Real-world Study On Environment And Medicine Based On Big Data

Part 1 Human eosinophils,basophils and neutrophils can be indicators of human immune response,and excessive exposure to air pollutants is associated with the development of immune diseases.The aim of this study was to determine the effects of particulate matter(PM2.5,aerodynamic diameter < 2.5mm;PM10,aerodynamic diameter < 10 mm),sulfur dioxide(SO2),nitrogen dioxide(NO2),carbon monoxide(CO),and ozone(O3)on human eosinophils,basophils and neutrophils in Chinese populations.The data were extracted from the records of physical examinations at the Physical Examination Center of Sichuan Provincial People’s Hospital in Chengdu from 2013 to 2016(N=246,425).We counted the number of physical examination records per day with counts of eosinophils,basophils and neutrophils exceeding the upper limit of the normal reference range as the daily prevalence.A distributed lag non-linear model(DLNM)was used to evaluate the lagged effect of each air pollutant on eosinophils,basophils and neutrophils.The lagged effects of each pollutant were counted and presented with smooth splines and show the 95% confidence intervals for each effect of pollutant.The results showed that the effect of PM2.5(RR: 1.154,95%CI: 1.062-1.255)and PM10(RR: 1.309,95%CI: 1.130-1.517)on eosinophils reached the maximum value on the lag 0th day in female,with significant difference;In the male population,PM2.5 had no significant effect on eosinophils,but PM10 had a significant effect on eosinophils with a lag of 20-28 days.The effects of O3 on basophil in male and female were significant at lag 24-27 and lag 10-24 days,respectively.The effects of PM2.5 and PM10 on male neutrophils reached the maximum at lag 0 day,while male neutrophils reached the maximum at lag 35 days.In summary,PM10 and O3 had significant effects on eosinophil and basophil in both male and female.PM2.5 and PM10 had significant effects on the neutrophils of male and female,but their modes of expression were different.Female eosinophils are more sensitive to air pollutants than male.Part 2 Whether long-term insulin use increases the risk of liver cancer in diabetics is still a matter of debate.Therefore,patients diagnosed with type 2 diabetes(T2D)from January 1,2014 to December 31,2015 were selected from the Shanghai Hospital Link database.Follow-up was started on the date of first diagnosis of T2 D or the date firstly prescribed the hypoglycemic drug,end at the date of death,diagnosed with liver cancer or December 31,2018.Finally,a total of 172,843 patients with T2 D were enrolled in this study.The study population was divided into an original cohort(135,106 insulin ever-users and 37,737 insulin never-users)and a matching cohort treated with a propensity score(PSM)(35,740 insulin ever-users and 35,740 insulin never-users).Cox regression was used to estimate the Hazard Ratios(HRs)of liver cancer in each cohort.Finally,the Standardized Incidence Rates(SIRs)of liver cancer in each cohort of men,women and total population were calculated in the original cohort.The results showed that in the original cohort,a total of 210 cases of liver cancer were diagnosed,non-adjusted HRs and fully-adjusted HRs are 2.741(95%CI: 2.082-3.609)and 2.608(95%CI: 1.944-3.498),respectively.In the matched cohort,a total of 116 cases of liver cancer were confirmed,with HR of 2.546(95%CI: 1.712-3.785).The SIRs of the insulin ever-users for male,female,overall T2 D patients were 2.008(95%CI: 1.483-2.532),4.028(95%CI: 2.587-5.469)and 2.577(95%CI: 2.032-3.122),respectively,which were all higher than those in insulin neverusers.These findings suggest that long-term insulin use may be associated with an increased risk of liver cancer in patients with type 2 diabetes.