橫斷面研究

1、研究生課程,1,Ch3 Research Methods of Epidemiology (Ⅰ)　Descriptive study,研究生課程,2,Questions：,常住居民健康狀況（常見/多發(fā)病、亞健康等）？疾病的早期篩檢(screening)及診斷(diagnose)？醫(yī)院感染(hospital infection)發(fā)生情況（分布特征）？可能存在的主要影響因素？采取措施后效果如何？如何合理安排醫(yī)院各科的床位、醫(yī)護人員等

2、？患者對醫(yī)院服務(wù)(hospital service)的滿意程度？某慢性病(chronic disease)對患者生活質(zhì)量(quality of life)的影響？關(guān)于社區(qū)獲得性肺炎(Community-acquired pneumonia )？……,研究生課程,3,,Example,研究生課程,4,。,研究生課程,5,根據(jù)日常記錄資料或通過特殊調(diào)查所得到的資料（包括實驗室檢查結(jié)果），對被研究人群的疾病或健康狀況進行三間分布特征描

3、述的研究，它是分析性研究的基礎(chǔ)。,Concept of descriptive study,研究生課程,6,Commonly used data,demography datadeath datadisease dataother,研究生課程,7,,【Characteristic】屬觀察性研究observation無須事先設(shè)立對照組no control不能分析暴露與效應(yīng)之間的因果聯(lián)系cannot analysis caus

4、al association,研究生課程,8,【Category】橫斷面研究 cross-sectional study生態(tài)學(xué)研究 ecological study常規(guī)歷史資料分析 analysis of history data個案調(diào)查 case investigation病例報告 case report,研究生課程,9,概念　特點　研究目的與用途　種類　設(shè)計與實施　優(yōu)點與局限性,橫斷面研究cross-secti

5、onal study,研究生課程,10,橫斷面研究又稱現(xiàn)況研究或患病率研究（Prevalence study） 是對某人群在某一時間斷面的有關(guān)因素及健康狀況進行的調(diào)查研究，從而客觀反映有關(guān)因素與疾病的分布及二者之間可能存在的關(guān)系。,一、概念,研究生課程,11,屬觀察性、描述性研究　無事先設(shè)立的對照組　只能反映某一特定時點的情況　因果并存，不能確定因果關(guān)系,二、特點,研究生課程,12,⑴ 了解疾病或健康狀況的分布特征

6、 ⑵ 了解分布特征與疾病之聯(lián)系，提供因果線 索或建立病因假設(shè) ⑶ 考核防治措施效果（重復(fù)橫斷面研究） ⑷ 了解疾病現(xiàn)況，合理分配衛(wèi)生資源 ⑸ 衡量國家或地區(qū)的衛(wèi)生水平和健康狀況 ⑹ 進行疾病監(jiān)測，了解疾病的長期變化趨勢 ⑺ 早期發(fā)現(xiàn)病人或高危人群,三、研究目的和用途,研究生課程,13,四、種類,普查（Census） 調(diào)查特定時點或時期、特定范圍內(nèi)的全部人群(總體) 抽樣調(diào)查（Samp

7、ling survey） 隨機抽樣，調(diào)查特定時點、特定范圍人群的一個代表性樣本，以樣本統(tǒng)計量估計總體參數(shù)所在范圍,研究生課程,14,1、普查（census）目的：　?、僭缙诎l(fā)現(xiàn)和治療疾病　?、诹私庖咔?，早期預(yù)測疫情　?、劢⑸碇笜说恼Ｖ怠　、芰私饧膊』蛏順藴实牡貐^(qū)、年齡、性別、民族、職業(yè)分布，為進一步研究提供線索,研究生課程,15,優(yōu)點：信息全面，準確性好缺點：花費大，精確性難以保證；當患病率低、

