更改

跳到导航 跳到搜索
删除1,598字节 、 2021年7月15日 (四) 13:52
无编辑摘要
第3行: 第3行:  
{{Short description|Experimental method designed to reduce bias, typically accomplished by randomly allocating subjects to two or more groups, with one being a control group}}
 
{{Short description|Experimental method designed to reduce bias, typically accomplished by randomly allocating subjects to two or more groups, with one being a control group}}
   −
[[File:Flowchart of Phases of Parallel Randomized Trial - Modified from CONSORT 2010.png|thumb|upright=1.5|据修改的2010年CONSORT (综合报告试验标准)要求,流程图包括:两组平行随机试验分为登记、分配、干预、随访和数据分析四个阶段,在对照试验中,需要其中一项干预作为对照处理措施。 <ref name="Schulz-2010">{{Cite journal | author = Schulz KF, Altman DG, ((Moher D; for the CONSORT Group)) | title = CONSORT 2010 Statement: updated guidelines for reporting parallel group randomised trials | journal = [[Br Med J]] | volume = 340 | pages = c332 | year = 2010 | doi = 10.1136/bmj.c332 | pmid = 20332509 | pmc = 2844940 }}</ref>]]
+
[[File:Flowchart of Phases of Parallel Randomized Trial - Modified from CONSORT 2010.png|thumb|upright=1.5|据修改的2010年CONSORT (综合报告试验标准)要求,流程图包括:两组平行随机试验分为登记、分配、干预、随访和数据分析四个阶段,在对照试验中,需要其中一项干预作为对照处理措施。 <ref name="Schulz-2010">{{Cite journal | author = Schulz KF, Altman DG, ((Moher D; for the CONSORT Group)) | title = CONSORT 2010 Statement: updated guidelines for reporting parallel group randomised trials | journal = [[Br Med J]] | volume = 340 | pages = c332 | year = 2010 | doi = 10.1136/bmj.c332 | pmid = 20332509 | pmc = 2844940 }}</ref>]]
      −
随机对照试验(A randomized controlled trial,<ref name="Chalmers-1981">{{Cite journal |vauthors=Chalmers TC, ((Smith H Jr)), Blackburn B, Silverman B, Schroeder B, Reitman D, Ambroz A | title = A method for assessing the quality of a randomized control trial | journal = [[Controlled Clinical Trials]] | volume = 2 | issue = 1 | pages = 31–49 | year = 1981 | doi = 10.1016/0197-2456(81)90056-8 | pmid = 7261638 }}</ref> RCT)是一种科学实验(例如:临床试验)或干预研究(区别于观察性研究) ,其目的是在测试新治疗的有效性时减少某些偏倚来源。通过受试者随机分配到两个或两个以上的组,经过不同的处理,产生的效应再与一个有可控的处理效应相比较。即一组或多组(实验组)接受正在评估的干预措施,而另一组(通常称为对照组)接受替代治疗,如安慰剂或无干预措施。在试验设计的条件下对这些组进行监测,以确定实验干预的有效性,并与对照组进行疗效比较评估。<ref>{{cite web|url=https://www.nice.org.uk/glossary?letter=r|publisher=National Institute for Health and Care Excellence, London, UK|title=Randomised controlled trial|date=2019|access-date=3 June 2019}}</ref>当然这也包括一个以上的治疗组或一个以上的对照组。
+
随机对照试验 A randomized controlled trial(<ref name="Chalmers-1981">{{Cite journal |vauthors=Chalmers TC, ((Smith H Jr)), Blackburn B, Silverman B, Schroeder B, Reitman D, Ambroz A | title = A method for assessing the quality of a randomized control trial | journal = [[Controlled Clinical Trials]] | volume = 2 | issue = 1 | pages = 31–49 | year = 1981 | doi = 10.1016/0197-2456(81)90056-8 | pmid = 7261638 }}</ref> RCT)是一种科学实验(例如:临床试验)或干预研究(区别于观察性研究),其目的是在测试新治疗的有效性时减少某些偏倚来源。通过受试者随机分配到两个或两个以上的组,经过不同的处理,产生的效应再与一个有可控的处理效应相比较。即一组或多组(实验组)接受正在评估的干预措施,而另一组(通常称为对照组)接受替代治疗,如安慰剂或无干预措施。在试验设计的条件下对这些组进行监测,以确定实验干预的有效性,并与对照组进行疗效比较评估。<ref>{{cite web|url=https://www.nice.org.uk/glossary?letter=r|publisher=National Institute for Health and Care Excellence, London, UK|title=Randomised controlled trial|date=2019|access-date=3 June 2019}}</ref>当然这也包括一个以上的治疗组或一个以上的对照组。
