1、使用GRADEpro評(píng)價(jià)證據(jù)質(zhì)量2011年11月GRADE系統(tǒng)簡(jiǎn)介GRADE（Grading of Recommendations Assessment, Development and Evaluation）是由2000年建立的GRADE工作組提出的一套評(píng)級(jí)系統(tǒng)。GRADE系統(tǒng)使用易于理解的方式評(píng)價(jià)證據(jù)質(zhì)量和推薦等級(jí)，目前已被世界衛(wèi)生組織（WHO）、Cochrane協(xié)作網(wǎng)等一批著名機(jī)構(gòu)所采用。2GRADE系統(tǒng)較之其他系統(tǒng)的優(yōu)勢(shì)由一個(gè)具有廣泛代表性的國(guó)際指南制定小組制定明確界定了證據(jù)質(zhì)量和推薦強(qiáng)度清楚評(píng)價(jià)了不同治療方案的重要結(jié)局對(duì)不同級(jí)別證據(jù)的升級(jí)與降級(jí)有明確、綜合的標(biāo)準(zhǔn)從證據(jù)到推薦全過程透明

2、明確承認(rèn)價(jià)值觀和意愿就推薦意見的強(qiáng)弱，分別從臨床醫(yī)生、患者、政策制定者角度做了明確實(shí)用的詮釋適用于制作系統(tǒng)評(píng)價(jià)、衛(wèi)生技術(shù)評(píng)估及指南3證據(jù)質(zhì)量（GRADEpro）為達(dá)到透明和簡(jiǎn)化的目標(biāo)， GRADE系統(tǒng)將證據(jù)質(zhì)量分為高、 中、 低、 極低 4 級(jí)。一些使用GRADE系統(tǒng)的組織甚至把低和極低歸為一級(jí)。4推薦等級(jí)使用GRADE系統(tǒng)時(shí)，指南小組用“強(qiáng)推薦”表示他們確信相關(guān)的干預(yù)措施利大于弊。用“弱推薦”表示干預(yù)措施有可能利大于弊，但他們把握不大。5GRADE中的證據(jù)質(zhì)量和推薦等級(jí)6關(guān)鍵步驟一 導(dǎo)入RevMan文件7關(guān)鍵步驟二 判斷結(jié)局的重要程度GRADE 系統(tǒng)建議采用 9 級(jí)分級(jí)判斷結(jié)局的重要程度:1

3、3：不重要結(jié)局（NOT IMPORTANT）46：重要結(jié)局（IMPORTANT）79：關(guān)鍵結(jié)局（CRITICAL）89關(guān)鍵步驟三 選擇研究設(shè)計(jì)類型按研究設(shè)計(jì)方法劃分隨機(jī)試驗(yàn)質(zhì)量等級(jí)最高，為高；觀察性研究 （包括隊(duì)列研究、病例 - 對(duì)照研究、病例 - 歷史對(duì)照研究）質(zhì)量等級(jí)低，為低；病例報(bào)告、病例分析等其他研究質(zhì)量等級(jí)非常低，為極低。10關(guān)鍵步驟四 評(píng)價(jià)證據(jù)質(zhì)量基于研究設(shè)計(jì)可因5 條降級(jí)因素和 4 條升級(jí)因素上升或下降級(jí)別。11決定證據(jù)質(zhì)量的因素可能降低證據(jù)質(zhì)量的因素研究的局限性結(jié)果不一致間接證據(jù)精確度不夠發(fā)表偏倚可能增加證據(jù)質(zhì)量的因素效應(yīng)值很大可能的混雜因素會(huì)降低療效劑量-效應(yīng)關(guān)系1213研究

4、設(shè)計(jì)與證據(jù)質(zhì)量高：為不降級(jí)的隨機(jī)試驗(yàn)和升 2 級(jí)的觀察性研究； 中：為降級(jí)的隨機(jī)試驗(yàn)和升 1 級(jí)的觀察性研究； 低：降 2 級(jí)的隨機(jī)試驗(yàn)和沒有升降級(jí)的觀察性研究； 非常低：降3級(jí)的隨機(jī)試驗(yàn)、 降1級(jí)的觀察性研究、 病例分析/病例報(bào)告。14幫助教程要點(diǎn)15研究的局限性（Risk of bias）16這些局限性包括（RCTs）隱藏分組缺失盲法缺失（特別是結(jié)局指標(biāo)為主觀性指標(biāo)且對(duì)其的評(píng)估極易受偏倚影響時(shí)）失訪過多未進(jìn)行意向性分析、觀察到療效就過早終止試驗(yàn)或未報(bào)道結(jié)果（通常是未觀察到療效的一些研究）17To rate study limitations: If you think any limita

5、tions were negligible choose no If you think there were serious limitations choose serious this will downgrade the quality of evidence for this outcome by 1 level If you think there were very serious limitations choose very serious this will downgrade the quality of evidence for this outcome by 2 le

