Built and social environment characteristics associated with motorcyclist mortality in Latin American cities from the SALURBAL study

Abstract

Background: Motorcyclists are the fastest growing road user group in Latin America, and account for 25% of all road traffic collision deaths. This study examines the relationship between motorcyclist mortality and the built and social urban environment in Latin American cities.

Methods: We studied 337 cities with ≥ 100,000 inhabitants in seven Latin American countries. Mortality data from 2010 to 2019 were obtained from civil registries and linked to cities defined by the SALURBAL project. Motorcyclist deaths were identified using ICD-10 codes, with redistribution of ill-defined codes. City-level measures included population, urban development, street design, public transportation, and social environment. Associations were estimated using multilevel negative binomial models. A subanalysis of 300 cities with motorcycle registration data was conducted.

Results: The crude city-level motorcyclist mortality rate was 4.16 per 100,000 population. Age-standardized rates varied from 0.51 to 22.60. Males had higher mortality rates, with the highest rates in 20-24-year-olds. After adjustment, cities with higher population density (RR 0.92 [95% CI 0.85-1.00]), intersection density (RR 0.91 [95% CI 0.83-0.99]), and social environment index (RR 0.88 [95% CI 0.83-0.93]) had lower motorcyclist mortality. More curvilinear street layout (RR 0.97 [95% CI 0.90,1.03]) and the presence of public transportation (RR 0.94 [95% CI 0.87,1.03]) showed a non-significant association with mortality. Higher urban development isolation (RR 1.07 [95% CI 1.00-1.14]) was associated with higher mortality, but the association weakened after adjustment. In cities with motorcycle registration data, higher rates of registered motorcycles were associated with higher motorcyclist mortality.

Conclusion: Motorcyclist road traffic deaths in Latin American cities are associated with specific city-level characteristics. In fully adjusted models, higher intersection density and a stronger social environment index were linked to lower mortality rates. City-level interventions that improve street connectivity, promote safer and more cohesive urban environments, and address social inequities in infrastructure and services may help reduce motorcycle deaths and enhance road safety in the region.

Keywords: Built environment; Cities; Epidemiology; Latin america; Motorcycles; Traffic collisions.

Fig. 1

Age-standardized motorcyclist mortality rate per 100,000 population (2010–2019), by country and sex. The box represents the interquartile range (25th–75th percentile), and the line within the box represents the median

Fig. 2

Boxplot of motorcyclist mortality rate per 100,000 population (2010–2019), by country, sex, and age group (A = Men, B = Women). The box represents the interquartile range (25th–75th percentile), and the line within the box represents the median

Fig. 3

Motorcyclist mortality rate per 100,000 motorcycle registrations (2010–2019), by country. The box represents the interquartile range (25th–75th percentile), and the line within the box represents the median

Fig. 4

Rate ratios of motorcyclist mortality associated with city characteristics in single exposure (Model 1) and multivariable (Model 2) models. Rate ratios and 95% CI were estimated using a multilevel negative binomial model with random intercept for each city, robust variance estimator, and the corrected age-sex population counts as an offset. All the models were adjusted for country, sex (male or female) and age (5-year age groups). All RRs and 95% CIs reflect a difference of 1 SD except Presence bus rapid transit or subway. Model 1 is a single exposure model of each exposure. Model 2 includes all exposures in a multivariable model

Fig. 5

Rate ratios of motorcyclist mortality associated with city characteristics in single exposure (Model 1) and multivariable (Model 2) models including motorcycles registered per 1,000 population as a covariate. Subanalysis specific for Brazil, Chile, Colombia and Mexico. Rate ratios and 95% CI were estimated using a multilevel negative binomial model with random intercept for each city, robust variance estimator, and the corrected age-sex population counts as an offset. All the models were adjusted for country, sex (male or female) and age (5-year age groups). All RRs and 95% CIs reflect a difference of 1 SD except bus rapid transit or subway. Model 1 is a single exposure model of each exposure. Model 2 includes all exposures in a multivariable model