      第18行: 第18行:       −
“ RCT”和“随机试验”这两个术语有时被用作同义词,但后一个术语没有提到对照,因此可以描述在没有对照组的情况下相互比较多个治疗组的研究。.<ref name="Ranjith-2005">{{Cite journal | author = Ranjith G | title = Interferon-α-induced depression: when a randomized trial is not a randomized controlled trial | journal = [[Psychother Psychosom]] | volume = 74 | issue = 6 | pages = 387; author reply 387–8 | year = 2005 | doi = 10.1159/000087787 | pmid = 16244516 | s2cid = 143644933 }}</ref>科学文献中常有“随机临床试验”或“随机比较试验”这类引发歧义的术语。<ref name="Peto-1976">{{Cite journal |vauthors=Peto R, Pike MC, Armitage P, Breslow NE, Cox DR, Howard SV, Mantel N, McPherson K, Peto J, Smith PG | title = Design and analysis of randomized clinical trials requiring prolonged observation of each patient. I. Introduction and design | journal = [[Br J Cancer]] | volume = 34 | issue = 6 | pages = 585–612 | year = 1976 | pmc=2025229 | pmid = 795448 | doi = 10.1038/bjc.1976.220 }}</ref><ref name="Peto-1977">{{Cite journal |vauthors=Peto R, Pike MC, Armitage P, Breslow NE, Cox DR, Howard SV, Mantel N, McPherson K, Peto J, Smith PG | title = Design and analysis of randomized clinical trials requiring prolonged observation of each patient. II. Analysis and examples | journal = [[Br J Cancer]] | volume = 35 | issue = 1 | pages = 1–39 | year = 1977 | pmc=2025310 | pmid = 831755 | doi = 10.1038/bjc.1977.1 }}</ref>并非所有的随机临床试验都是随机对照试验(其中一些试验永远不可能成为随机对照试验,因为实施控制是不切实际或不道德的)。随机对照临床试验这个术语是临床研究中使用的另一个术语;<ref name="Wollert-2004">{{Cite journal |vauthors=Wollert KC, Meyer GP, Lotz J, Ringes-Lichtenberg S, Lippolt P, Breidenbach C, Fichtner S, Korte T, Hornig B, Messinger D, Arseniev L, Hertenstein B, Ganser A, Drexler H | title = Intracoronary autologous bone-marrow cell transfer after myocardial infarction: the BOOST randomised controlled clinical trial | journal = [[The Lancet|Lancet]] | volume = 364 | issue = 9429 | pages = 141–8 | year = 2004 | doi = 10.1016/S0140-6736(04)16626-9 | pmid = 15246726 | s2cid = 24361586 }}</ref>然而,随机对照临床试验也被用于其他研究领域,包括许多社会科学。
+
“ RCT”和“随机试验”这两个术语有时被用作同义词,但后一个术语没有提到对照,因此可以描述在没有对照组的情况下相互比较多个治疗组的研究。.<ref name="Ranjith-2005">{{Cite journal | author = Ranjith G | title = Interferon-α-induced depression: when a randomized trial is not a randomized controlled trial | journal = [[Psychother Psychosom]] | volume = 74 | issue = 6 | pages = 387; author reply 387–8 | year = 2005 | doi = 10.1159/000087787 | pmid = 16244516 | s2cid = 143644933 }}</ref>科学文献中常有“随机临床试验”或“随机比较试验”这类引发歧义的术语。