现场流行病学的基石 —— 病例定义 刘 明 斌 2010 年 2 月 25 日 南昌市疾病预防控制中心 中国现场流行病学培训项目.

Presentation on theme: "现场流行病学的基石 —— 病例定义 刘 明 斌 2010 年 2 月 25 日 南昌市疾病预防控制中心 中国现场流行病学培训项目."— Presentation transcript:

1 现场流行病学的基石 —— 病例定义 刘 明 斌 2010 年 2 月 25 日 南昌市疾病预防控制中心 中国现场流行病学培训项目

2 主要内容 Contents  为什么要使用病例定义？ Why use a case definition?  什么是病例定义？ What is a case definition?  敏感度和特异度问题 Sensitivity and specificity  构建病例定义 Construction of case definitions  监测系统病例定义应用实例 Applying case definitions  病例一览表 Line Listing  疾病的时间序列 Time sequence of illness

3 为什么使用病例定义 ? Why use a case definition? 医生、护士、实验室和其他人员依法报告 Doctors, nurses, laboratories, and others report notifiable events 资料产生外在变异 Produces extraneous variability in these data 减少变异性以便： Need to minimize this variability to: – 监视趋势 Monitor trends – 评价干预效果 Evaluate the effectiveness of intervention activities – 地区间的比较 Compare areas – 比较结果应真实反应疾病本身的变化和不同 Comparisons must reflect changes or differences in the disease

4 疾病监测的病例定义必须始终保持一致 Case definitions for surveillance need to be consistent over time. 报告病例数 Reported Cases 年份 Quarter year

5 2005 － 2006 年某社区病毒性肝炎每周报告发病率 Viral Hepatitis by week for one county, 2005-2006 乙肝 HBV 未分型肝炎 Unspecified hepatitis

6 2005 － 2006 年某社区剔除社区医院的病毒性肝炎每周报 告发病率 Viral Hepatitis for a county with and without one hospital, 2005-2006 剔除社区医院 社区总报告发病率

7 病例的错误分类可导致流行病学分析偏差 Misclassified cases can bias epidemiologic analysis 暴露 Exposed 病例 Cases 非病例 Not Cases 合计 Total 真实 True 相似疾病 Similar disease 小计 Total 是 Yes 80201009001000 否 No 20 409601000 合计 Total 1004014018602000 暴露组发病率 Rate in exposed 10% 未暴露组发病率 Rate in unexposed 4% 相对危险 Risk ratio （ RR ） 2.5 危险差异 Risk difference 6%

8 完美的病例分类则可给出不同的率 Perfect classification gives different rates 暴露 Exposed 病例 Cases 非病例 Not Cases 合计 Total 真实 病例 True 相似疾病 Similar disease 合计 Total 是 Yes 800 9201000 否 No 200 9801000 合计 Total 1000 19002000 暴露组发病率 Rate in exposed 8%10% 未暴露组发病率 Rate in unexposed 2%4% 相对危险度 Risk ratio （ RR ） 4.02.5 危险差异 Risk difference 6%

9 在消灭某种疾病 X 的过程中与其相似疾病的发病率等于 10 During elimination of disease X the incidence of similar disease = 10 相似疾病 Similar disease 疾病 X Disease X 监测到的疾病总数 Total disease detected

10 在消灭某种疾病 X 的过程中与其相似疾病的发病率等于 1 During elimination of disease X the incidence of similar disease = 1 监测到的病例总数 Total disease detected 疾病 X Disease X 相似疾病 Similar disease

11 什么是病例定义 ? What is a case definition? 一套标准 A set of standard criteria 用于确定是否感染 Used to decide if a person has an infection 保证卫生统计结果是真实可信的 Assures that differences in health statistics are real 根据目的不同而不同 Varies with the purpose

12 不同专家会对同一（群）病人进行相同分类吗？ Will experts classify the same patients the same? 10 位专家会诊 100 位病人 10 experts review 100 patients 每位专家均将他们划分为某病病人或非病人 Each expert classifies them as a case or not a case of a disease X 每位专家会按同样方法对每位病人进行分类吗？ Will every expert classify every patient the same way? Case or not a case?

