醫(yī)藥-臨床-護(hù)理研究設(shè)計(jì)概覽ppt課件

1、歡迎進(jìn)入學(xué)習(xí) 課堂研究設(shè)計(jì)概覽The aim of research should be to establish the truth and research design aims to minimise or exclude the threats to the internal validity of the study.There are only a handful of ways to do a study properly but a thousand ways to do it wrong. - Sackett (1986)偏倚Bias與真值的系統(tǒng)偏差Sackett DL.

2、Bias in analytic research. J Chron Dis 1979; 32: 51 常見偏倚選擇性偏倚所選擇的研究對(duì)象（樣本）系統(tǒng)性地有別于所要推斷的總體人群。例如志愿者偏倚、電話調(diào)查介入偏倚過多關(guān)照試驗(yàn)組導(dǎo)致療效高估、病人依從性差導(dǎo)致療效低估隨訪偏倚失訪病人的特征有別于保持隨訪病人的特征失訪可能與治療有關(guān)知情偏倚醫(yī)生如果知道病人接受何種處理將會(huì)影響他對(duì)病人病情的判斷測(cè)量偏倚對(duì)所關(guān)心結(jié)局的判斷帶有主觀性，則判斷往往會(huì)引人而異混雜Confounding分組處理因素與所關(guān)心的結(jié)果同時(shí)受到另外一個(gè)因子（混雜因子）的影響 例如：大醫(yī)院的病死率高于小醫(yī)院 選擇大小醫(yī)院和病死率高低均與

3、疾病嚴(yán)重程度有關(guān)疾病嚴(yán)重程度即混雜因子隨機(jī)化醫(yī)學(xué)研究中的隨機(jī)化平衡了比較的兩組間顯在的和潛在的混雜因素的影響，因此是控制混雜的重要手段在設(shè)計(jì)和分析時(shí)，可以采用分層或配對(duì)的辦法對(duì)已知混雜因子進(jìn)行控制但在分析時(shí)采用分層手法有一定的局限性，這時(shí)需依靠基于模型的統(tǒng)計(jì)方法予以解決。偶然性Chance由于存在偶然性，單純隨機(jī)化分組也有可能造成比較的兩組間的不均衡樣本數(shù)的不均衡某些混雜因子的不均衡關(guān)于研究設(shè)計(jì)統(tǒng)計(jì)學(xué)對(duì)醫(yī)學(xué)研究的最重要貢獻(xiàn)在于研究設(shè)計(jì)而不是數(shù)據(jù)分析從統(tǒng)計(jì)學(xué)角度出發(fā)的研究設(shè)計(jì)主要涉及研究的結(jié)構(gòu)和數(shù)據(jù)收集過程的方方面面，包括度量的選擇及其頻數(shù)做適于研究目的的設(shè)計(jì)大多數(shù)研究只是想回答一些簡(jiǎn)單的問題，

4、但這并非意味它們的設(shè)計(jì)也是簡(jiǎn)單的沒有合適的研究設(shè)計(jì)就不可能有意義的研究結(jié)果再多花一點(diǎn)時(shí)間在你的設(shè)計(jì)上吧。對(duì)所要研究問題的認(rèn)識(shí)越深入、考慮的方面越多、越細(xì)，你的設(shè)計(jì)越接近于完美。研究設(shè)計(jì)的分類觀察性研究（observational）還是實(shí)驗(yàn)性研究（experimental）前瞻性研究（prospective）還是回顧性研究（retrospective）縱向研究（longitudinal）還是橫斷面研究（cross-sectional）研究設(shè)計(jì)的類型ExperimentalProspectiveLongitudinalObservationalProspectiveRetrospectiveLon

5、gitudinalCross-sectionalLongitudinalCross-sectional處理/干預(yù)隨機(jī)化設(shè)立對(duì)照設(shè)盲實(shí)驗(yàn)性研究之四大要素實(shí)驗(yàn)性研究的結(jié)構(gòu)研究因子Factors研究因子的水平Levels均衡設(shè)計(jì)組內(nèi)比較與組間比較析因設(shè)計(jì)Factorial designType of StudyAlternative NameUnit of StudyRandomised Controlled TrialsClinical TrialsCommunityInterventionTrialsPatientsCommunities/Groups/ClustersExperimental

6、StudiesClusterRandomisedTrialspopulationgroup 1group 2OutcomeOutcomenew treatmentcontrol treatment隨機(jī)對(duì)照試驗(yàn)Randomized controlled trialsAdvantagesAbility to prove causal relationships is strongest of all study typesRandomisation minimises the risk of confounding variablesThe experimental design permits

7、greater flexibility, efficiency and powerful statistical manipulationDisadvantagesOften based on specific volunteer samples and findings may not be generalisable to the wider population Not feasible/ethical for major Public Health questions (e.g. Smoking and Lung Cancer)隨機(jī)化Randomization防止選擇性偏倚、控制混雜統(tǒng)

8、計(jì)分析的基礎(chǔ)常用隨機(jī)化的方法單純隨機(jī)化Simple randomization1. 拋硬幣2. 利用隨機(jī)數(shù)字表3. 利用隨機(jī)數(shù)字發(fā)生器區(qū)組隨機(jī)化Block randomization分層隨機(jī)化Stratified randomization分層變量的選擇。年齡、性別？、中心其他隨機(jī)化方法整群隨機(jī)化Cluster randomization觀察性研究Observational study雖然實(shí)驗(yàn)性研究是推斷研究因子效果的金標(biāo)準(zhǔn)，但是可行性和倫理性上的制約使我們采用觀察性研究病例-對(duì)照研究Case-control study、定群研究Cohort study、橫斷面研究過去現(xiàn)在將來CohortCa

9、sesAndControlsCross-Sectional觀察性研究With observational studies the investigator does not control the treatment or exposure but attempts to make valid comparisons between people with or without disease or between those naturally exposed or unexposed to a factor of interest.定群研究Cohort studiesCohort stud

10、ies entail observing two or more groups differing in exposure to a potential cause of disease overtime to compare the incidence of disease in each group.It is prospective with the exposure measured before outcome, enabling the direction of events to be established that is necessary if we want to say

11、 anything about causation.Only one risk factor can be assessed for each study but multiple outcomes can be measured.Typically these studies require large numbers, especially if the incidence of the disease/outcomes is low and a long time scale. Hence they are likely to be expensive.研究對(duì)象選擇是否具有代表性失訪失訪

12、的性質(zhì)（是否與結(jié)果變量有關(guān)？）其他問題習(xí)慣改變、工作改變偏重于高危人群調(diào)查病例-對(duì)照研究Case-control studiesWith case-control study the comparison is made between people with and without the disease in order to identify differences in previous exposures in an attempt to identify the cause of the event.Retrospective簡(jiǎn)單、成本低、適用于發(fā)病率極低的病種Multiple exposures can be assessed, and they are primary method of studying new or unusual outcomes.缺點(diǎn)對(duì)照難找（醫(yī)院對(duì)照、健康對(duì)照）病例的確定回憶性偏倚Recall Bias匹配、匹配過度、一對(duì)多匹配橫斷面研究研究對(duì)象的選擇應(yīng)答率健康者偏倚因果推斷因病致貧還是因貧致??？觀察性研究的局限性The principal problem of observational studies is that although externally valid their internal v