[医学]05第五章 病例对照研究.ppt

实施方法 选择队列 确定病例组 确定对照组 收集资料 统计分析 病例对照研究分析方法估计OR值 用准似然危险度估计RR值 用Cox模型进行多因素分析等等 估计关联强度 准似然危险度（pseudo risk）：将子队列按有暴露史和无暴露史分为两个相对人群，分别计算暴露组和非暴露组的发病比率。 A1和A0分别表示子队列暴露组中的发病人数和非暴露组的发病人数，B1和B0有暴露组的人数和无暴露组的人数。 与巢式病例对照研究不同点 对照是随机选取，不与病例进行配比。 随机对照中的成员如发生被研究的疾病，既作为对照，又同时作为病例。 可以同时研究几种疾病，不同疾病有不同的病例组，但对照组都是同一组随机样本。 病例交叉设计 Case-Crossover Design is a scientific way to ask and answer the question that “Was the patient doing anything unusual just before the onset of the disease?” Hazard Control Event Exposure Case-Crossover Design I’m so sick…. Did I do anything unusual right before the illness in comparison to my usual routine? To answer this question, we need to do the comparison within the individual. Illness Control period Hazard period It is a design that compares the exposure to a certain agent during the interval when the event does not occur (control period), to the exposure during the interval when the event occurs (hazard period). Case Crossover Design It is related to prospective crossover design. “Crossover” means an experiment in which all subjects pass through both the treatment and placebo phases. Each individual serves as his or her own control. It is a matched case-control study but involves cases only and each individual serves as his/her own control. The data from a case crossover design can also resemble cohort data if the control data are units of person-time. Hazard Control Event Exposure Case Crossover Design Hazard Period Control Period Event Onset Exposure ? Exposure ? How to Analyze Case Crossover Data? Since in the case crossover design we are comparing the exposure in the hazard period to the exposure in the control period, and the individual serves as his or her own control, the analysis will be very similar to that of a match case-control study. Hazard Control Event Exposure Assuming the lengths of hazard period and control period are the same ... Collect the data in the hazard period and control period. Assemble the data into a 2?2 table as we d