13 是否可能存在完美专家？ Will one expert be perfect? 假设一位专家审阅 100 位病人的诊疗记录 One expert reviews 100 patient records. 他将如何判定病人与非病人？ A case or not a case of a disease X? 一个月后同一专家重新审阅上述 100 位病人的诊 疗记录又将如何？ One month later the expert reviews the same 100 records. 不同专家将会用同一方法或标准判定每位病人？ Will every expert classify every patient the same way? Case or not a case?

14 一个典型的病例定义（临床诊断）：麻疹 A typical case definition: measles 出皮疹 3 － 7 天 Generalized skin rash of 3 to 7 days duration 体温大于 38 度 Fever > 38C 加上以下一项或多项 Plus one or more of the following – 鼻卡他症状 coryza ( 鼻塞 stuffy, 流涕 runny nose) – 咳嗽 cough – 结膜充血 conjunctival injection

15 敏感度和特异度 Sensitivity and specificity 敏感度 Sensitivity – 能够发现的病例占所有真实病 例的比例 of all the true cases of the disease how many (what percent or proportion) does the method discover? 特异度 Specificity – 能够发现的非病例占所有非病 例的比例 of all the non-cases how many (what percent or proportion) does the method discover?

16 病例定义永远不可能完美 Case definitions are never perfect 按病例定义判定 By Case definition 真实病例 True Cases 合计 Total 是 Yes 否 No 病例 Cases 9020110 非病例 Not cases 108090 合计 Total 100 200

17 敏感度等于真实病例被准确判定的比例 Sensitivity = proportion of true cases correctly identified 按病例定义判定 By Case definition 真实病例 True Cases 合计 Total 是 Yes 否 No 病例 Cases 9020110 非病例 Not cases 108090 合计 Total 100 200 敏感度 Sensitivity = 90/100 = 0.90 = 90% 假阴性率 Percentage false negatives = 10/100 = 0.10 = 10%

18 特异度等于非病例被准确判定的比例 Specificity = the proportion of non-cases correctly identified 按病例定义判定真实病例 True Cases 合计 Total By Case definition 是 Yes 否 No 病例 Cases 9020110 非病例 Not cases 108090 合计 Total 100 200 特异度 Specificity = 80/100 = 0.80 = 80% 假阳性率 Percentage false positives = 20/100 = 0.20 = 20%

19 假定某社区在疾病 X 暴发期间同时存在其它不同疾病 Consider a county with different diseases during an outbreak of Disease X. 与疾病 X 相似 Similarity to disease X 其它疾病 Other diseases 疾病 X Disease X

20 高度敏感的病例定义必将包括很多其它疾病的病例 A case definition that includes all cases will include many other diseases 高度敏感的病例定义 This is called a “highly sensitive” case definition 非病例 Not Cases

21 高度特异的病例定义必将排除包括很多真实病例 A case definition that includes only cases will exclude many cases X 疾病病例 Disease X 高度特异的病例定义 This is called a “highly specific” case definition 非病例 Not Cases

22 病例定义可有多个水平 The case definition may have levels of certainty. 非病例 Not Cases 疑似病例 Suspect 可能病例 Possible 临床病例 Probable 确诊病例 Confirmed 特异度 Specificity 敏感度 Sensitivity

23 细菌性痢疾暴发的病例定义 Case definition for a shigellosis outbreak 非病例 Not Cases 腹泻 Diarrhea ≥ 2 days Diarrhea≥ 2 days + 呕吐 Vomit 或者 OR 发烧 Fever >38C Diarrhea≥ 2 days +Vomit 并且 AND Fever >38C Suspect, or Possible, or Probable + 实验室阳性 Shigella 疑似病例 Suspect 可能病例 Possible 临床病例 Probable 确诊病例 Confirmed

24 敏感度和特异度可随目标的不同而变化 Variability in sensitivity and specificity depends on purpose 诊断和治疗 Diagnosis and treatment of patients. 流行病学调查阶段 Phase of an epidemiologic investigation 确证暴发原因 To identify cause 常规监测或疾病登记 For routine surveillance or registries 内部或者外部使用 For internal or external use

25 暴发期间使用敏感病例定义 During outbreaks the performance of a sensitive case definition improves 相似疾病 暴发疾病 疾病暴发

26 构建病例定义 Construction of case definitions

27 疾病识别标志： Disease Identifiers: 症状 Symptomatology 体征（物理检查） Signs (Physical examination) 临床评价 Clinical evaluations (EKG, EEG, etc.) 不符合其他疾病 Lack of compatibility with other diseases 治疗情况 Response to treatment 实验室检测（病毒分离、血糖等） Laboratory tests (virus isolation, blood sugar, etc.)