<ref name="Peto-1976">{{Cite journal |vauthors=Peto R, Pike MC, Armitage P, Breslow NE, Cox DR, Howard SV, Mantel N, McPherson K, Peto J, Smith PG | title = Design and analysis of randomized clinical trials requiring prolonged observation of each patient. I. Introduction and design | journal = [[Br J Cancer]] | volume = 34 | issue = 6 | pages = 585–612 | year = 1976 | pmc=2025229 | pmid = 795448 | doi = 10.1038/bjc.1976.220 }}</ref><ref name="Peto-1977">{{Cite journal |vauthors=Peto R, Pike MC, Armitage P, Breslow NE, Cox DR, Howard SV, Mantel N, McPherson K, Peto J, Smith PG | title = Design and analysis of randomized clinical trials requiring prolonged observation of each patient. II. Analysis and examples | journal = [[Br J Cancer]] | volume = 35 | issue = 1 | pages = 1–39 | year = 1977 | pmc=2025310 | pmid = 831755 | doi = 10.1038/bjc.1977.1 }}</ref>并非所有的随机临床试验都是随机对照试验(其中一些试验永远不可能成为随机对照试验,因为实施控制是不切实际或不道德的)。随机对照临床试验这个术语是临床研究中使用的另一个术语;<ref name="Wollert-2004">{{Cite journal |vauthors=Wollert KC, Meyer GP, Lotz J, Ringes-Lichtenberg S, Lippolt P, Breidenbach C, Fichtner S, Korte T, Hornig B, Messinger D, Arseniev L, Hertenstein B, Ganser A, Drexler H | title = Intracoronary autologous bone-marrow cell transfer after myocardial infarction: the BOOST randomised controlled clinical trial | journal = [[The Lancet|Lancet]] | volume = 364 | issue = 9429 | pages = 141–8 | year = 2004 | doi = 10.1016/S0140-6736(04)16626-9 | pmid = 15246726 | s2cid = 24361586 }}</ref>然而,随机对照临床试验也被用于其他研究领域,包括许多社会科学。
      第61行: 第61行:  
=== 试验注册 ===
 
=== 试验注册 ===
   −
2004年, 医学杂志编辑国际委员会(ICMJE)宣布,所有在2005年7月1日之后考虑在该委员会12种杂志上发表之前,必须对试验进行注册。<ref name="pmid15356289">{{cite journal  |vauthors=De Angelis C, Drazen JM, Frizelle FA, etal |title=Clinical trial registration: a statement from the International Committee of Medical Journal Editors |journal=The New England Journal of Medicine |volume=351 |issue=12 |pages=1250–1 |date=September 2004 |pmid=15356289 |doi=10.1056/NEJMe048225 }}</ref>尽管如此,试验登记可能仍然延迟或根本不会发生。<ref name="pmid22147862">{{cite journal|vauthors=Law MR, Kawasumi Y, Morgan SG | title=Despite law, fewer than one in eight completed studies of drugs and biologics are reported on time on ClinicalTrials.gov. | journal=Health Aff (Millwood) | year= 2011 | volume= 30 | issue= 12 | pages= 2338–45 | doi=10.1377/hlthaff.2011.0172 | pmid=22147862  | doi-access=free }}</ref><ref name="pmid19724045">{{cite journal|vauthors=Mathieu S, Boutron I, Moher D, Altman DG, Ravaud P | title=Comparison of registered and published primary outcomes in randomized controlled trials. | journal=JAMA | year= 2009 | volume= 302 | issue= 9 | pages= 977–84 | doi=10.1001/jama.2009.1242 | pmid=19724045  | doi-access=free }}</ref>医学期刊将强制性临床试验登记作为发表的先决条件进展缓慢。<ref>{{cite journal|last1=Bhaumik|first1=S|title=Editorial policies of MEDLINE indexed Indian journals on clinical trial registration.|journal=Indian Pediatr.|date=Mar 2013|volume=50|issue=3|pages=339–40|pmid=23680610|doi=10.1007/s13312-013-0092-2|s2cid=40317464}}</ref>
+