8、 或無簡易敏感的診斷方法、無切實有效 的治療方法時不適用適用癥：應(yīng)用原則(同篩檢) 成本-效益分析后再確定,研究生課程,16,應(yīng)用原則：檢測手段靈敏、可靠被調(diào)查疾病的患病率較高（或是當?shù)刂饕l(wèi)生問題）疾病進行性發(fā)展，自然史比較明確有較長的領(lǐng)先時間早期治療有效，可明顯改善預(yù)后,研究生課程,17,2、抽樣調(diào)查（sampling survey） 優(yōu)點：快速、高效、花費少、較為精確

9、缺點：信息存在誤差（隨機或/和系統(tǒng)的） 準確性不如普查；不適用于患病率低、 變異較大的情況；設(shè)計、實施及資料 分析較復(fù)雜 適用癥：快速評價應(yīng)用原則：樣本具有代表性和足夠量；調(diào)查方 法可靠,研究生課程,18,（一）研究目的（二）研究類型與方法（包括抽樣方法）（三）研究對象及樣本含量（四）資料收集（研究指標，調(diào)查表設(shè)計）（五）偏倚與控制（六）資料分析（統(tǒng)計方

10、法、分析提綱）（七）可行性分析（人力、物力、財力等）,五、設(shè)計與實施,研究生課程,19,（一）明確研究目的（二）研究類型和方法,由研究目的決定研究類型和方法觀察法與實驗法普查與抽樣調(diào)查抽樣方法,研究生課程,20,抽樣方法,非隨機抽樣non-random sampling 　選擇樣本時，加入研究者主觀因素，使總體中每個個體被抽取的機會不均等。,隨機抽樣non-random sampling 　遵循隨機化原則，保證總體中每一

11、個個體都有同等機會被選入作為研究對象。,,研究生課程,21,隨機抽樣方法： 單純隨機抽樣（simple random sampling ） 系統(tǒng)抽樣（systematic sampling） 分層抽樣（stratified sampling） 整群抽樣（cluster sampling） 多級抽樣（multi-stage sampling）　　隨機誤差與系統(tǒng)誤差（偏倚）　　樣本大小的估計,研究生課程,22,目標

12、人群 target population 納入標準 樣本大小估計 主觀因素：α值和（1-β）值 抽樣方法 客觀因素：研究指標的變異程度 預(yù)計值（如患病率，陽性率等）,（三）研究對象,研究生課程,23,計量資料樣本大小估計公式,計數(shù)

13、資料樣本大小估計公式,t:統(tǒng)計學(xué)上的t值 d: 容許誤差 P：某病現(xiàn)患率 Q=1–P,研究生課程,24,（五）資料收集,通過測定或檢查的方法收集　　　盡量采用客觀、定量的研究指標和準確可靠的檢測方法?！≈苯佑谜{(diào)查表詢問研究對象,研究生課程,25,（六）偏倚與控制,,偏倚是指調(diào)查或研究結(jié)果與真實情況不符，或樣本統(tǒng)計量不能代表總體參數(shù)所在的范圍，是

14、一種系統(tǒng)誤差。產(chǎn)生原因：,主觀選擇研究對象任意變換抽樣方法調(diào)查對象不合作或因種種原因拒絕參加調(diào)查到的對象均為幸存者，無法調(diào)查死亡者回答不準確或回憶不清測量誤差,研究生課程,26,嚴格遵照抽樣方法要求，確保隨機化原則提高研究對象的依從性和受檢率正確選擇測量工具和檢測方法培訓(xùn)調(diào)查員，統(tǒng)一標準和認識作好資料的復(fù)查復(fù)核工作選擇正確的統(tǒng)計分析方法，辨析混雜因素,偏倚的控制,研究生課程,27,（七）資料分析,檢查資料的完整準確

15、性，補缺、補漏，刪除重復(fù)，糾正錯誤對疾病或某健康狀態(tài)按規(guī)定的標準歸類核實按不同空間、不同時間、不同人群描述分布特征和差異，進行顯著性檢驗可按是否暴露于研究因素或是否有病進行分組，從而進行比較分析,研究生課程,32,優(yōu)點： 1、樣本來自人群，研究結(jié)果有較強的推廣意義 2、調(diào)查人群中有自然形成的同期對照，具有可比性 3、可同時觀察多種因素 4、無醫(yī)德問題,六、優(yōu)點與局限性,研究生課程,33,缺點及局限性： 1