28 流行病学标准 Epidemiologic criteria 时间 Time 地点 Place 人 Person 宿主因素（基础疾病） Host factors (underlying illness) 与其它病例的流行病学联系 Linkages to other cases 相关危险因素 Associations with risk factors

29 感染性胃肠炎：病例定义示例 Infectious gastroenteritis: Example case definition 临床 Clinical: – 腹泻（ 3 次以上稀便 x2 天） Diarrhea (3+ liquid stools x 2 days) – 发热 Fever – 无其他原因（药物 — 抗生素） No other cause (drugs -- antibiotics) 实验室确诊 Confirmed Laboratory: – 粪便中发现病原体 Organism identified in stool 流行病学 Epidemiological – 在最短潜伏期内住院 Hospitalized for minimum incubation period – 潜伏期期间在医院 In hospital during incubation

30 使用一个大多数或全部病例具有的症状特征 Use one characteristic that will include most or all cases 腹泻 Diarrhea 发热 Fever 黄疸 Jaundice 出疹 Rash 出疹并发热 Rash+fever 咳嗽并发热 Cough + fever

31 使用在某些病例中发现的更具特异性的症状特征 Use more specific characteristics that are seen in some cases 特殊发现 Specific findings – 杨梅舌－猩红热 Strawberry tongue -- scarlet fever 特殊病症 Pathognomonic finding – 柯氏斑－麻疹 Koplick’s spots = measles

32 使用一组一般性的症状特征 Use a group of common findings 卡他症状 Coryza 咳嗽 Cough 结膜炎 Conjunctivitis

33 其它标准 Other criteria 不能被其它已知疾病所能解释的症状 Symptoms not explained by another confirmed disease or condition 既往健康人所具有的特征 In a previously healthy person

34 川崎病 Kawasaki disease 发热持续 5 天或以上 Febrile illness for 5 or more days 并同时具有以下 5 种症状中的 4 种 With four of the following five findings – 双眼结膜充血 Bilateral conjunctival injection – 新发口腔粘膜破损 Oral lesions (new) – 手指和脚趾破损 Lesions on fingers and toes – 皮疹 Skin rash – 颈部淋巴结病 Cervical lymphadenopathy

35 新生儿破伤风 Neonatal Tetanus 新生儿 Neonate who – 出生时啼哭 Cried at birth – 出生后头两天内能正常吸允 Had normal suck for first 2 days of life 出生后第 3 天至 30 天以内 Developed from 3 to 30 days after birth – 无力吸允 Inability to suck – 出现痉挛 Generalized spasticity – 可有惊厥 With or without convulsions – 死亡 Commonly results in death

36 百日咳 Pertussis 咳嗽并具有以下任一特征 Cough with one of the following characteristics – 咳嗽持续 2 周或 2 周以上 Continues for 2 or more weeks – 阵发 With paroxysms – 咳嗽后呕吐 Posttussive vomiting

37 病例定义应用实例 Applying case definitions

38 举例：人狂犬病监测病例定义的标准 Example criteria for a case definition for human rabies for surveillance 临床标准 Clinical: – 脑炎 Encephalitis – 发病后 10 天内死亡 Death within 10 days of onset 实验室标准 Laboratory: – 头发毛囊、角膜或者脑组织免疫荧光抗体染色阳性 Direct FA of hair follicle, cornea or brain – 抗狂犬病毒 IgG 或 IgM 抗体阳性（未经免疫者） Anti-rabies IgG or IgM in an unvaccinated person 流行病学标准 Epidemiological: – 被已知宿主动物咬伤 Bite from a known reservoir animal species – 被患狂犬病的动物咬伤 Bite from a rabid animal

39 举例：狂犬病病例定义 Example rabies case definition 疑似病例 Suspect – 患脑炎并于 10 天内死亡 Encephalitis with death in ≤10 days 可能病例 = 疑似病例＋ Possible = suspect and – 动物咬伤 animal bite 临床诊断病例 = 疑似病例＋ Probable = suspect and – 被患狂犬病的动物咬伤 bite from rabid animal 确诊病例 = 疑似病例、可能病例或者临床诊断病例＋ Confirmed = suspect, possible or probable – 免疫荧光抗体检测狂犬病毒 阳性 Rabies virus by direct FA from cornea, hair follicle, or brain – 或者抗狂犬病毒 IgM 抗体阳性（未经免疫者） Or anti-rabies IgM in an unvaccinated person

