’s mobile phone as an in-vehicle sensor (Aim 3), as naturalistic assessment may reveal subtle alterations in driving behavior that predict future MVC risk. To ensure we appropriately consider future implementation efforts during this development phase, we will also conduct qualitative interviews with end users (N=4-6 older drivers) and facilitators (N=4-6 each of spouses/adult children of older drivers and medical providers who work with older adults) to assess the feasibility, acceptability, and appropriateness of DRIVER and naturalistic assessment as well as the potential adoption and sustainability of this method of assessment. Successful completion of these aims will yield a user-friendly screening tool designed to detect older drivers most at risk for adverse outcomes, including MVC. Results will also facilitate the PI’s transition to independence by providing preliminary data to motivate a future R01 application focused on the ability of DRIVER to prospectively predict driving outcomes using in-vehicle sensors and longitudinal monitoring. This project is aligned with the strategic priorities of the Rhode Island Hospital Injury Control COBRE to prevent injury through digital innovation and translation of research to practice and policy.