Spatiotemporal Analysis Of Association Between Death Risks On COPD And Lung Cancer In Heilongjiang Province

Background:The prevalence and mortality of lung cancer in Heilongjiang Province(Heilongjiang)have been higher than the national average for the past 30 years,and Heilongjiang is also one of the high incidence areas of chronic obstructive pulmonary disease(COPD)in China.Recent evidence suggested that COPD was closely associated with the risk of lung cancer.But traditional epidemiological studies that revealed associations between COPD and lung cancer rarely considered the spatiotemporal auto-correlation of these two diseases.Bayesian spatiotemporal models can analyze the temporal effects,spatial effects and risk factors of diseases simultaneously and provide better estimates of the disease risk and analyze the relationships between the disease risk and known explanatory variables.Objective:This study aims to explore the spatiotemporal association between COPD and lung cancer death risk in Heilongjiang,to provide ecological evidences for studies on lung cancer risk factors,and to provide intervention targets for the prevention and control of lung cancer.Methods:We collected the monitoring data on COPD and lung cancer mortalities in Heilongjiang through 2008 to 2017.Pearson correlation and Joinpoint regression were used to analyze temporal trends and correlation of COPD and lung cancer age-standardized mortality rates.We also collected data on regional lung cancer risk factors.The optimal Bayesian spatiotemporal model based on integrated nested Laplace approximation was used to estimate the standardized mortality ratio(SMR)of lung cancer and to analyze relative risks(RR)of different risk factors,especially for COPD.Bayesian shared component model was used to identify the spatiotemporal similarities and differences between COPD and lung cancer death risk in Heilongjiang.Results:1 Temporal trends and correlation of age-standardized mortality rates between COPD and lung cancerThe results of Pearson correlation analysis showed a significantly positive correlation(r>0 and P<0.05)of age-standardized mortality rates between COPD and lung cancer in males(except for 2012)and females(except for 2011 and 2013)at county-level in Heilongjiang.Joinpoint regression analysis showed that 80(60.6%)counties in males and 71(53.8%)counties in females had same directions on average annual percent change of age-standardized mortality rates between COPD and lung cancer.2 Relative risks of COPD effect on SMR of lung cancerUsing city-level multivariate analysis based on Bayesian spatiotemporal model,we found that the age-standardized mortality rate of COPD was positively associated with the SMR of lung cancer(RR in both sexes=1.10,95%CI=1.04～1.16;RR in males=1.09,95%CI=1.03～1.16;RR in females=1.10,95%CI=1.04～1.16).At county-level,a 1/100000 increment in age-standardized mortality rate of COPD was associated with a 0.34%(95%CI=0.22%～0.46%)increased SMR of lung cancer in both sexes(males:0.29%,95%CI=0.19%～0.40%;females:0.33%,95%CI=0.20%～0.46%).3 Joint spatiotemporal death risk patterns for COPD and lung cancerThe results of SCM model suggested that death risks of COPD and lung cancer at county-level in Heilongjiang showed a significant spatiotemporal co-correlation trend from 2008 to 2017.Compared with the Heilongjiang average level,we detected that 42(31.8%)counties were both spatially high death risks of COPD and lung cancer for males,while 45(34.1%)counties for females(the probability of shared spatial effect RR>1 was greater than 0.8).Conclusion:COPD might be one of the important risk factors for death risk of lung cancer in Heilongjiang.These two diseases shared both spatial and temporal effects.We recommend that lung cancer screening be carried out in COPD patients,and measures be taken based on different conditions in each region to effectively prevent and control the development of lung cancer.