堆在储存柜最底层的那堆“废纸”，陈晓红固执地阻拦任何人扔掉。
那是一叠画满“正”字的表格，每一笔表示一个逝去的生命。
1985年，在白求恩国际和平医院医务部工作的陈晓红，把经手的死亡报告单，画“正”字计数。眼下，白纸早已泛黄，表格里的数据被录入计算机里。71岁的陈晓红再也不用画“正”字计数，只要在计算机里“跑一下”，误诊病历的数据会弹出屏幕。
究竟经手过多少份误诊的病历报告？她记不清具体的数字了，“大约30万份吧。”她见证过梅毒在上世纪70年代消失，又卷土重来，误诊率最高时达到60%，如今，所有医生都认识了它，极少误诊。在《临床误诊误治》杂志当主编时，她把当时少有人知的胃食管反流及其综合征等多种疾病，推到公众眼前。
这是一条少有人走的路。就像攀登一座医学高山，分支众多的专科是从正面拾级而上，而研究误诊则是从背面翻越，同样要经过陡坡和峭壁。如今，退休多年的陈晓红还在向上爬。她已经成为中国研究误诊最多的人之一。
每个医生心里都有片墓地
许多双眼睛在关注着误诊研究：出版社时不时询问陈晓红的研究进度，希望免费出书；程序员也加入研究，敲入编程语言，轻而易举地找到误诊疾病之间的相关性；科技公司找上门来，想和陈晓红合作，将她攒了30多年的误诊病历，作用于临床诊断。
可上世纪90年代初，当她和同伴写好第一版《误诊学》时，却被大大小小的出版社拒绝，“医生都要写经验，你这写的是反面”，更何况，从来没有一本医学书籍以误诊为主题。为了让新书顺利出版，她鼓起勇气，拜访当时的医学“大咖”，请他们帮忙写序。
吴阶平向她敞开了门。这个中国泌尿外科先驱者、两院院士、周恩来的医生，丝毫不掩饰地向初次见面的晚辈坦承他犯过的错误：他曾建议一个慢性前列腺炎患者尝试热水坐浴，没多久病人反馈，不起效果，他追问坐浴的方式，病人回答，“不就是洗洗屁股吗？”
说到这里，向来温和的吴阶平蓦地站起来，提高了声量，“这不是怪我吗，没有说清楚。”
“您已经离开临床多年，为什么对这些小事记得如此准确呢？”陈晓红忍不住问。
“这可不是小事。”此后，吴阶平总会叮嘱每个细节，希望不要延误病人的治疗：用什么样的水盆、加多少量的水、怎样不断加水、为什么不能用浴盆或洗浴代替……
他一边回忆，一边来回踱步，甚至把出现误诊的起因揽在身上：“这个责任不怪青年，责任在我们老同志，没有把以往的经验及时传给青年。”
为了支持陈晓红，他挨个通知当时在北京能找到的医学界院士，参加新书出版的研讨会，“这也是我想做的事情，被你们做了，我很激动。”
被誉为“中国外科之父”的裘法祖接到了研讨会的邀请，他回信，“对未应邀到会表示歉意”。后来，他特意约见陈晓红，给她打气，强调误诊的规律值得探讨：国外有人研究犯罪学，不是教人犯罪，是为了避免犯罪；同样，研究误诊不是教人误诊，而是要减少、避免误诊。
那本书总结的误诊规律，如今在计算机的重复核算下得以验证、更新。只不过，原先人工计算的5000份误诊病历，如今扩充到30多万份。《误诊学》甚至被盗版，翻译成繁体字，流行于港台，出版社想要维权，陈晓红不同意——她想让更多人看到这本书，哪怕看的是盗版书。
1995年，陈晓红去《临床误诊误治》杂志当主编。在那时，医疗纠纷开始增多，把许多医生缠得焦头烂额。承认误诊，无异于自找麻烦。但一批老院士、老医生，愿意说一些刺耳的话，把封笔之作留在了这本杂志上。
上世纪70年代初写出130多万字《腹部外科学》，影响当时数万名外科医生的钱礼，早已宣布搁笔。但他看了杂志上的误诊分析后，主动撰写多篇反思误诊误治的文章，希望能“传帮带”，帮助青年医生迅速成长。
中国现代耳鼻咽喉科创始人之一姜泗长，在从医的第55年，回忆起他做过成百上千例耳硬化症手术，牢牢记得唯一失败的一例：1970年左右，手术即将结束时，病人头部突然移动，他拿着镊子的手没来得及避开，将一块组织推进前庭。病人听不见了。
He also recalled a 16-year-old boy with a large maxillary sinus fibroid in 1949.
