| Objectives:(1)To analyze the levels,causes and characteristics of injury mortality in Chinese population,describe the long-term trends of mortalities of total injury and the main injuries(road traffic injury,self-harm,falls,drowning)and carry out the corresponding model investigations,and investigate the possible causes of injury mortalities and their influencing factors,providing references for the future work of the prevention and control on injury deaths in China;(2)To systematically study all kinds of parameter estimation methods for the Age-period-cohort(APC)Model,and then study the theory of Estimable Functions Approach in depth,providing a methodological reference for the future studies of the APC Model in China.Methods:(1)Using the data of injury death and population provided by GBD 2015,the direct standardization method was adopted to standardize the injury mortalities,analyze the composition of injury deaths and the main types of injuries faced by different age groups in China,and compare the death spectrums of Chinese population and variations of the proportion of different injuries in different years;(2)The mortality trends of total injury and four major injuries in China from 1990 to 2015 were described,then the Jointpoint Regression Model was adopted to identify whether the ascending/descending trends in different sections were statistically significant,and the logarithmic transformation linear regression analysis was used to calculate the annual percentage change of the injury mortalities;(3)On the basis of systematically studying the characteristics and shortcomings of parameter estimation methods for the APC Model,the theory of the Estimable Functions Approach was studied in depth and its characteristics and its advantages were clarified.In addition,the common types of the estimable functions were verified in the framework of a unifies system,and several newer estimable functions with stronger interpretations were introduced based on them;(4)The APC Model with the Estimable Functions Approach was adopted to analyze the mortality trends of total injury and the four main injuries,decomposing the death risk into three aspects:age effects,period effects and cohort effects,and then the possible causes and influencing factors of the changes in injury mortalities were investigated according to the model results.Results:(1)In 2015,the standardized mortality rate of total injury in China was 54.21/100,000(74.74/100,000 for males and 32.93/100,000 for females).Both male and female mortality rates were generally increasing with age groups,but the constituent ratios of total death for injury among men and women in different age groups witnessed large variations.Total injury deaths in the age groups from 5 to 24 years old accounted for the largest proportion of total death(about 50%).The injury mortalities of males were more serious than that of females,and the injury mortality rate in both sexes increased with age groups overall.Injury is an important factor leading to premature death and labor loss.In the period of 1990-2015,the top four leading causes of injury death in China were always road traffic injury,self-harm,falls and drowning(although their orders have changed),and these four injuries occupied the vast majority of all the injury deaths(about 75%);(2)The net drift of total injury mortality was-1.522%(95%Cl,-1.665%to-1.379%)per year for males and-3.492%(95%Cl,-3.688%to-3.295%)per year for females,and the local drift values were below 0 in all age groups in both sexes during the period of 1990 to 2015.The results of longitudinal age curves(LAC)showed that the mortality risk of total injury in Chinese males and females in the same birth cohort showed trends of decrease and then increase,and decrease and increase again in their life stages after adjusting period effects.The Period Rate Ratio(PRR)of the mortality risk of total injury for both sexes was generally declining during the study period.The Cohort Rate Ratio(CRR)of the mortality risk of total injury for both sexes was declining,except that of male birth cohorts between 1941-1945 and 1966-1970 which kept a stable trend;(3)The net drift of road traffic injury mortality was-0.006%(95%CI,-0.109%to 0.229%)per year for males and-1.261%(95%CI,-1.492%to-1.029%)per year for females,and the local drift values were below 0 in all age groups in males below 30 years old and females during the period of 1990 to 2015.The results of LAC showed that the mortality risk of road traffic injury in Chinese males and females in the same birth cohort showed trends of decrease and then increase in their life stages after adjusting period effects.The PRR of the mortality risk of road traffic injury for both sexes was generally declining since 2005 whereas that of males witnessed an increasing trend and that of females witnessed a stable trend from 1990 to 2005.The CRR of the mortality risk of road traffic injury for males was increasing before the 1965-1969 cohort and then showed a decreasing trend,and that of females showed a declining trend