2004年, 医学杂志编辑国际委员会(ICMJE)宣布,所有在2005年7月1日之后考虑在该委员会12种杂志上发表之前,必须对试验进行注册。<ref name="pmid15356289">{{cite journal  |vauthors=De Angelis C, Drazen JM, Frizelle FA, etal |title=Clinical trial registration: a statement from the International Committee of Medical Journal Editors |journal=The New England Journal of Medicine |volume=351 |issue=12 |pages=1250–1 |date=September 2004 |pmid=15356289 |doi=10.1056/NEJMe048225 }}</ref>尽管如此,试验登记可能仍然延迟或根本不会发生。<ref name="pmid22147862">{{cite journal|vauthors=Law MR, Kawasumi Y, Morgan SG | title=Despite law, fewer than one in eight completed studies of drugs and biologics are reported on time on ClinicalTrials.gov. | journal=Health Aff (Millwood) | year= 2011 | volume= 30 | issue= 12 | pages= 2338–45 | doi=10.1377/hlthaff.2011.0172 | pmid=22147862  | doi-access=free }}</ref><ref name="pmid19724045">{{cite journal|vauthors=Mathieu S, Boutron I, Moher D, Altman DG, Ravaud P | title=Comparison of registered and published primary outcomes in randomized controlled trials. | journal=JAMA | year= 2009 | volume= 302 | issue= 9 | pages= 977–84 | doi=10.1001/jama.2009.1242 | pmid=19724045  | doi-access=free }}</ref>医学期刊将强制性临床试验登记作为发表的先决条件进展缓慢。<ref>{{cite journal|last1=Bhaumik|first1=S|title=Editorial policies of MEDLINE indexed Indian journals on clinical trial registration.|journal=Indian Pediatr.|date=Mar 2013|volume=50|issue=3|pages=339–40|pmid=23680610|doi=10.1007/s13312-013-0092-2|s2cid=40317464}}</ref>
      第68行: 第68行:  
通过研究设计对 RCT 进行分类。从最常见到最不常见,RCT 研究设计的主要类别是<ref name="Hopewell-2010">{{Cite journal |vauthors=Hopewell S, Dutton S, Yu LM, Chan AW, Altman DG | title = The quality of reports of randomised trials in 2000 and 2006: comparative study of articles indexed in PubMed | journal = BMJ | volume = 340 | pages = c723 | year = 2010 | doi = 10.1136/bmj.c723  | pmid = 20332510 | pmc = 2844941 }}</ref>:
 
通过研究设计对 RCT 进行分类。从最常见到最不常见,RCT 研究设计的主要类别是<ref name="Hopewell-2010">{{Cite journal |vauthors=Hopewell S, Dutton S, Yu LM, Chan AW, Altman DG | title = The quality of reports of randomised trials in 2000 and 2006: comparative study of articles indexed in PubMed | journal = BMJ | volume = 340 | pages = c723 | year = 2010 | doi = 10.1136/bmj.c723  | pmid = 20332510 | pmc = 2844941 }}</ref>:
   −
* 平行试验:每个参与者被随机分配到一个组,组中的所有参与者都接受(或不接受)干预。<ref>{{cite journal |last1=Kaiser |first1=Joerg |last2=Niesen |first2=Willem |last3=Probst |first3=Pascal |last4=Bruckner |first4=Thomas |last5=Doerr-Harim |first5=Colette |last6=Strobel |first6=Oliver |last7=Knebel |first7=Phillip |last8=Diener |first8=Markus K. |last9=Mihaljevic |first9=André L. |last10=Büchler |first10=Markus W. |last11=Hackert |first11=Thilo |title=Abdominal drainage versus no drainage after distal pancreatectomy: study protocol for a randomized controlled trial |journal=Trials |date=7 June 2019 |volume=20 |issue=1 |page=332 |doi=10.1186/s13063-019-3442-0|pmid=31174583 |pmc=6555976 |doi-access=free }}</ref><ref>{{cite journal |last1=Farag |first1=Sara M. |last2=Mohammed |first2=Manal O. |last3=EL-Sobky |first3=Tamer A. |last4=ElKadery |first4=Nadia A. |last5=ElZohiery |first5=Abeer K. |title=Botulinum Toxin A Injection in Treatment of Upper Limb Spasticity in Children with Cerebral Palsy: A Systematic Review of Randomized Controlled Trials |journal=JBJS Reviews |date=March 2020 |volume=8 |issue=3 |pages=e0119 |doi=10.2106/JBJS.RVW.19.00119 |pmid=32224633|pmc=7161716 |doi-access=free }}</ref>
