| Objective By analyzing the incidence and mortality trend of liver cancer in China from 1990 to 2019 and exploring the influencing factors of the incidence and mortality trend of liver cancer,to predict the standardized incidence rate and mortality tare of liver cancer in China from 2020 to 2022,and to evaluate the prevention and treatment effect of liver cancer in China in the past 30 years,so as to provide scientific basis for the formulation of prevention and treatment strategies and the allocation of health resources in China.Methods Data on liver cancer incidence and mortality in China from 1990-2019 are obtained from the Global Health Data Exchange database(GHDx).The annual percentage change(AAPC)and annual percentage change(APC)are calculated by Joinpoint regression to describe changes in liver cancer incidence and mortality trends for the whole population,males and females,respectively.The age-period-cohort model(APC model)is constructed to analyze the influence of age effect,period effect and birth cohort effect on the incidence and mortality of liver cancer.The ARIMA time series model and BP neural network model are developed to compare the prediction accuracy and modeling effects of the two models,and to predict the standardized incidence and standardized mortality of liver cancer in China from 2020 to 2022.Software for data analysis and statistics includes: SPSS 26.0 software,Matlab2019 software,Joinpoint Regression Program 4.8.0.1 software,Excel 2013 software,all tests are bilateral tests,and the test level is =0.05.Results(1)From 1990 to 2019,the crude incidence(mortality)rate,standardized incidence(mortality)rate,truncated incidence(mortality)rate of 35-64 years old and cumulative incidence(mortality)rate of 0-74 years old of liver cancer in China all showed a trend of first increasing,then decreasing and then slowly increasing,and all of them were significantly higher in males than in females.Among them,the standardized incidence(mortality)rate,the truncated incidence(mortality)rate of female aged 35-64 and the cumulative incidence(mortality)rate of female aged 0-74 showed a decreasing trend,and gradually stabilized since 2006.The sex ratio of liver cancer and the ratio of mortality sex of male and female liver cancer in China are fluctuating,among which the sex ratio of liver cancer is between 2.55-3.15,the ratio of mortality sex is between 2.44-2.87;the mortality incidence ratio(MIR)is 0.95;the agespecific incidence(mortality)rate of liver cancer in China increases with the age,among which the age-specific incidence(mortality)rate of male increases rapidly after80 years old,while that of female decreased slowly after 80 years old.(2)From 1990 to 2019,the standardized incidence rates of the whole population,male and female in China decreased by 3.13%(AAPC=-3.1347,P<0.001),2.84%(AAPC=-2.8390,P<0.001)and 3.89%(AAPC =-0.001)respectively.The standardized mortality of the whole population,male and female decreased at an average annual rate of 3.52%(AAPC =-3.5154,P < 0.001),3.27%(AAPC =-3.2669,P < 0.001)and 4.10%(AAPC =-4.1022,P < 0.001),respectively,with different trends in incidence and mortality over time.(3)The results of age-period-cohort model showed that the changes of hepatitis B incidence and mortality in the whole population,male and female in China were all affected by age,period and cohort factors(all P < 0.001),and the results of age-periodcohort model of incidence and mortality were highly coincident.The incidence and mortality of hepatitis B tended to increase first and then decrease with age.The age effects of incidence and mortality of the whole population peaked at 55-59 years old(38.93/100,000 and 34.64/100,000),while those of male peaked at 55-59 years old and50-54 years old(59.19/100,000 and 52.37/100,000,respectively)The relative risk of incidence and mortality showed a monotonous decreasing trend among female,and declined from 1990 to 2014 in the whole population and male.The cohort effect in the incidence and mortality risks of the whole population,male and female showed a decreasing trend with the backward shift of the birth age,namely the later the birth age,the lower the cohort effect of hepatitis B incidence and mortality.(4)The accuracy and effect of BP neural network model in predicting the standardized incidence and mortality of liver cancer in China from 2020 to 2022 are more effective than ARIMA model.The prediction results show that the standardized incidence and mortality of liver cancer in China will fluctuate and decline from 2020 to2022,and their predicted standardized incidence rate and standardized mortality rate are(9.6328/100,000 、 9.4401/100,000 、 9.2268/100,000)and(8.2277/100,000 、8.1624//100,000、7.9894/100,000).Conclusion(1)The liver cancer prevention and control has achieved remarkable results in China from 1990-2019.During this period,the incidence and mortality of liver cancer in my country showed a decreasing trend,but the gender difference between men and women is still significant.The incidence and mortality of liver cancer in the elderly still staying high level.(2)In the past five years,the incidence and mortality of liver cancer in China have shown a slight upward trend,and the aging of the population and changes in liver cancer risk factors are important factors in its occurrence.(3)Compared with the ARIMA time series model,the BP neural network model has better accuracy and effect in predicting the standardized incidence and mortality of liver cancer in China from 2020 to 2022.The prediction results show that the standardized incidence and mortality of liver cancer in China from 2020 to 2022 will fluctuate and decline. |
