Analysis Of The Severity Of Traffic Accident Injury Among The Elderly

The problem of road traffic safety is widely concerned by the society,and the analysis of accident injury severity is an important subject to ensure traffic safety.Compared with young people,the vision of the elderly is gradually degraded,the injury is not easy to recover,and the mobility is decreased,which is easy to cause more serious traffic accidents in the event of traffic accidents.Therefore,it is necessary to study the severity of traffic accident injuries in the elderly.The existing scholars mainly study the accident injury severity from four aspects: human,vehicle,road and environment,and the built environment factors will also have a certain impact on the accident injury severity.Ignoring the heterogeneity and time instability between the data will lead to bias and even wrong conclusions.Therefore,this paper analyzes the severity of traffic accident injuries in the elderly from three aspects: built environment factors,heterogeneity and time instability.Firstly,taking the traffic accidents of the elderly as the research object,23,599 traffic accident data of the elderly from 2015 to 2019 were extracted from the traffic accident database.According to the longitude and latitude of the traffic accident data of the elderly and the built environment data,a new traffic accident database of the elderly is constructed.From the characteristics of the elderly,vehicles,roads,road environment and built environment,21 influencing factors were selected for multicollinearity diagnosis and statistical analysis of traffic accident data of the elderly.Secondly,based on the traffic accident data of elderly intersections in 2019,the potential category Logit model is used to explore the impact of built environment factors on the accident injury severity of elderly intersections.Third,based on the traffic accident data of elderly drivers from 2015 to 2019,the mixed Logit model,the random parameter Logit model considering mean heterogeneity and the random parameter Logit model considering mean and variance heterogeneity were respectively constructed for model comparison,and the heterogeneity was analyzed based on the model with the best goodness of fit.Fourthly,the global time instability test and local instability test,as well as the stochastic parameter Logit model considering mean and variance heterogeneity,were used to calculate the marginal effect values of each influencing factor to explore the time instability existing in the accident data of elderly drivers from 2015 to 2019.Finally,some preventive measures are proposed based on the model results.The results show that:(1)According to the potential Logit model,the absence of shopping centers in the 300-meter buffer zone of the accident will increase the probability of minor injuries.(2)The goodness of fit of Logit model with random parameters considering mean and variance heterogeneity is higher than the other two models.(3)The Logit model with random parameters considering mean and variance heterogeneity from 2015 to 2019 identified the intercept term,parameters occurring in the work area,weather and male drivers as random parameters with normal distribution;Rural area,post-collision escape,straight line,peak/bottom have mean heterogeneity.The speed is [0,10]km/h,the lane is one-way,the motorcycle,the airbag is not ejected,and the body condition is normal with variance heterogeneity.(4)Through the global time instability test and the local time instability test,it is found that there is time instability in the traffic accident data of the elderly from 2015 to 2019.Among the influencing factors,there were significant time instability,such as illness,male driver,and escape after collision.The research results of this paper are not only helpful for in-depth analysis of the factors affecting the severity of traffic accidents in the elderly,but also helpful for the traffic management department to formulate traffic safety policies to reduce the severity of traffic accidents in the elderly.