+
* 平行试验:每个参与者被随机分配到一个组,组中的所有参与者都接受(或不接受)干预。<ref>{{cite journal |last1=Kaiser |first1=Joerg |last2=Niesen |first2=Willem |last3=Probst |first3=Pascal |last4=Bruckner |first4=Thomas |last5=Doerr-Harim |first5=Colette |last6=Strobel |first6=Oliver |last7=Knebel |first7=Phillip |last8=Diener |first8=Markus K. |last9=Mihaljevic |first9=André L. |last10=Büchler |first10=Markus W. |last11=Hackert |first11=Thilo |title=Abdominal drainage versus no drainage after distal pancreatectomy: study protocol for a randomized controlled trial |journal=Trials |date=7 June 2019 |volume=20 |issue=1 |page=332 |doi=10.1186/s13063-019-3442-0|pmid=31174583 |pmc=6555976 |doi-access=free }}</ref><ref>{{cite journal |last1=Farag |first1=Sara M. |last2=Mohammed |first2=Manal O. |last3=EL-Sobky |first3=Tamer A. |last4=ElKadery |first4=Nadia A. |last5=ElZohiery |first5=Abeer K. |title=Botulinum Toxin A Injection in Treatment of Upper Limb Spasticity in Children with Cerebral Palsy: A Systematic Review of Randomized Controlled Trials |journal=JBJS Reviews |date=March 2020 |volume=8 |issue=3 |pages=e0119 |doi=10.2106/JBJS.RVW.19.00119 |pmid=32224633|pmc=7161716 |doi-access=free }}</ref>
      −
* 交叉试验:随着时间的推移,每个参与者都会接受(或不接受)随机序列的干预。<ref>{{cite book  | last = Jones | first = Byron |author2=Kenward, Michael G.  | title = Design and Analysis of Cross-Over Trials | edition=Second| publisher = London: Chapman and Hall | year = 2003 }}</ref><ref>{{cite book  | last = Vonesh | first = Edward F. |author2=Chinchilli, Vernon G.  | chapter=Crossover Experiments| pages=111–202|title = Linear and Nonlinear Models for the Analysis of Repeated Measurements | publisher = London: Chapman and Hall | year = 1997 }}</ref>
+
* 交叉试验:随着时间的推移,每个参与者都会接受(或不接受)随机序列的干预。<ref>{{cite book  | last = Jones | first = Byron |author2=Kenward, Michael G.  | title = Design and Analysis of Cross-Over Trials | edition=Second| publisher = London: Chapman and Hall | year = 2003 }}</ref><ref>{{cite book  | last = Vonesh | first = Edward F. |author2=Chinchilli, Vernon G.  | chapter=Crossover Experiments| pages=111–202|title = Linear and Nonlinear Models for the Analysis of Repeated Measurements | publisher = London: Chapman and Hall | year = 1997 }}</ref>
      −