16、、疾病與因素同時存在，難以推斷因果關(guān)系 2、只能獲得患病率，無發(fā)病率資料 3、潛伏期或緩解期患者易被誤診而產(chǎn)生偏倚 4、一般只適用于對慢性病的研究 5、可產(chǎn)生選擇性偏倚和信息偏倚,34,定義 目的 研究方法 優(yōu)點 缺點,生態(tài)學(xué)研究ecological study,35,描述性研究的一種在群體水平上研究某種因素與疾病的關(guān)系以群體為觀察和分析單位描述不同人群中某因素的暴露狀況與疾病的頻率

17、，分析該暴露因素與疾病的關(guān)系,一、定義,36,提供病因線索，建立病因假設(shè)評估人群干預(yù)措施的效果,二、目的,37,生態(tài)比較研究ecological comparison study 觀察不同人群或地 區(qū)某種疾病的分布，然后根據(jù)疾病分布的差異，提出病因假設(shè) 生態(tài)趨勢研究ecological trend study 連續(xù)觀察不同人群中某因素平均暴露水平的改變和 (或)某種疾病發(fā)病率、死亡率變化的關(guān)系，了解

18、變動趨勢，比較暴露水平變化前后疾病頻率的變化情況，判斷某因素與某疾病的聯(lián)系。,三、研究方法,38,經(jīng)濟 ，出結(jié)果快提供病因未明疾病的病因線索在個體劑量無法測量時，是唯一可供選擇的方法適用于研究因素暴露變異范圍小，較難測量暴露與疾病的關(guān)系人群干預(yù)措施的評價及估計疾病發(fā)展趨勢,四、優(yōu) 點,39,生態(tài)學(xué)謬誤 ecological fallacy 難以控制混雜因素存在多重共線性問題難以確定因果聯(lián)系,五、缺 點,40,個案調(diào)查c

19、ase investigation,對個別發(fā)生的病例、病家及周圍環(huán)境進行的流行病學(xué)調(diào)查，多用于傳染病及不明原因疾病的調(diào)查。通過流調(diào)，了解分析傳染來源、接觸者以及可能的傳播途徑掌握疫情，為控制疾病提供資料　　　　　　　　　　如：SARS流調(diào),41,病例報告 case report,對臨床上某種罕見病的病例進行詳細的介紹，以引起同行的關(guān)注。有助于發(fā)現(xiàn)與識別新疾病或不良反應(yīng)　　如：AIDS的發(fā)現(xiàn)有利于闡明疾病的發(fā)病與治療機制　

20、　如：“氟烷肝炎”有助于闡明疾病的罕見臨床表現(xiàn),42,橫斷面研究醫(yī)學(xué)文獻的評價原則1、文獻中欲闡述的問題是否可用橫斷面研究？2、是何種橫斷面研究？3、采用了何種抽樣方法？調(diào)查對象的代表性如何？4、調(diào)查是否在同一個較短的時期內(nèi)完成？5、是否考慮研究中可能出現(xiàn)的偏倚及解決辦法？6、所采用的研究指標及統(tǒng)計分析方法是否正確？7、研究所得出的結(jié)論是否客觀？科學(xué)性如何？,研究生課程,43,Questionnaire Design,

21、44,研究生課程,Elements of Questionnaire Design,Identifying the questionnaire contentWriting or selecting questions to measure variables of interestConstructing the questionnaire,45,研究生課程,A Questionnaire Is More than the Q

22、uestions!,Other Elements Opening/cover pageInstructions (skip patterns, probes, optional wording)Introductions to questionsNeeded definitions and explanationsTransitions,46,研究生課程,Before Designing a Questionnaire,De

23、cide the study’s purpose (aims, research questions, hypotheses)Identify the variables to be measuredDevelop a preliminary analysis planDecide the data collection mode (e.g., interview, paper/pencil, computer-assist