6、vels 18結(jié)果不一致（ Inconsistency）Inconsistency may arise from differences in:populations (e.g. drugs may have larger relative effects in sicker populations) interventions (e.g. larger effects with higher drug doses) outcomes (e.g. diminishing treatment effect with time)差異可能源于人群（如藥物對(duì)重癥人群的療效可能相對(duì)顯著）、干預(yù)措施（如較

7、高藥物劑量會(huì)使療效更顯著）或結(jié)局指標(biāo)（如隨時(shí)間推移療效降低）。1920I2 One set of criteria would say that an I2 of less than 40% is low3060% may be moderate5090% may be substantial75100% is considerable21To rate inconsistency: If you think any inconsistency was negligible choose no If you think there was serious inconsistency choos

8、e serious this will downgrade the quality of evidence for this outcome by 1 level If you think there was very serious inconsistency choose very serious this will downgrade the quality of evidence for this outcome by 2 levels 22間接證據(jù)（ Indirectness)There are two types of indirectness:1. Indirect compar

9、ison2. Indirect populationinterventioncomparatoroutcome2324To rate indirectness: If you think the evidence is direct choose no If you have serious doubts about directness choose serious this will downgrade the evidence for this outcome by 1 level If you have very serious doubts about directness choo

10、se very serious this will downgrade the evidence for this outcome by 2 levels 25精確度不夠（Imprecision ）26當(dāng)研究納入的患者和觀察事件相對(duì)較少而致可信區(qū)間較寬時(shí)，指南小組將降低該研究的證據(jù)質(zhì)量。A threshold rule-of-thumb valueTotal number of events is less than 300（dichotomous）Total population size is less than 400（continuous）27To rate imprecision:

11、If you think the results were precise choose no If there was serious imprecision choose serious this will downgrade the quality of evidence for this outcome by 1 level If there was very serious imprecision choose very serious this will downgrade the quality of evidence for this outcome by 2 levels 2

12、8發(fā)表偏倚（ Publication bias ）Publication bias arises when investigators fail to report studies they have undertaken (typically those that show no effect). Methods to detect the possibility of publication bias in systematic reviews exist, although authors of the reviews and guideline panels must often gu

13、ess about the likelihood of publication bias. A prototypical situation that should elicit suspicion of publication bias occurs when published evidence is limited to a small number of trials, all of which are showing benefits of the studied intervention.若研究者未能發(fā)表研究（通常是陰性結(jié)果的研究）時(shí)，證據(jù)質(zhì)量亦會(huì)減弱。典型情況是當(dāng)公開的證據(jù)僅局限

14、于少數(shù)試驗(yàn)而這些試驗(yàn)全部由企業(yè)贊助，此時(shí)不得不質(zhì)疑存在發(fā)表偏倚。29Funnel Plot30To rate probability of the publication bias: If you think there is no evidence of publication bias choose unlikely If there is high probability of publication bias choose likely this will downgrade the quality of evidence for this outcome by 1 level If

15、there is very high probability of publication bias choose very likely this will downgrade the quality of evidence for this outcome by 2 levels 31效應(yīng)值很大（Large effect）療效越顯著，證據(jù)越有力。32To rate magnitude of the effect: If the effect was not large (RR between 0.5 and 2.0) choose no If the effect was large (R

16、R either 2.0 or 2 or 5.0 or 5 or 0.2this will upgrade the quality of evidence for this outcome by 2 levels 33可能的混雜因素會(huì)降低療效（Plausible confounding would change the effect）On occasion, all plausible confounding from observational studies or randomised trials may be working to reduce the demonstrated eff

17、ect or increase the effect if no effect was observed.For example, if only sicker patients receive an experimental intervention or exposure, yet they still fare better, it is likely that the actual intervention or exposure effect is larger than the data suggest.34To rate the effect of all plausible r

18、esidual confounding: If there is no evidence that the influence of all plausible confounding would reduce a demonstrated effect or suggest a spurious effect when results show no effect choose no If there is evidence that the influence of all plausible confounding would reduce a demonstrated effect o

19、r suggest a spurious effect when results show no effect choose yesthis will upgrade the quality of evidence for this outcome by 1 level 35劑量-效應(yīng)關(guān)系（Dose-response gradi）The presence of a dose-response gradient may increase our confidence in the findings of observational studies and thereby increase the

20、 quality of evidence.36To rate the presence of dose-response gradient: If there is no evidence of dose-response gradient choose no If there is evidence of dose-response gradient choose yesthis will upgrade the quality of evidence for this outcome by 1 level 37評(píng)級(jí)流程38關(guān)鍵步驟五 完善數(shù)據(jù)39關(guān)鍵步驟六 生成表格40Evidence Pro. Summary of Findings TableGRADE Evidence profileParticularly useful for guideline developersPresents information about the body of evidence (e.g. number of studies), the judgments