热度 32|
协变量(三)
关于分层随机化,小胖还需要提的一点便是对研究中心的考虑,因为我们现在接触到的绝大多数试验都是多中心临床试验,而大多数多中心临床试验都把研究中心作为分层因素,这种分层是出于实际操作的原因或因为研究中心(或者研究者)可能由于其它一些已知的或未知的预后因素成为一个重要的混杂因素。如果每个中心的患者人数比较少,根据中心进行分层就显得不切实际。在这种情况下,应该考虑是否可以按照别的因素,比如说国家或地区进行分层随机化。这种选择可能是因为相同的用药情况、治疗或者其它因素等使得分层更为合理。
总之,最后也是最重要的是,你选择的分层因素及其原因都必须在研究方案中加以描述和解释。
By A. Buzdar, J. Douma, N. Davidson, R. Elledge, M. Morgan, R. Smith, L. Porter, J. Nabholtz, X. Xiang, C. Brady. Phase III, Multicenter, Double-Blind, Randomized Study of Letrozole, an Aromatase Inhibitor, for Advanced Breast Cancer Versus Megestrol Acetate. Journal of Clinical Oncology, Vol 19, Issue 14 (July), 2001: 3357-3366
“This was a randomized, double-blind, parallel-group, multicenter, international, comparative phase III study conducted in 120 centers throughout the United States, Canada, and Europe. Enrollment of 602 patients occurred over a 30-month period. Patients were randomly assigned to one of three treatment arms: letrozole 0.5 mg qd, letrozole 2.5 mg qd, or megestrol acetate 40 mg qid. Randomization was performed for each country without stratification by center.”
“A total of 602 patients from 120 centers in seven countries were randomized in the study over a 30-month period”
上边这个试验就是一个典型的按照国家来进行分层随机化的例子。由于120个中心一共入组600多例病人,如果按照中心分层,那么显然每个中心的人数太少,不切实际,因此按照国家进行分层随机化。
手机版|会员|至尊|接种|公卫人 ( 沪ICP备06060850号-3 )
GMT+8, 2024-5-29 17:43 , Processed in 0.028511 second(s), 5 queries , Gzip On, MemCached On.
Powered by Discuz! X3.4
© 2001-2023 Discuz! Team.