Crash Prediction Models for Older Drivers: A Panel Data Analysis Approach

Hu, Patricia; Trumble, David; Foley, Daniel; Eberhard, John; Wallace, Robert · 1996 · ROSA P / Oak Ridge National Laboratory

archive: archived pipeline: cataloged verified

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

Summary

This study addresses the growing highway safety concerns associated with the aging population, specifically aiming to identify factors that place older drivers at greater crash risk. The authors argue that previous research has been limited by methodological flaws, particularly the failure to account for the temporal order between the onset of medical conditions and crash events. By linking "snap-shot" data to multi-year crash records, earlier studies obscured causality. This paper utilizes a panel data analysis approach to overcome these limitations, examining the joint impacts of socio-demographic factors, functional limitations, medical conditions, and medication use while controlling for driving exposure. The analysis is based on longitudinal data from the Iowa 65+ Rural Health Study, linked with crash records from the Iowa Department of Motor Vehicles for the period 1985–1993. The dataset includes 6,553 female person-years and 5,414 male person-years. The researchers employed Poisson regression models to analyze crash counts, testing for overdispersion and random effects but finding the standard Poisson specification appropriate. Two separate gender-specific models were developed to account for significant differences in risk factors between men and women. The models included variables such as annual miles driven, physical functional limitations, chronic diseases, cognitive test scores, and medication use. The results reveal distinct gender-specific risk profiles. For older female drivers, the significant risk factors were annual miles driven, having difficulty extending arms above shoulder level (indicating motor capability deficits), back pain, and living alone. Women with difficulty extending their arms were more than twice as likely to be involved in crashes compared to those without this limitation. For older male drivers, the significant risk factors included annual miles driven, employment status, a history of glaucoma, low scores on word-recall tests (indicating cognitive deficits), and the use of antidepressant drugs. The use of antidepressants was identified as the most influential risk factor for men aside from mileage, doubling the probability of crash involvement. Notably, employment increased crash risk for men but was associated with a lower, though marginal, risk for women. The study concludes that crash prediction models for older drivers must account for temporal correlations and gender differences to be effective. The findings suggest that licensing and screening practices should consider specific functional and cognitive impairments, such as motor deficits in women and cognitive or vision issues in men, alongside medication use. The research highlights that while older drivers are generally safer than younger cohorts, specific physiological and psychological limitations significantly elevate their crash risk, necessitating more sophisticated evaluation methods beyond standard vision tests.

Key finding

Antidepressant use doubles the probability of crash involvement for older male drivers, while difficulty extending arms above shoulder level doubles the crash risk for older female drivers.

Methodology

modeling

Sample size: 1811

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. Discovered via bulk_ingest_rosap on 2026-05-23 (45 acquisition events logged).

StageOutcomeToolModelPromptAttemptsCompleted
discover success rosap 2 2026-05-23
archive success 1 2026-05-23
extract success cached 4 2026-06-10
clean success 1 2026-06-01
chunk success 1 2026-06-01
embed success 1 2026-06-02
enrich success 1 2026-05-23
promote success 1 2026-05-23
summarize success llm qwen3.6-27b-prismaquant summ-v5 42 2026-06-10
tag success vector_similarity 19 2026-06-11
verify partial 2 2026-06-10

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified_with_issues.

Topics

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

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).