| ObjectiveTo investigate the smoking-related cancer burden in Qingdao during 2005 to 2017 and predict the trend of mortality.To provide scientific basis for the future implementation of tobacco control strategies.MethodsWe calculated the population attributable fractions of different diseases and smoking-related cancer mortality using smoking impact ratio as exposure level based on characteristics of different diseases.The sex-age-specific mortality attributable to smoking was collected by Qingdao Municipal Center for Disease Control and Prevention online report system.The study population was obtained by the Public security bureau of Qingdao.SPSS 18.0 was used to construct the time series prediction model for smoking-related cancer mortality from January 2005 to December 2016.The mortality from January 2017to December 2017 was used to test the predictive effect of the model.The model was also used to predict the number of smoking-related deaths from 2018 to 2020.ResultThe numbers of smoking-related cancer deaths from 2005 to 2017 follows a fluctuating ascending trend.The highest number of deaths occurs in January each year.The total number of smoking-attributable cancer deaths between 2005 and 2017 was 43,628?33,756 males and 9,872 females?.The sex ratio was 3.42:1.The annual mortality rates were 53.40/105,46.17/105,49.47/105,63.15/105,67.75/105,58.89/105,55.32/105,70.07/105,61.98/105,81.40/15,77.12/105,82.66/105?78.87/105,respectively.The average annual incidence was 65.59/105.The average annual increment rate of mortality is 3.30%.The mortality rate was the lowest in 2011 and the highest in 2016.From 2005 to 2017,the population attribution fractions of smoking were 26.41%,22.53%,23.74%,26.61%,27.66%,25.76%,24.79%,29.94%,26.78%,30.79%,29.13%,31.04%and 30.32%,respectively.The annual average PAF value was 27.66%.The PAF was highest in 2016 and the lowest in 2006.The highest mortality rate was found in the 80-84 age group?278.20/105?,followed by the over-85 age group?268.19/105?and the 75-79 age group?242.98/105?.The average annual smoking-related cancer mortality rate was 102.75/105 for male and 29.33/105 for female.The average annual PAF of smoking was 31.33%for males and19.75%for females.Most amongst all smoking-related cancer mortality was contributed by lung cancer?73.57%?,followed by liver cancer?9.01%?,esophageal cancer?7.67%?and gastric cancer?4.76%?.The proportion of smoking-related cancer death was similar each year.The monthly smoking-related cancer mortality series from 2005 to 2016 remained stable after the adjustment of first-order difference and first-order seasonal difference.After model identification,parameter estimation and testing,ARIMA?2,1,0?×?3,1,0?12 was chosen as the most appropriate model for analysis.We used this model to predict the smoking-related cancer deaths in 2017 to test its validity.Extensive experiments results conducted on 2017 data showed an average relative error of 5.74%,showing the superiority of our chosen model.In addition,the smoking-related cancer deaths from 2018 to 2020 was predicted to be 5249,5423,and 6048,respectively.ConclusionsFrom 2005 to 2017,the number of smoking-attributable cancer deaths follows a fluctuating ascending trend and the mortality rate in male was higher than female.The smoking-related cancer mortality increased with age.The proportion of people aged over60 years old in smoking-related cancer mortality increased by years.The proportion of smoking-related cancer mortality was similar each year.Lung cancer accounted for the most of the deaths.Overall,ARIMA?2,1,0?×?3,1,0?12 model showed an exceptional suitability in predicting smoking-related cancer deaths in Qingdao.The predicted values were in good agreement with the actual values,yet still needed to be adjusted by supplementary data. |
PDF Full Text Request