This is an operation that is expected to go well - Jiang Sichang has always been praised by his peers for his quick and accurate operation.
But the boy eventually died on the operating table due to a sudden hemorrhage, and the blood transfusion was not prepared before the operation.
Afterwards, the boy's father took the initiative to comfort Jiang Sichang, "Dean Jiang has failed, so if others do it, the failure is conceivable."
"It irritates me more than scolding me in person." In the days to come, Jiang Sichang will always think of the boy's childish and cute appearance.
He attributed the failure of the operation to the complacency of his youth at the time.
Jiang Sichang exposed several medical errors in his memoirs.
These bloody lessons motivated him to pay attention to and support misdiagnosis research.
A book "Doctor's Choice" written by foreign doctors mentioned that every doctor has a cemetery in his heart.
Every time a patient dies due to a surgical error, a new headstone is added to the cemetery.
Doctors are also ordinary people, and they need to use mistakes to grow and enrich their experience.
Most of the old academicians who have contributed to magazines and are willing to reveal their scars are now dead.
Chen Xiaohong lamented that these "everyones" who have left traces in the history of modern medicine in China rarely mention their achievements when they look back on their medical careers, and they all talk about the mistakes they have made.
Talking about misdiagnosis is about solving problems, not creating them
Today, computers have helped Chen Xiaohong find out many patterns of misdiagnosis.
Doctors in hospitals at all levels may misdiagnose: young doctors misdiagnosed mostly because of inexperience, and they did not expect it to be another disease; old doctors mostly misdiagnosed because they were too experienced and took it for granted.
Misdiagnosis is not that A disease is misdiagnosed as B disease, but A disease can be misdiagnosed as many diseases, and many diseases can be misdiagnosed as A disease.
Different diseases are intertwined with each other, similar symptoms may lead to many diseases, and the clinical manifestations of common diseases are becoming more and more hidden and individualized.
To make matters worse, today, doctors have less time to make a diagnosis.
Many senior doctors interviewed mentioned this point invariably.
Patients who are sent to the emergency room sometimes die before they can be diagnosed.
Doctors must compete against the disease against time, learn to diagnose decisively, and must diagnose correctly, otherwise it is easy to attract lawsuits.
A dental patient in his 60s, who had filled his teeth many times, still had a toothache. In the end, he had to ask the doctor to pull out the bad tooth.
The doctor did an electrocardiogram on him before the tooth extraction, and found that his toothache was not related to bad teeth, but an old myocardial infarction.