* 聚类试验:预先存在的参与者组(例如,村庄、学校)被随机选择以接受(或不接受)干预。<ref>{{cite journal |last1=Gall |first1=Stefanie |last2=Adams |first2=Larissa |last3=Joubert |first3=Nandi |last4=Ludyga |first4=Sebastian |last5=Müller |first5=Ivan |last6=Nqweniso |first6=Siphesihle |last7=Pühse |first7=Uwe |last8=du Randt |first8=Rosa |last9=Seelig |first9=Harald |last10=Smith |first10=Danielle |last11=Steinmann |first11=Peter |last12=Utzinger |first12=Jürg |last13=Walter |first13=Cheryl |last14=Gerber |first14=Markus |last15=van Wouwe |first15=Jacobus P. |title=Effect of a 20-week physical activity intervention on selective attention and academic performance in children living in disadvantaged neighborhoods: A cluster randomized control trial |journal=PLOS ONE |date=8 November 2018 |volume=13 |issue=11 |pages=e0206908 |doi=10.1371/journal.pone.0206908 |pmid=30408073|pmc=6224098 |bibcode=2018PLoSO..1306908G |doi-access=free }}</ref><ref>{{cite journal |last1=Gladstone |first1=Melissa J. |last2=Chandna |first2=Jaya |last3=Kandawasvika |first3=Gwendoline |last4=Ntozini |first4=Robert |last5=Majo |first5=Florence D. |last6=Tavengwa |first6=Naume V. |last7=Mbuya |first7=Mduduzi N. N. |last8=Mangwadu |first8=Goldberg T. |last9=Chigumira |first9=Ancikaria |last10=Chasokela |first10=Cynthia M. |last11=Moulton |first11=Lawrence H. |last12=Stoltzfus |first12=Rebecca J. |last13=Humphrey |first13=Jean H. |last14=Prendergast |first14=Andrew J. |last15=Tumwine |first15=James K. |title=Independent and combined effects of improved water, sanitation, and hygiene (WASH) and improved complementary feeding on early neurodevelopment among children born to HIV-negative mothers in rural Zimbabwe: Substudy of a cluster-randomized trial |journal=PLOS Medicine |date=21 March 2019 |volume=16 |issue=3 |pages=e1002766 |doi=10.1371/journal.pmed.1002766|pmid=30897095 |pmc=6428259 |doi-access=free }}</ref>
+
* 聚类试验:预先存在的参与者组(例如,村庄、学校)被随机选择以接受(或不接受)干预。<ref>{{cite journal |last1=Gall |first1=Stefanie |last2=Adams |first2=Larissa |last3=Joubert |first3=Nandi |last4=Ludyga |first4=Sebastian |last5=Müller |first5=Ivan |last6=Nqweniso |first6=Siphesihle |last7=Pühse |first7=Uwe |last8=du Randt |first8=Rosa |last9=Seelig |first9=Harald |last10=Smith |first10=Danielle |last11=Steinmann |first11=Peter |last12=Utzinger |first12=Jürg |last13=Walter |first13=Cheryl |last14=Gerber |first14=Markus |last15=van Wouwe |first15=Jacobus P. |title=Effect of a 20-week physical activity intervention on selective attention and academic performance in children living in disadvantaged neighborhoods: A cluster randomized control trial |journal=PLOS ONE |date=8 November 2018 |volume=13 |issue=11 |pages=e0206908 |doi=10.1371/journal.pone.0206908 |pmid=30408073|pmc=6224098 |bibcode=2018PLoSO..1306908G |doi-access=free }}</ref><ref>{{cite journal |last1=Gladstone |first1=Melissa J. |last2=Chandna |first2=Jaya |last3=Kandawasvika |first3=Gwendoline |last4=Ntozini |first4=Robert |last5=Majo |first5=Florence D. |last6=Tavengwa |first6=Naume V. |last7=Mbuya |first7=Mduduzi N. N. |last8=Mangwadu |first8=Goldberg T. |last9=Chigumira |first9=Ancikaria |last10=Chasokela |first10=Cynthia M. |last11=Moulton |first11=Lawrence H. |last12=Stoltzfus |first12=Rebecca