40 根据以下报告的狂犬病例数据，哪些病例符合病例定义？ Some data from reported rabies cases. Which fit the case definitions? E= 脑炎 encephalitis Y = 是 Yes N = 否 No R = 患狂犬病的动物 Rabid Animal O= 其它动物 other animal ND = 没有检测 Not done CI = 角膜印迹 Corneal Impression Br = 脑组织 Brain

41 腮腺炎性脑膜脑炎 Mumps meningoencephalitis 临床标准 Clinical – 脑膜炎或脑炎症状并发热大于 37.5 度 Symptoms of meningitis or encephalitis with fever (>37.5 C) – 脑脊液白细胞计数大于 50 个 /ml 、葡萄糖大于 3,0 um/L CSF with 50 leucocytes/ml y glucose > 3,0 um/L – 前 2 周内患有流行性腮腺炎 Mumps disease within 2 weeks before 流行病学标准 Epidemiologic – 发病前一个月与流行性腮腺炎患者面对面接触 过 Face-to-face contact with a mumps case in the month before onset 实验室确诊 标准 Laboratory confirmation – 抗腮腺炎病毒 IgM 抗体 Anti-mumps IgM – 分离到腮腺炎病毒 Isolation of mumps virus

42 流行性腮腺炎性脑膜炎 Mumps Meningitis 疑似病例 Suspect – 症状加接触 Symptoms plus contact 可能病例 Possible – 症状加流行性腮腺炎 Symptoms plus mumps disease 临床诊断病例 Probable – 疑似或者可能病例加脑脊液检查 Suspect or possible with compatible CSF 确诊病例 Confirmed – 上述三种病例 Suspect, possible or probable – 急性期或恢复期检测到 IgM 抗体或分离到腮腺炎病毒 Anti-mumps IgM or isolation of mumps virus during acute illness or convalescence

43 根据以下病例一览表判定哪些病例符合病例定义 Can you apply the case definition to this line-listing of patients?

44 病例一览表 Example of a Line Listing 1=“Yes”?=“Don’t know” 0=“No” ID # 发病日期 Date of Symptom Onset 症状 Signs/Symptoms 实验室结 果 Labs 人口统计学 Demographics 腹泻 Diarrhea 呕吐 Vomit 发热 Fever >101 Positive stool culture 年龄 Age 性别 Gender 11/2/0411?119M 21/5/04010017M 31/2/04010123F 41/7/04111118? 51/1/04?11118F

45 如何制作一览表 How to Create a Line Listing 及时添加新的确诊病例 Add new cases as they are identified 在调查过程中保持更新所有病例信息 Update case information throughout the investigation 根据情况设置变量数量 Number of variables to include will vary – 变量少利于目测 Fewer will allow for quick visual assessment – 变量多则对评估和描述暴发十分有用 More will be useful for assessment and characterization of outbreak

46 病例一览表应该包含的项目 What should a line listing of cases include? 病例姓名或身份识别号码 Case name or identifying number 症状出现日期（或采样日期） Date of symptom onset (or specimen collection date) 病例定义的组成要素 Components of case definition 接触信息 Contact information 此外还应 包括 May also include additional information such as: – 性别、年龄、种族、职业、危险因素 Age, gender, race, occupation, risk factors

47 疾病的时间序列 Time sequence of illness 易感 Susceptibility 暴露 Exposure 潜伏期 Latency, incubation, induction 前驱症状 Prodrome 亚临床期 Subclinical state 临床期 Clinically apparent disease --- appearance of recognizable symptoms 致残，死亡 Incapacity convalescence, death

48 感染性疾病的典型时间序列 Typical time sequence of an infectious disease ProdromeClinical Illness Convalescence Relapse or Recrudesence Incubation 暴露 Exposure 发病 Onset 死亡 Death 恢复 Recovery 时间 Time 致残 Incapacity 确诊 Diagnosis Latent State 复发

49 Thank you! Any Questions?