GODILLON Sylvanie et VALLEE Julie

Inégalités socio-spatiales de risque d’accident en tant que piéton : un cumul de facteurs individuels et contextuels ?

13/05/2015

Résumé :

Pour approfondir l’analyse des inégalités socio-spatiales devant le risque d’accident de la route, cet article s’intéresse aux caractéristiques individuelles et contextuelles des personnes impliquées dans un accident en tant que piéton. En analysant les données relatives aux accidents dans l’agglomération lilloise à l’aide de modèles de régression statistique multiniveaux, il s’agit de comparer le profil des individus impliqués dans un accident en tant que piéton avec celui des individus impliqués dans un accident avec un autre mode de transport. Les résultats montrent que, quand ils sont impliqués dans un accident de la route, les habitants des Zones Urbaines Sensibles (ZUS) le sont davantage en tant que piéton que les habitants d’autres quartiers. Cette différence n’est pas le simple reflet de la structure démographique et sociale des ZUS puisqu’elle demeure significative après ajustement sur le sexe, l’âge, l’activité professionnelle et la catégorie socioprofessionnelle des individus. D’autres caractéristiques du quartier de résidence – comme une forte présence de logements en immeubles collectifs et une sur-occupation des logements – se révèlent associées à un risque accru d’accident piéton. En revanche, aucune association n’a été mise en évidence avec l’intensité du trafic routier dans le quartier de résidence.

Accéder au texte intégral : Godillon_vallee_rfst_2015


Abstract

Socio-spatial inequalities in pedestrian accident risk: individual and contextual factors

 

According to the literature, social and socio-spatial inequalities exist in traffic accidents. Research indicates that road accident rates are particularly high for low-income classes (Van Beeck et al., 1991; Hasselberg, et al., 2001). Other studies have focused on the characteristics of the residential areas in which victims of accidents live, and these also show that accident rates are higher for residents in poorer areas (Dougherty et al., 1990; Abdalla et al., 1997; Fleury et al., 2010; Haddak et al., 2012). Certain types of road users, like pedestrians, are especially vulnerable. France’s National Observatory of Road Safety (Observatoire National de la Sécurité Routière) has found that the risk of being seriously injured is 1.6 times higher for pedestrians than people travelling by car (ONISR 2014). Given the trend to increased walking as a mode of transport in cities, it is particularly important to understand better the risks pedestrians face when trying to improve road safety.

 

This article investigates the profile of individuals as pedestrians who are involved in accidents by comparing them with persons who experienced accidents arising in other forms of transport (drivers or passengers of personal vehicles, cyclists, people using public transport, etc.). The aim is to analyse the socio-spatial inequalities of risks faced by pedestrians: are persons living in poorer areas likelier to be involved in road accidents than people living in richer areas? Once the individual characteristics are taken into account, what are the variables of the residential environment that are associated with an increased risk of having a pedestrian accident? To what extent are the characteristics of an individual’s residential area (in terms of social housing or traffic environment) linked to the risk of being involved in a traffic accident as a pedestrian?

 

This research focuses on road accidents involving residents of nine disadvantaged neighbourhoods in the Lille area, in northern France. According to the inventory of accidents recorded by the police, 2,185 residents in these areas were involved in a road traffic accident between 2000 and 2007. Around 17% of these persons were pedestrians. Other databases provide information about the residential neighbourhood of the victims, including a population census, a mobility survey and traffic data. The methodology here uses a multi-level regression model. To build this model, demographic and social characteristics of the inhabitants are used (including sex, age, activity and socio-professional category). Information relating to residential areas is also used, such as the share of housing units in apartment flats, the average number of persons per room and road traffic volume.

 

The results show that several individual and contextual factors come into play in understanding the risks of pedestrian accidents. At the individual level, risks are higher for women, persons under 18, and persons over 80, as well as for the unemployed and people without work. From a contextual point of view, accident victims in poorer neighbourhoods are likelier to be pedestrians than are the residents of richer areas. This difference between areas is not merely a reflection of the age or sex structure of the population, nor its social composition. Once adjustments have been made to take into account the demographic and social structure of a population, the inhabitants of poorer areas have a risk 1.5 times greater of being involved in accident as pedestrians (see Table). The characteristics of the residential environment are also associated with a higher risk of a pedestrian accident. These include a strong proportion of housing units in apartment flats, and high housing occupation rates. No evidence was found with respect with the traffic intensity in the residential area.

 

Table: Risks of Being Involved in a Road Traffic Accident as a Pedestrian according to Individual and Residential Characteristics (authors, 2015)

 

Sans titre-1