Bayesian spatial and ecological models for small-area accident and injury analysis

Area of Science:

• Epidemiology
• Spatial Statistics
• Public Health

Background:

• Regional variations in motor vehicle accident injuries (MVAI) are significant.
• Ecological/contextual risk factors at the area level require investigation.
• Bayesian spatial and ecological regression models offer advanced analytical capabilities.

Purpose of the Study:

• To apply Bayesian modeling to analyze small-area variations in MVAI.
• To identify and assess ecological/contextual risk factors for MVAI in young males.
• To investigate associations between MVAI rates and regional characteristics.

Main Methods:

• Utilized hospital separation data for 83 local health areas in British Columbia (1990-1999).
• Employed Bayesian spatial and ecological regression models.
• Examined 18 regional characteristics including socio-economic indicators, environment, medical services, crime, and traffic violations.

Main Results:

• Significant regional variation in MVAI was observed in males aged 0-24.
• Adjusting for risk factors eliminated most of the observed variation.
• Lower socio-economic status, higher adult male crime rates, seatbelt violations, and speeding charges were associated with increased MVAI.

Conclusions:

• Socio-economic status is a profound influence on MVAI in young males.
• Crime and traffic violations are significant predictors of MVAI in specific age groups.
• Findings can inform injury prevention strategies and regional health planning.