陈晓红发现，急性心肌梗死的病人临床表现可以是肚子疼、肩疼、背疼、牙疼等。2000年，她在误诊误治研讨会上呼吁，当医生发现病人有这些症状时，可以给病人做个心电图，“花七八块钱做心电图，能避免一次误诊。”
扎进误诊病例30多年，陈晓红决定要做一个给临床医生提醒的人，“不要让诊断走弯路”。
有时候，坏消息比好消息更让她兴奋。2008年，她听说，四川发生17例因病人肺栓塞死亡引发的医疗纠纷，要通过尸体解剖认定责任归属。她马上推着编辑出差，去当地核实信息，尽快发稿，“我国尸检率太低了，每一例尸检案例都特别宝贵。”
新的疾病出现了，她和同事们紧跟不舍。2015年，北京协和医院的医生发来文章，统计了2011年至2014年诊治的30例抗NMDA受体脑炎，外院初诊误诊率高达100%。这是个少有人了解的新疾病。全世界范围首例报道这个疾病是在2007年。
这篇文章迅速引起编辑部的重视——必须马上刊登，提醒其他医生重视。慢慢地，各地医院不断公开分享相关的诊疗经验和误诊病例，医生送检意识变强，到了2021年，郑州大学第一附属医院公布了121位确诊抗NMDA受体脑炎的患者，只有43人曾被误诊。
再早些年，汪忠镐院士研究胃食管反流综合征，发现这个病长期被误诊，分散在呼吸科、心血管科、耳鼻喉科、牙科等。这个发现倒逼着编辑部在网站上开辟“关注胃食管反流病”的专题，把汪院士的研究“喊”出去。
在编辑部，每每碰到好文章，编辑总会大声念出来，与同事分享。西南地区某大医院看不好的病人，到了甘肃一家县医院里，马上确诊为黑热病，那是当地的地方性疾病。某些三甲医院诊断流行性出血热，误诊率高于一级医院，因为一级医院的医生更常接触被老鼠咬的农民。
每个地方性疾病都值得重视。早年，编辑部极少收到来自新疆的投稿。陈晓红着急，辗转联系一个新疆医生，邀请医生投稿。慢慢地，来自新疆的稿件变多。
来自全国各地的投稿人，绝大多数是临床一线医生。陈晓红分析，相比基础研究，大多临床医生更容易接触误诊案例，一旦刊登稿件，对医生的职称评定有帮助。也有医生在投稿时说，专家审稿的建议，能帮助他更深刻地认识他的错误，避免再犯。
为了提高稿件水平，上世纪90年代中期，陈晓红把白求恩国际和平医院各个专科最资深的退休主任，聘到编辑部，共同评审从全国各地寄来的投稿。
那些在临床上干了大半辈子的老主任，对这个新任务兴致勃勃。有人常常一边读稿一边嘟囔，“我在临床时见过这个案例”；有人看到投稿写得乱，忍不住上手逐句修改；还有人读着读着就拍桌而起，“这简直是草菅人命！”
有时遇到连他们也没见过的案例，老主任们会一遍遍讨论，翻书，找答案。编辑部的几个大书架，很快塞满了最新的医学书。
许多医生说，误诊在推动着医生，进一步认识人类的身体。其实，《临床误诊误治》的创刊人冯连元，最初也是为了汲取同行的误诊经验，才创办了杂志。当时，中国消化病学奠基人张孝骞支持了他的想法，但也提出担心，“这个名字会不会惹事？”
思前想后，冯连元提出解决办法：文章隐去患者、医生的名字，并适当修改一些无关紧要的细节，以免暴露患者的个人信息。“办这个杂志，是要解决问题，而不是制造问题。”
误诊是系统性难题
30多年过去了，现代医学在精进：机器人取代了人，看CT影像；3D打印能打印出人体组织；以往，手机影响心脏起搏器工作，但现在两者直接相连，手机追踪患者心脏健康……但误诊依然每天在临床上发生。
北京大学人民医院呼吸科医生何权瀛认为，要解决误诊，是系统性工程。一方面，人类所掌握的医学知识越来越多，越来越深，医学院的学制从5年制变为6年制，再增加到8年制，但医学生到了临床，依然感觉知识不够用。医学专科越分越细，许多医生只专注研究一个专科的其中一种病，当患者有多种疾病，就容易漏诊。
“就像用钻打洞，越钻越深，最后，看不见洞旁边的地方。”何权瀛比喻。
他研究睡眠呼吸暂停20多年，发现这个疾病可能会引发冠心病、高血压、糖尿病等其他疾病。问诊时，他喜欢给病人画图，在纸上列明各种疾病，刨根问底地问，再寻找每个疾病之间的关联。但有病人不愿回答过多的提问，“你这个大夫这么烦人，给我开药不就完了。”
医患互相不信任，是导致误诊的原因之一。有的未婚女性肚子疼，故意向医生隐瞒了个人性史，导致医生找错病因，而没想到是宫外孕。有时，延误治疗的后果将伴随这些女性一生——切掉一半输卵管，相当于降低一半受孕机会。
患者如今能从各个途径获取医学知识，但何权瀛认为，公众的卫生知识水平依然不够，更重要的是，患者不如过去那般能够包容、允许医生误诊。这也造成尸检率减少，当家属通过尸检报告发现误诊，往往会和医院打官司。
前几年，《临床误诊误治》编辑部每年都会迎来几拨儿不速之客——患者拿着一叠病历，希望编辑部帮忙评评理，“你们判断一下，我们这是不是误诊了？”
陈晓红总结，这种态度源于知识不对等。当病人躺在床上接受检查时，他在仰视医生；但当医生也有看不懂疾病、下不了判断的时候，他也在仰视神秘复杂的医学宇宙。
误诊研究是医学发展的同行者。检查技术的发展一度帮助医生判断正确，但陈晓红发现，过度依赖检查机器，成了新兴的误诊原因。
30年来，误诊的概念逐渐变大，对医生的要求更严格了。过去，医生诊断错了疾病才算误诊，如今即使诊断正确，但是治疗用药不恰当，或是初诊判断错误，也算误诊。
世界卫生组织曾公布，临床医学的平均误诊率为30%，其中80%医疗失误是由于思维和认识错误导致的。根据陈晓红搜集的误诊文献，近30年，文献误诊率维持在30%左右，没有明显下降，但文献误诊率不代表真实误诊率。
“很难统计准确的临床误诊率。”何权瀛解释，我国临床死亡尸检率低，而尸检是最好、最准确发现误诊的方式。更何况，如今没有统一的计算误诊率的标准。
那个在信息闭塞时代，为了获取经验而办刊的冯连元，如今已经在临床工作了近40年，攒足了经验。但他发现，即使有了经验，也会误诊。
For common diseases, doctors need to use medicines and treatments according to the corresponding diagnosis and treatment guidelines, but the standards in the guidelines cannot completely match each patient.
