Welch H Gilbert, Woloshin Steven, Schwartz Lisa M, Gordis Leon, Gøtzsche Peter C, Harris Russell, Kramer Barnett S, Ransohoff David F
VA Outcomes Group, White River Junction VA Medical Center, 215 N Main St, VA Outcomes Group, 11B, White River Junction, VT 05009, USA.
Arch Intern Med. 2007 Nov 26;167(21):2289-95. doi: 10.1001/archinte.167.21.2289.
Last year, the New England Journal of Medicine ran a lead article reporting that patients with lung cancer had a 10-year survival approaching 90% if detected by screening spiral computed tomography. The publication garnered considerable media attention, and some felt that its findings provided a persuasive case for the immediate initiation of lung cancer screening. We strongly disagree. In this article, we highlight 4 reasons why the publication does not make a persuasive case for screening: the study had no control group, it lacked an unbiased outcome measure, it did not consider what is already known about this topic from previous studies, and it did not address the harms of screening. We conclude with 2 fundamental principles that physicians should remember when thinking about screening: (1) survival is always prolonged by early detection, even when deaths are not delayed nor any lives saved, and (2) randomized trials are the only way to reliably determine whether screening does more good than harm.
去年，《新英格兰医学杂志》发表了一篇头条文章，报道称如果通过螺旋计算机断层扫描筛查发现肺癌患者，其10年生存率接近90%。该出版物引起了媒体的广泛关注，一些人认为其研究结果为立即开展肺癌筛查提供了有说服力的依据。我们强烈反对这种观点。在本文中，我们强调该出版物没有为筛查提供有说服力依据的4个原因：该研究没有对照组，缺乏无偏倚的结果测量方法，没有考虑以往研究中关于该主题的已有知识，也没有提及筛查的危害。我们最后总结了医生在考虑筛查时应牢记的2条基本原则：（1）即使死亡没有延迟、也没有挽救任何生命，早期发现总是能延长生存期；（2）随机试验是可靠确定筛查利大于弊的唯一方法。
