Analysis of Observational Studies w crosssectional prevalence study

观察性研究的分析 Analysis of Observational Studies 南方医科大学生物统计学系

观察性研究的三种方法 w 横断面/现况/患病率研究 （cross-sectional / prevalence study) w 队列研究（cohort study） w 病例-对照研究（case-control study）

Past Time Present Future Cohort Cases and controls Crosssectional Cohort study: disease experience is collected prospectively, retrospectively or retrospectively and prospectively Case-control study: past experience of cases and controls is recalled Cross-sectional study: past experience and current disease status are collected at the same time

General form of 2× 2 table in cross-sectional study Disease Risk factor Present Absent Total Present a b a+b Absent c d c+d Total a+c b+d a+b+c+d

Initial stage in cohort study Risk factor Disease Present Absent Total Present (exposed) a+b Absent (non-exposed) c+d

Initial stage in case-control study Risk factor Disease Absent(control Present(cases) ) Present Absent Total a+c b+d

Cross-sectional study Population Now (Random) Sample Diseases status Now Present Absent Exposure status Exposed Non-exposed

400 smokers (exposed) Cohort free of disease, 1000 participants 600 nonsmokers (non-exposed) 45 with lung cancer 355 without lung cancer 5 with lung cancer 595 without lung cancer 20 years Beginning (Cause) End (Results) Cohort study

吸烟与肺癌发病关系的队列研究 Smoking Lung cancer Total Present Absent Yes 45(a) 355(b) 400(m 1) No 5(c) 595(d) 600(m 0) Total 50(n 1) 950(n 0) 1000(n)

Case-control study Now Past Population Non-cases Cases (Random) sample Controls Cases Exposed Non-exposed