For example, according to the guidelines, patients with gas poisoning should be infused with 200 mg of niacin, but Feng Lianyuan has encountered an experience beyond imagination: 2000 mg of niacin, 10 times the guideline, was used to save the patient.
There is also a patient in Shanghai who has had a headache for a month and hopes to lower his blood pressure, but his high pressure is 115 and his low pressure is 75, which are within the normal range.
Several hospitals refused to prescribe him antihypertensive drugs.
Feng Lianyuan asked and found that the patient's blood pressure was lower than the normal blood pressure of ordinary people all the year round, so he broke the guidelines and prescribed antihypertensive drugs.
The patient's headache gradually eased.
Going beyond the guidelines for medication is an extremely test of the doctor's courage.
Today, in order to avoid excessive medical treatment, the system can automatically identify the amount of medicine a doctor has taken, and if it is found to exceed the guidelines, the doctor will be fined.
Moreover, once a medical dispute occurs, the attribution of responsibility will be determined according to the guidelines, and the doctor must explain clearly, why not prescribe drugs according to the guidelines?
Feng Lianyuan concluded that the formulation of the guidelines is a great progress, setting standards for doctors across the country, but the appropriate dosage of medication for each person is different, and individual differences must be understood.
"It's like putting 100 screws into screw holes, some screws have to be backed up with paper to fit accurately," he said.
Therefore, Feng Lianyuan, who has been retired for many years, is still studying misdiagnosis.
He cited mathematical models, made up for inaccuracies, and found the piece of paper that was holding the screws.
Doctors sit together regularly and correct themselves
For some years, Meng Qingyi would appear on the podium of the 301 Hospital as the chief physician to give the first lesson.
Sitting at the bottom are advanced doctors from all over the country, all of them are technical experts who "stomp their feet three times" in the local hospital.
His first lesson was about misdiagnosis.
He didn't talk about those high-level intractable diseases, but started from the stories of "misconceptions" in daily clinical work.
After an elderly man went to bed at night, his family couldn't wake up, so he was sent to the emergency room overnight.