overall;(4)The net drift of self-harm mortality was-3.290%(95%Cl,-3.431%to-3.151%)per year for males and-5.253%(95%CI,-5.445%to-5.062%)per year for females,and the local drift values were below 0 in all age groups in both sexes during the period of 1990 to 2015.The results of LAC showed that the mortality risk of self-harm in Chinese males and females in the same birth cohort showed trends of increase,then decrease,and then increase again in their life stages after adjusting period effects.Both the PRR and CRR of the mortality risk of self-harm for both sexes showed a declining trend overall;(5)The net drift of falls mortality was-0.517%(95%Cl,-0.676%to-0.358%)per year for males and-1.562%%(95%CI,-1.913%to-1.210%)per year for females,and the local drift values were below 0 in all age groups in males below 40 years old and females during the period of 1990 to 2015.The results of LAC showed that the mortality risk of falls in Chinese males and females in the same birth cohort showed trends of decrease and then increase in their life stages(increase rapidly in old age)after adjusting period effects.The PRR of the mortality risk of falls for both sexes was generally declining before 2000 whereas that of males witnessed an increasing trend from 2000 to 2010 and then a decreasing trend again from 2010 to 2015,and that of females witnessed an increasing trend from 2000 to 2005 and then a decreasing trend again from 2005 to 2015.The CRR of the mortality risk of falls for males was increasing before the 1965-1969 cohort and then showed a decreasing trend,and that of females showed a declining trend overall;(6)The net drift of drowning mortality was-3.100%%(95%Cl,-3.368%to-2.832%)per year for males and-3.164%(95%CI,-4.509%to-3.818%)per year for females,and the local drift values were below 0 in all age groups in both sexes during the period of 1990 to 2015.The results of LAC showed that the mortality risk of drowning in Chinese males and females in the same birth cohort showed trends of rapid decrease,then decelerating decrease,and then gradual increase in their life stages after adjusting period effects.Both the PRR and CRR of the mortality risk of drowning for both sexes showed a declining trend overall.Conclusions:(1)In 2015,the standardized mortality rate of total injury in China was lower than that of the world average,and it was between the levels of total injury mortality in the World Bank upper-middle income countries and upper-income countries.The injury mortalities of males were more serious than that of females,and the injury mortality rate in both sexes increased with age groups overall.Injury is an important factor leading to premature death and labor loss.Road traffic injury,self-harm,falls and drowning were the four major types of injury in China;(2)The APC model serves as a statistical method that can simultaneously study age effects,period effects and cohort effects,decompose the various factors influencing certain social phenomenon and analyze root causes of the event.Based on the Joinpoint Regression Model which effectively describes whether the trends were statistically significant,it can further analyze the underlying factors underlying these trends.The Estimable Functions Approach can overcome the shortcomings of traditional parameter estimation methods and preferably solve the "non-identification problem" of the APC model;(3)Age is an important risk factor for injury death.After controlling cohort effects and adjusting period effects,the maximum death risk of road traffic injury and falls in Chinese population is concentrated in the old age of after 70 years old,that of self-harm is concentrated in the stage of 20-24 years old,and that of drowning is in the childhood stage below15 years old.Appropriate interventions could be taken according to the variations of injury mortality risk in different age stages;(4)In the period from 1990 to 2015,the Period Rate Ratio(PRR)of injury mortalities in Chinese population generally showed a decreasing trend with the time pass by.The main factors influencing the changes in PRR of the four major injures are:the change of ICD coding,improvement of quality of the death code registration,rapid economic development,fast urbanization,the change of behavior and travel method,the relevant laws and regulations and the corresponding regulatory measures,prevalence and survival rates of chronic diseases and drug taken status,medical diagnosis and treatment level,etc.(5)The reason for the stable relative risk of total injury mortality in the male birth cohorts from 1941-1945 to 1966-1970 is that the decreasing trend of Cohort Rate Ratio(CRR)of certain types of injuries(such as self-harm and drowning)in these birth cohorts was offset by an increase in CRR of other types of injuries(such as road traffic injury and falls).The main factors influencing the changes in CRR of the four major injures are:safety awareness,behavioral habits,health care conditions,educational level,BMI level,life stages when experience the urbanization,etc. |
PDF Full Text Request