统计学
1.He who cannot quote his therapeutic experiences in numbers is a charlatan;be truthful for clarity's sake, do not hesitate to admit failures, as they must show the mode and places of improvement.
—— Theodor Billroth, 1829——1894
【译文】不能用数字来引证自己治疗经验的人就是个江湖郎中。为清晰起见，要诚实，承认失败不要犹豫，因为这样才能揭示所需改进的地方和方式。
【点评】真正的科学态度。
2.Statistics, which, if properly collected and set forth, are immensely valuable, are often but the dull man's delight.The little items, "excluded cases" and"moribund cases" are common in statistical tables, and like charity, they cover a multitude of sins.
—— J.Chalmers Da Costa, 1863——1933
【译文】统计学，如果（资料）收集和解释得当，是非常有价值的，但常常只有木脑袋人乐意这样做。几个小条目，“剔除病例”和“濒死病例”在统计表格中司空见惯，就像慈善事业一样，掩盖了许多罪恶。
3.A statistical analysis, properly conducted, is a delicate dissection of uncertainties, a surgery of suppositions.
—— M.J.Moroney
【译文】运用得当的统计分析是对不确定性问题的细致剖析，譬如，某种手术。
4.Statistics: the only science that enables different experts using the same figures to draw different conclusions.
—— Evan Esar, 1899——1995
【译文】统计学：唯一能让不同的专家用同样的数字得出不同结论的科学。
【点评】因此，在使用统计学这个工具时，要保持几分警惕。
5.There are three kinds of lies: lies, damned lies, and statistics.
—— Benjamin Disraeli, 1804——1881
【译文】谎言有3种：谎言、该死的谎言和统计学。
6.Some purveyors of statistics are fine imaginative writers and should be war correspondents.
—— J.Chalmers Da Costa, 1863——1933
【译文】一些统计学数据的提供者是极富想象力的作家，他们应该去做战地记者。
7.Statistics don't lie, but the men who make them sometimes do.
—— J.Chalmers Da Costa, 1863——1933
【译文】数据不会说谎，但造数据的人有时会。
8.Statistics will prove anything, even the truth.
—— Berkeley Moynihan, 1865——1936
【译文】统计学数字会证明一切，甚至真相。
9.Even a sensible person promptly associates the term "statistics" with the thought: "this is a bunch of lies".
—— August Bier, 1861——1949
【译文】即使是明智的人也会迅速把“统计学”这个词和“这是一堆谎言”的想法联系在一起。
【点评】这句话充分揭示了“统计学”这种工具的两面性。
10.According to the latest official figures, 43% of all statistics are totally worthless.
—— Ashleigh Brilliant
【译文】最新官方数据显示，43%的统计学数据毫无价值。
【点评】还不错，至少有价值的超过了50%！
11.Medical statistics are a special branch of alchemy, carefully crafted to turn bullshit to airline tickets.
【译文】医学统计是炼金术的一个特殊分支，精心打造，能把胡编乱造吹上天。
12.Medical statistics are like a bikini.What they reveal is interesting but what they conceal is vital.
【译文】医学统计就像一件比基尼，暴露部分是引起人们兴趣的，但隐藏的才是最重要的。
13.The only figures which I find statistically significant are those recorded anonymously and analyzed impartially by someone with no personal interest in the verdict.
—— William Heneage Ogilvie, 1887——1971
【译文】我发现唯一有统计学意义的数字是那些未署名的记录，并由对其结论毫无个人兴趣的人去秉公分析所得。
【点评】要做到这一点真不容易。
14.Any study that reports a certain number of consecutive cases without a complication (death) actually means that the number of cases plus two were performed with two complications (deaths); the first and last.
—— Michael Hoffman
【译文】任何一个说没有并发症（或死亡 ）的一定数量的连续病例研究，实际上意味着在其报道的病例数中加上2例有并发症（或死亡）的病例，第一个和最后一个。
15.Doctors think a lot of patients are cured who have simply quit in disgust.
—— Herbert Hoover
【译文】医生们常把一些因厌恶而退出研究的患者都当着治愈来处理。
【点评】回顾性研究中如果随访不严谨，这种情形并不少见。
16.Suspect the retrospective study, which reports a zero mortality rate in X number of octogenarians: it only means the first and last mortalities were not included.(https://www.daowen.com)
—— Moshe Schein
【译文】值得怀疑的是，回顾性研究报道了多少个八旬老人的死亡率为零：这只是意味着第一个和最后一个死亡者没有包括在内。
【点评】这种可能性是存在的。
17.If your first essay at an operation results in fatality it will take a longtime to get good statistics.
—— Mark M.Ravitch, 1910——1989
【译文】如果一种手术首次报道就有死亡病例，那就需要很长时间才能得到漂亮的统计数据。
18.The statement "all patients operated upon in this manner survived"probably means that any patient who died must ipso facto not have been operated upon "in this manner" and may therefore be excluded from the series.
—— Mark M.Ravitch, 1910——1989
【译文】“所有以这种方式手术的患者都存活了下来”这一说法可能意味着，任何死亡的患者事实上都没用“这种方式”手术，因此才可能被排除在系列之外。
【点评】所以，请外科医生说这句话时要多思考一下。
19.You don't need to run Chi Square on common sense.
【译文】无需对常识进行卡方检验。
20.The fact that a stone, thrown out of the window, rises instead of falls is not statistically significant, but it is a powerful observation.