The doctor judged that he must be seriously ill in a coma, so he ordered various examinations, and urgently called a doctor who only worked during the day for MRI examinations, but could not find the cause.
Until 6 am, the old man woke up suddenly and looked at the doctor facing him with a surprised expression, "Why am I here?" It turned out that he just took two tranquilizers.
He also gave an example that fever patients during the high-incidence period of influenza are easily misdiagnosed as influenza.
In fact, the patient may have other diseases, such as urinary infection, meningitis, tetanus attack and so on.
He still remembers that, in 1994, when he graduated with a doctorate and just entered the emergency department, a young man in a coma came to the emergency department.
The doctor spent more than 4 hours doing various examinations, but still could not find the cause of the coma. Finally, an expert from the hyperbaric oxygen department was invited from an outside hospital to know that the patient was poisoned by carbon monoxide.
He later read from an English book, "Young patients in non-traumatic coma, first consider poisoning." These experiences finally played a role in later work.
"Doctors need to be alert at all times and keep asking themselves questions. Why is this behavior? Will it be misdiagnosed?" He described that doctors' work status is like walking on thin ice, and they must force themselves to break through clinical thinking and cognition.
This course later became the ace class.
He often hears complaints from patients, "This disease was not diagnosed in the county hospital. You will know why when you come here." Meng Qingyi explained to his colleagues, "The doctors in the county hospital are not poor, and your disease has no symptoms in the early stage." , and his correct diagnosis is based on the mistakes of the predecessors, and the diagnosis of the predecessors cannot be belittled.
There are still many mysteries in the medical field waiting to be solved.
Meng Qingyi has been researching difficult diseases for many years, and he has encountered some diseases that are "unreliable". They are neither like existing diseases, nor belong to a certain specialty, nor are they recorded in any medical books at home and abroad.
In many cases, the fight against such diseases requires a series of failed attempts before finally winning.
At Peking University People's Hospital, every two months, a group of doctors gather to share the "puzzles" encountered in clinical practice.
Doctors from various hospitals took turns to come to power and tell the truth about the difficult cases they had experienced.
They are also open to peer scrutiny as they share their experiences.
For difficult cases, how to find the final answer.
In this process, there are also some cases where the initial diagnosis was missed, or there was a phenomenon of seeing the trees but not the forest.
Mi Yuhong, director of the Emergency and Critical Care Center of Beijing Anzhen Hospital Affiliated to Capital Medical University, is a frequent participant of this seminar.
They also sometimes have serious discussions about deaths.
Even nitpicking, analyzing the details of the entire diagnosis process, and investigating whether there are imperfections?
Or, where can you do better next time?
This ability to self-correct is too important for emergency physicians.
Mi Yuhong has studied pulmonary embolism for 17 years, and has been exposed to at least thousands of patients with pulmonary embolism. The clinical manifestations of each patient are different, not only the symptoms mentioned in the guide.
More importantly, in the emergency department, the clinical manifestations of three fatal emergencies, namely aortic dissection, pulmonary embolism, and acute myocardial infarction, are similar, independent of each other, and sometimes cause and effect each other, which is extremely easy to confuse and misdiagnose.
The treatment methods of these three diseases are contrary to each other, and they also test the judgment, experience and medical skills of emergency doctors.
Such seminars have been held for 60 sessions so far, attracting emergency physicians from major hospitals in Beijing, and now specialists are participating.
They sit together regularly, discuss each other's misdiagnosis experiences without barriers, and try to avoid them in the future.
Mi Yuhong remembered that once, a doctor mentioned a case on the podium. The patient had been treated in a hospital in Beijing before, and was then transferred to another hospital to be diagnosed.
After stepping down, the doctor walked in front of her calmly and reminded softly, "What I just said (misdiagnosis) is your hospital."
let the doctor say
There are many doctors who dare to face misdiagnosis and missed diagnosis, but few doctors dare to speak up.
Mi Yuhong once contributed to the journal "Clinical Misdiagnosis and Mistreatment", writing about the process of contacting a patient with a missing right upper pulmonary artery in the emergency department and being misdiagnosed as pulmonary embolism at the initial diagnosis.