24、ed),47,研究生課程,Make a Variable List,Write or obtain questions to measure the variablesUse existing measures whenever possible!Reliability and validity information may be available for existing measuresAllows comparison

25、of results across studiesAllows estimation of trendsAdds to cumulative body of methodological experience with survey items,48,研究生課程,Objectives When Writing Questions,To get reliable and valid reports of respondents’ ex

26、periencesGood survey questions provide consistent (reliable) and accurate (valid) measuresWhen 2 respondents are in the same situation they should answer the question the same way.The answers to questions should c

27、orrespond to what they are intended to measure.,49,研究生課程,Reasons Why Respondents Report Events with Less than Perfect Accuracy,They do not know the informationThey cannot recall it, although they do know itThey do no

28、t understand the questionThey do not want to report the answer in the context in which they are being asked it.,50,研究生課程,Considerations When Writing Questions,Type of question Response formatsQuestion wording,51,研究

29、生課程,Types of Questions,Open-ended questionsClose-ended questionsOrdered response categoriesUnordered response categoriesPartially close-ended questions,52,研究生課程,Open-ended Questions,Responses are not provided to th

30、e respondent.Advantages:Researcher does not need to know universe of possible answersRespondent not influenced by specific alternatives suggestedRespondent can reveal what is most salientUseful in exploratory work

31、Can be used to build rapport in interview,53,研究生課程,Open-ended Questions,Disadvantages:Effort required of respondentRespondents may vary in ability/willingness to articulateRespondents may be reluctant to reveal detail

32、ed information or socially unacceptable opinions or behaviorsLarge amount of information may be revealed, information may be vague or irrelevantDifficulties in recording and in reducing and coding material,54,研究生課程,Clo

33、se-ended Questions,A list of acceptable responses is provided to the respondentAdvantages:Easier for respondentCommunicates same frame of reference to respondentsStandardizationLess variability in interviewer perfor

34、manceLess time to administer and record response,55,研究生課程,Close-ended Questions,Disadvantages:Need to know appropriate response categories in advance Lack of spontaneity permitted respondentRespondent may be forced i

35、nto an unnatural frame of referenceMay suggest response categories respondent has not thought ofRespondent may not feel as involved or motivated by questionnaire,56,研究生課程,Response Formats,Multiple categories that exhau

36、st all meaningful answers and are mutually exclusiveDuring a typical work week (40 hours), how many hours do you spend on engaging in interpersonal relation:_____less than 10 hours_____10 to 19 hours_____20 to 29

37、 hours_____30 to 39 hours_____40 hours or more,57,研究生課程,Ordered Response Formats,Response categories are ordered along a gradient. Examples:Strongly agree to strongly disagree (3 to 7 point scale; include or not inc

38、lude a neutral response category)Excellent, Good, Fair, PoorNumerical rating scalesNoCompleteConfidenceConfidenceat All______________________________ 1 2 3 4 5 6 7 8 9

39、 10,58,研究生課程,Unordered Response Format,No single dimension underlies response categories. Respondent must evaluate each. Example:Which one of the following do you think is most responsible for the long waiting period

40、in the emergency room of this hospital? (Choose only ONE answer.)Low staff moraleThe low priority given to the emergency department by the hospital administration.The low priority given to the emergency department by

41、the medical staff.The large number of people using the emergency room.,59,研究生課程,Partially Close-ended Questions,Answer choices are provided and respondents have the opportunity to create their own responses. Example:Wh

42、at is your position at this school?Classroom teacherPrincipalGuidance counselorNurseOther position (Please specify:____________),60,研究生課程,Question Wording - DO,Make the wording clear and unambiguous Use common voca

43、bularyUse vocabulary appropriate for your audienceWrite questions like people talk not like people writeOnly include one idea per questionInclude both sides of the continuum in stemMake the frame of reference approp

