Aichun Tan1, Danping Tian1, Yuanxiu Huang1, Lin Gao1, Xin Deng1, Li Li1, Qiong He1, Tianmu Chen1, Guoqing Hu1, Jing Wu2. 1. Department of Epidemiology and Health Statistics, Public Health School, Central South University, Changsha 410078, China. 2. Division of NCD Control and Community Health, Chinese Center for Disease Control and Prevention. Email: wujingcdc@163.com.
Abstract
OBJECTIVE: To develop the forecasting models for fatal road traffic injuries and to provide evidence for predicting the future trends on road traffic injuries. METHODS: Data on the mortality of road traffic injury including factors as gender and age in different countries, were obtained from the World Health Organization Mortality Database. Other information on GDP per capita, urbanization, motorization and education were collected from online resources of World Bank, WHO, the United Nations Population Division and other agencies. We fitted logarithmic models of road traffic injury mortality by gender and age group, including predictors of GDP per capita, urbanization, motorization and education. Sex- and age-specific forecasting models developed by WHO that including GDP per capita, education and time etc. were also fitted. Coefficient of determination(R(2)) was used to compare the performance between our modes and WHO models. RESULTS: 2 626 sets of data were collected from 153 countries/regions for both genders, between 1965 and 2010. The forecasting models of road traffic injury mortality based on GDP per capita, motorization, urbanization and education appeared to be statistically significant(P < 0.001), and the coefficients of determination for males at the age groups of 0-4, 5-14, 15-24, 25-34, 35-44, 45-54, 55-64, 65+ were 22.7% , 31.1%, 51.8%, 52.3%, 44.9%, 41.8%, 40.1%, 25.5%, respectively while the coefficients for these age groups in women were 22.9%, 32.6%, 51.1%, 49.3%, 41.3%, 35.9%, 30.7%, 20.1%, respectively. The WHO models that were based on the GDP per capita, education and time variables were statistically significant (P < 0.001)and the coefficients of determination were 14.9% , 22.0%, 31.5%, 33.1% , 30.7%, 28.5%, 27.7% and 17.8% for males, but 14.1%, 20.6%, 30.4%, 31.8%, 26.7%, 24.3%, 17.3% and 8.8% for females, respectively. CONCLUSION: The forecasting models that we developed seemed to be better than those developed by WHO.
OBJECTIVE: To develop the forecasting models for fatal road traffic injuries and to provide evidence for predicting the future trends on road traffic injuries. METHODS: Data on the mortality of road traffic injury including factors as gender and age in different countries, were obtained from the World Health Organization Mortality Database. Other information on GDP per capita, urbanization, motorization and education were collected from online resources of World Bank, WHO, the United Nations Population Division and other agencies. We fitted logarithmic models of road traffic injury mortality by gender and age group, including predictors of GDP per capita, urbanization, motorization and education. Sex- and age-specific forecasting models developed by WHO that including GDP per capita, education and time etc. were also fitted. Coefficient of determination(R(2)) was used to compare the performance between our modes and WHO models. RESULTS: 2 626 sets of data were collected from 153 countries/regions for both genders, between 1965 and 2010. The forecasting models of road traffic injury mortality based on GDP per capita, motorization, urbanization and education appeared to be statistically significant(P < 0.001), and the coefficients of determination for males at the age groups of 0-4, 5-14, 15-24, 25-34, 35-44, 45-54, 55-64, 65+ were 22.7% , 31.1%, 51.8%, 52.3%, 44.9%, 41.8%, 40.1%, 25.5%, respectively while the coefficients for these age groups in women were 22.9%, 32.6%, 51.1%, 49.3%, 41.3%, 35.9%, 30.7%, 20.1%, respectively. The WHO models that were based on the GDP per capita, education and time variables were statistically significant (P < 0.001)and the coefficients of determination were 14.9% , 22.0%, 31.5%, 33.1% , 30.7%, 28.5%, 27.7% and 17.8% for males, but 14.1%, 20.6%, 30.4%, 31.8%, 26.7%, 24.3%, 17.3% and 8.8% for females, respectively. CONCLUSION: The forecasting models that we developed seemed to be better than those developed by WHO.