—— Nicola Basso
【译文】往窗外扔出一块石头，上升而不是下降这一事实虽然不具有统计学显著性，却是一个有力的观察结果。
21.If you torture the data long enough, you can make it say anything.
【译文】如果你折磨数据足够长时间，你要它咋说就能咋说。
22.When anyone essays to publish statistics it impresses me in the same way as when someone starts to relate a fishing experience.One must admit that the truth is theoretically possible.
—— Arthur E.Hertzler, 1870——1946
【译文】任何论文所发表的统计学数据，给我的印象就像有人开始讲述其钓鱼的经历。当然，我们也必须承认，理论上说，也有可能是真的。
23.My personal thesis is that surgical judgment should be based on"numbers", with the caveat that the numbers themselves may be suspect.
—— Seymour I.Schwartz
【译文】我的观点是，外科判断应以“数”为基础，要注意的是数字本身可能不可信。
【点评】数字可信与否，提供数字的人至关重要。
24.Surgeons were slow to embrace the new technique, possibly because of their self-perceived image of infallibility.One of the first honest doubters who saw the merits of the random control trial in surgical research was a colorectal surgeon,John Goligher.Thirty years ago he appreciated the three fundamental rules: that a trial must be prospective with contemporary controls randomized by a method that cannot be influenced by the investigator, that it must have a single objective clearly stated, and that it must have a potential for altering clinical practice.
—— Mary Evans
【译文】外科医生拥抱新技术的速度通常较慢，这可能与他们以完美无缺的形象自居有关。第一个（挑战传统的）坦诚的质疑者是一位结直肠外科医生约翰·戈利格，他率先看到了随机对照试验在外科研究中的优势。30年前，他就赞赏三个基本原则：①实验必须是前瞻性的，并采用随机化方法设置同期对照，分组不能受研究人员的影响；②必须有一个可清晰表述的单一评价目标；③必须具有改变临床实践的潜力。
【考注】约翰·戈利格（Goligher John Cedric.1922——1998）是国际著名结直肠外科医生，生前曾担任英国利兹大学利兹总医院普外科主任43年，对临床结肠直肠外科做出过许多重要贡献。1961年出版问世的Surgery of the Anus，Rectum and Colon（Textbook Binding）是第一部结直肠外科教科书，是年轻外科医生必备读物。曾5次修订再版（最后一版是1984年）。据称他是最早认识到胃肠吻合器的价值，最早进行临床随机对照试验的外科医生，当初是用作评估治疗消化性溃疡的各种手术，并获得了长期随访。这项开创性的工作和许多在结直肠其他领域的研究成果让他在国际上获得了广泛赞誉和尊重。
25.As surgical procedures are evolving so rapidly there is a very narrow interval between when the surgical technology is too immature to test in randomized studies and when it is so well entrenched that surgeons are reluctant to randomize their patients.
—— Moshe Schein
【译文】由于外科手术的发展如此之快，从因手术技术太不成熟而无法进行随机研究，到因手术流程已固化成熟而不情愿将其患者纳入随机化研究，这两者之间的间距已经非常小了。
【点评】外科医生总是企图以各种理由拒绝进行外科手术的前瞻性随机对比研究。
26.The plural of anecdote is not data.
【译文】多个传闻故事的集合并不是数据。
【点评】一组数据中的每个数据都可被认为是一个故事，但很难根据多个故事的累计得出科学结论，科学的数据应该按照明确规定的协议来有序收集。
27.Trying to compare your numbers with the Mayo Clinic is as farting against the hurricane.
【译文】试着把你的数据与梅奥诊所（Mayo Clinic）的进行比较，那就像对着飓风放屁一样。
28.Nothing spoils good results as much as follow-up.
—— B.Ramana
【译文】没有什么比随访对（研究的）好结果毁坏性更大的了。
【点评】只有经得起长期随访的研究结果才是真正有价值的。
29.It is too bad that we cannot cut the patient in half in order to compare two regimens of treatment.
—— Béla Schick, 1877——1967
【译文】糟糕的是我们不能为了比较两种治疗方案而把患者分成两半。
【点评】的确，临床科研设计常常会受到许多因素的限制。
30.Does absence of proof of efficacy provide proof of the absence of efficacy?
—— John Marshall
【译文】缺乏有效性证据难道就是其无效的证据吗？
31.Do not let the data get in the way of your opinions.
—— Philip Barie
【译文】不要让数据左右了你的观点。
【点评】一方面要尊重数据，同时更应尊重临床。
32.Too good to be true, or too true to be good.
—— William Henenge Ogilvie, 1887——1971
【译文】好得不像是真的，或者太真实而不够好。
【点评】只要不是人为造出来的结果就应该尊重。
33.Meta-analysis is to analysis as metaphysics to physics.
—— H.Harlan Stone
【译文】荟萃分析（meta+analysis）如同形而上学（meta+physics）之于物理学。
【点评】传统物理学是基于计量、测量、计算、实验、实证以获得事实和数据的科学，似乎与靠抽象推论、演绎的“形而上学”水火不容。作者用meta-analysis与metaphysics组词上的巧合类比“荟萃分析meta-analysis”。其实个人理解“相对论”的诞生就得益于“抽象推论和理论演绎”，同理，荟萃分析之于生物医学研究也是有其不可替代的作用的。