44、riate and explicit,61,研究生課程,Question Wording - DO,Use techniques for enhancing recall Shorten the reference periodUse landmarks to aid datingProvide a helpful contextProvide cues to stimulate recallAsk about typical

45、 behavior,62,研究生課程,Question Wording – DON’T,Use double-barreled questionsWhere do you go to get information about HIV risk and to exchange needles?When I get ill, I know it is because I have not been eating right or

46、 getting proper exercise.Can you tell me when you last visited the clinic?,63,研究生課程,Question Wording – DON’T,Use leading questionsDo you agree that all children should be covered by health insurance?With economic

47、 conditions the way they are these days, it really isn’t fair to have more than one or two children.,64,研究生課程,Question Wording – DON’T,Use double negativesLack of immunization for smallpox is not a problem in the U.S.

48、Most sex workers have little chance of not getting a sexually transmitted disease.Use discriminatory or culturally insensitive wordshusband/wife vs. partner,65,研究生課程,Question Wording – DON’T,Use incomplete wording

49、Age?What was your age on your last birthday?Reason last　　What was the medical saw doctor?　　reason for which you most recently went to a doctor?,66,研究生課程,Question Wording – DON’T,Provide

50、incomplete or overlapping response categoriesWhere have you received health care in the past 12 months?____Student health____UNC Hospitals____Private MD in Chapel HillUse check-all-that-apply format,67,研究生課程,Qu

51、estion Wording – DON’T,Ask respondents to make unnecessary calculations.Out of 100 students your age, how many to you think smoke cigarettes once a month?Use loaded questions or loaded words　Have you ever shop lift

52、ed anything?,68,研究生課程,Practical Standardsfor Evaluating Questions,Is this a question that will mean the same thing to everyone?Is this a question that people can answer?Is this a question that people will be willing

53、 to answer, given the data collection procedures?,69,研究生課程,Constructing the Questionnaire:Putting the Questions In Order,Beginning – inviting, interesting, non-threatening questionsMiddle – most important, put difficul

54、t and sensitive toward endClosing – easy questions again, often routine, background,70,研究生課程,Question Order,Psychological not logical orderChronological order where appropriateGroup questions by subjectWithin a group

55、, group questions with like response categoriesGeneral questions before specificBe aware of order effects between opinion questions,71,研究生課程,Question Design,Include instructions, as needed, with questions - not at the

56、beginning of the questionnaireClearly differentiate response categories from questionsBe consistent in placement of answer boxesAsk one question at a time: don’t stack side-by-sideNumber questions consecutively and s

57、imply from beginning to end,72,研究生課程,Other Elements,Title/cover pageGeneral instructions Identifiers (e.g., respondent ID)Transitions,73,研究生課程,Summary,Decide what information (variables) is neededDraft or obtain q

58、uestions to elicit that informationPut questions in meaningful orderAdd other elements of questionnairePretest questionnaireRepeat,研究生課程,74,調(diào)查表的設(shè)計1、原則：必需項目一個不少，不需項目一個不多。2、類型：開放式，即問答題或填空題形式 　　 封閉式，即選擇題形式

59、 混合式，包含上述兩種形式3、結(jié)構(gòu) 封面信 指導(dǎo)語 調(diào)查問題和備選答案

60、 調(diào)查者項目,,研究生課程,75,4、調(diào)查問題數(shù)量： 5、調(diào)查問題的順序： 先易后難 先客觀后主觀 先一般后敏感 同類問題在一起6、問題及答案設(shè)計 應(yīng)圍繞假設(shè)，避免無關(guān)問題 應(yīng)符合實際情況 要具體化和可操作

61、應(yīng)符合窮盡性和互斥性,研究生課程,76,吸煙情況調(diào)查表1、您曾吸過煙嗎？ 吸過□ 不吸□2、吸哪一種煙？ 有濾嘴□ 無濾嘴□3、每天吸煙量： 支4、吸入深度： 深吸□ 一般吸□ 不吸入□5、您初次吸煙年齡是 歲6、如果已戒煙，戒了 年7、請回憶您的吸煙史： 偶吸 年 20