Factors Affecting Crash Severity among Elderly Drivers: A Multilevel Ordinal Logistic Regression Approach

Alrumaidhi, Mubarak; Rakha, Hesham · 2022 · OpenAlex-citations

DOI: 10.3390/su141811543

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study investigates the factors influencing crash severity among elderly drivers (aged 65 and older) in Virginia, United States, addressing the growing safety risks associated with an aging population. With elderly drivers facing higher fatality and injury rates due to age-related physical and cognitive deficits, the research aims to identify specific exogenous variables that contribute to severe outcomes. The authors highlight a gap in previous literature, noting that traditional statistical models often ignore spatial heterogeneity and cluster-specific effects, which can lead to biased parameter estimates. To address this, the study employs a Multilevel Ordinal Logistic Regression (M-OLR) approach to account for variations across different physical jurisdictions. The analysis utilizes crash data from the Virginia Department of Transportation spanning 2014 to 2021. The dataset includes 157,800 crashes involving senior drivers across 313 physical jurisdictions. Crash severity was categorized into three ordinal levels: property damage only, minor injuries, and possible fatalities. The researchers examined eighteen variables, including driver characteristics (distraction, drowsiness, belt usage, alcohol/drug involvement) and site-specific conditions (roadway type, weather, speed limits, animal involvement). The M-OLR model was compared against a standard Ordinal Logistic Regression (OLR) model to evaluate the impact of accounting for spatial clustering. The results demonstrate that the M-OLR model provides a statistically significant better fit than the standard OLR model, confirming the importance of addressing spatial heterogeneity. Key findings indicate that crashes on two-way roads are more severe than those on one-way roads. Driver behavior significantly impacts severity; distracted, drowsy, or unbelted elderly drivers face escalated risks of severe crashes compared to their compliant counterparts. Additionally, crashes involving speed violations or occurring on higher-speed roads result in more extreme outcomes. Conversely, crashes involving animals were found to likely result in property damage only rather than severe injuries. The significance of this study lies in its methodological improvement and actionable safety insights. By validating the use of multilevel modeling, the authors provide a more accurate framework for analyzing crash data that accounts for geographical and jurisdictional variations. The identified risk factors offer specific targets for intervention, supporting the design of safer road networks and policies tailored to protect elderly drivers. These findings contribute to broader transportation safety efforts by highlighting the distinct vulnerabilities of older drivers and the environmental conditions that exacerbate crash severity.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• sex gender
• demographic disparities
• pre crash contributing factors
• incidence prevalence
• vru crash typology

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Empirical Findings: crash risk outcomes