The driving force behind her bold announcement is that she is confident in her professional level. She wants to take this opportunity to remind her colleagues, "I have been studying pulmonary embolism for 17 years. I know what is the common difficulty for everyone in the emergency circle?"
But for those younger doctors, it takes a lot of courage to speak openly about a misdiagnosis.
In fact, what young doctor did not grow up with the constant correction of mistakes by the superior doctor or director?
Meng Qingyi understands the concerns of young doctors.
For young doctors, misdiagnosis is a very painful thing, and they will feel frustrated. "In the past, big experts found that young doctors misdiagnosed and lost their temper, but they would drop the medical record folder." But gradually, young doctors have more experience. , thick-skinned, without the director's reminder, he will go to the book.
Meng Qingyi emphasized that the vast majority of misdiagnoses have no consequences.
This is similar to the conclusion of Chen Xiaohong's research on misdiagnosis. In the past 30 years of misdiagnosed medical records, most of the misdiagnoses will not affect the patients, and only a few will cause death or disability.
But most doctors are more willing to close the door and discuss misdiagnosis in the industry. "This is not a good thing (for patients) after all."
Out of conscience, He Quanying is willing to speak out publicly and be a bridge between the public and the medical community.
He found that in recent years, very few officials or doctors were willing to openly discuss misdiagnosis, and misdiagnosis and mistreatment were not included in the standards for measuring the level of doctors and hospitals.
“误诊已经成为老生常谈又避而不谈的话题了。”何权瀛说，医学界不该掩耳盗铃，因害怕惹上官司，影响名誉，而忽视不提误诊，这会影响公众对误诊的关注和理解。
这种避之不谈的氛围，也影响了陈晓红的研究。她曾经在1999年成立误诊误治研究会，得到许多院士支持，但后来，有人担心这个名字会惹来麻烦，改为“医疗质量研究会”。
早年，在医学期刊发论文，能帮助医生评职称，这一度让陈晓红不愁缺稿。当时，一位行内人评价，其他医学杂志是小白鼠期刊、兔子期刊，只有《临床误诊误治》在讲人。但最近几年，评价医学杂志的标准改变了，基础研究、课题研究更容易受到重视。《临床误诊误治》杂志里，与误诊有关的文章越来越少，页数越来越薄。
陈晓红开始把研究重心挪到误诊大数据研究上，她舍不得放下误诊研究。她看到每年仍有3000多篇误诊的文献，刊登在各大平台，“如果我不干，这些珍贵的文章就死了，它们需要有人来唤醒、收集。”
有医生在短视频平台上介绍自己的误诊故事。陈晓红兴奋地找来团队的年轻人，“我该如何把这个病历扒下来？”她最新的苦恼是，文献资料记录了相对滞后的病例，有些医生写的甚至是数年前、十年前的案例，“太旧了”，跟不上当前医疗的发展。
她唯一想到的解决方案是，鼓励更多医生敢说，勇敢真诚地分享临床上的误诊案例。只有这样，才能广泛地第一时间搜集到最新的误诊案例。
陈晓红依然走在这条少有人走的路上。她一点儿也不感觉孤独。她招聘一批年轻人，不仅包括医学生，还有计算机专业毕业的程序员。她能看到前方最光亮的地方，是医生接诊时，随时用上她攒了30多年的误诊病历、规律。
支持她的人，还包括一位马来西亚的华人医生。他开了一间小诊所，从1998年就订阅了杂志。前几年，他专程来找陈晓红，希望购买、使用这个数据库。
Even now, there is still controversy in the medical community on "whether misdiagnosis is learned or not".
Some doctors insist on misdiagnosis without knowledge, just a summary of doctors' scattered clinical experience, "not all phenomena are called science".
However, Meng Qingyi believes that the law of misdiagnosis is complex and profound, which cannot be written down in textbooks, and cannot be refined by some specialists.
"Misdiagnosis is advanced knowledge." Meng Qingyi said, it should become the crown jewel of medical research.
Wei Xi, reporter from China Youth Daily and China Youth Daily