Factores asociados a la letalidad por accidentes de tráfico en ciclistas en España (1993-2017)

Abstract

Despite the decrease observed in road traffic crashes (TC) mortality in recent years in Spain, cyclists’ crashes have not shown the same tendency. They have even a higher mortality rate nowadays. In addition, there is a public interest in promoting the use of bicycles as a sustainable means of transport. This could lead to further increases in mortality rates from TC involving cyclists. Objective: To identify the direction and to quantify the magnitude of the association of environment- and cyclist-dependent factors with the risk of death after being involved in traffic crashes in Spain, with special emphasis on the use of helmets. Methodology: A retrospective cohort study was designed from a case series and broken down into two sub-periods (1993 - 2013 and 2014 - 2017). For this purpose, the database of the Spanish General Directorate of Traffic was used, with information on 90,582 cyclists involved in crashes on the spanish mainland, of whom 1,882 died. A multiple imputation procedure was performed using the chained equations method due to the presence of missing data. Crude and adjusted Poisson regression models were used to estimate the associations between each independent variable and the risk of death, using multilevel models, considering both cyclist and province as aggregation levels. Causal models were used to identify the magnitude of the possible effect of helmet use on the risk of death. We assessed the role of the area of the crash as a confounder and/or as an effect modifier of this association. All analyses were performed with the statistical package Stata 14. Results: The risk of death can vary according to the type of accident and environmental circumstances at the time of the TC, with a higher risk for those occurring on the road, outside intersections, under adverse weather conditions, with an unaltered road surface and during the early hours. It was also observed that fatality has shown a gradual decrease throughout the time series. Cyclists who showed a higher risk of death were those who were older at the time of the TC; males; those with a nationality other than Spanish; and those who had committed an offence. Our results suggest a protective effect of helmet use on the risk of death, which is of a greater magnitude if the TC occurred on the road than if it occurred in an urban area. In fact, the area in which the TC occurred is a confounder of the association between helmet use and the risk of death, as well as an effect modifier. Conclusions: We identified different factors associated with the risk of death after a cyclist is involved in a TC in Spain. These factors are either the type of accident, the environmental circumstances, or the characteristics of the cyclist himself. The associations found were maintained in both sub-periods studied. Helmet use decreased the risk of cyclist´s death after the TC by 0,3% to 2,3%, and the area in which the TC occurred was the main confounder of this association.