J. |last13=Humphrey |first13=Jean H. |last14=Prendergast |first14=Andrew J. |last15=Tumwine |first15=James K. |title=Independent and combined effects of improved water, sanitation, and hygiene (WASH) and improved complementary feeding on early neurodevelopment among children born to HIV-negative mothers in rural Zimbabwe: Substudy of a cluster-randomized trial |journal=PLOS Medicine |date=21 March 2019 |volume=16 |issue=3 |pages=e1002766 |doi=10.1371/journal.pmed.1002766|pmid=30897095 |pmc=6428259 |doi-access=free }}</ref>
      −
* 因子试验:每个参与者被随机分配到一个接受干预或非干预特定组合的组(例如,第1组接受维生素X和维生素Y,第2组接受维生素X和安慰剂Y,第3组接受安慰剂X和维生素Y,第4组接受安慰剂X和安慰剂Y)。
+
* 因子试验:每个参与者被随机分配到一个接受干预或非干预特定组合的组(例如,第1组接受维生素X和维生素Y,第2组接受维生素X和安慰剂Y,第3组接受安慰剂X和维生素Y,第4组接受安慰剂X和安慰剂Y)。
      第116行: 第116行:       −
* '''最大化统计能力''',尤其是在亚组分析中。一般来说,相等的组规模将最大化统计能力,然而,不相等的组规模对于某些分析来说可能更强大(例如,使用Dunnett程序对安慰剂与几个剂量进行多次比较<ref>{{cite web|last=Rosenberger|first=James|title=STAT 503 - Design of Experiments|url=https://onlinecourses.science.psu.edu/stat503/node/16|publisher=Pennsylvania State University|access-date=24 September 2012}}</ref>),并且有时由于非分析性原因而被采用(例如,如果有更高的机会获得试验治疗,患者可能更有动力登记,或者监管机构可能要求最少数量的患者接受治疗)。<ref name="Avins-1998">{{cite journal|last=Avins|first=A L|title="Can unequal be more fair? Ethics, subject allocation, and randomized clinical trials".|journal=J Med Ethics|year=1998|volume=24|pages=401–408|doi=10.1136/jme.24.6.401|issue=6|pmc=479141|pmid=9873981}}</ref>
+
* '''最大化统计能力''',尤其是在亚组分析中。一般来说,相等的组规模将最大化统计能力,然而,不相等的组规模对于某些分析来说可能更强大(例如,使用Dunnett程序对安慰剂与几个剂量进行多次比较<ref>{{cite web|last=Rosenberger|first=James|title=STAT 503 - Design of Experiments|url=https://onlinecourses.science.psu.edu/stat503/node/16|publisher=Pennsylvania State University|access-date=24 September 2012}}</ref>),并且有时由于非分析性原因而被采用(例如,如果有更高的机会获得试验治疗,患者可能更有动力登记,或者监管机构可能要求最少数量的患者接受治疗)。<ref name="Avins-1998">{{cite journal|last=Avins|first=A L|title="Can unequal be more fair? Ethics, subject allocation, and randomized clinical trials".|journal=J Med Ethics|year=1998|volume=24|pages=401–408|doi=10.1136/jme.24.6.401|issue=6|pmc=479141|pmid=9873981}}</ref>
   −
* '''最小化选择偏差'''。如果调查人员可以有意识或无意识地在治疗之间优先招募患者,就可能发生这种情况。一个好的随机化过程是不可预测的,因此研究人员不能根据先前的治疗分配来猜测下一个受试者的分组。当已知以前的治疗方案时(如在非盲法研究中)或可以猜到(如果一种药物有明显的副作用),选择偏倚的风险最高。
+
* '''最小化选择偏差'''。如果调查人员可以有意识或无意识地在治疗之间优先招募患者,就可能发生这种情况。一个好的随机化过程是不可预测的,因此研究人员不能根据先前的治疗分配来猜测下一个受试者的分组。当已知以前的治疗方案时(如在非盲法研究中)或可以猜到(如果一种药物有明显的副作用),选择偏倚的风险最高。
   −
* '''最小化分配偏差(或混淆)'''。当影响结果的协变量在治疗组之间分布不均,并且治疗效果与协变量的效果混淆时(即“偶然偏差”),可能会出现这种情况。<ref name="SchulzGrimes2002"/><ref name="Buyse-1989">{{Cite journal | author = Buyse ME | title = Analysis of clinical trial outcomes: some comments on subgroup analyses | journal = [[Controlled Clinical Trials]] | volume = 10 | issue = 4 Suppl | pages = 187S–194S | year = 1989 | pmid = 2605967 |  doi = 10.1016/0197-2456(89)90057-3 }}</ref>如果随机化程序导致与各组结果相关的协变量失衡,如果不对协变量进行调整,效果估计可能会有偏差(这可能无法测量,因此无法调整)。
+
* '''最小化分配偏差(或混淆)'''。当影响结果的协变量在治疗组之间分布不均,并且治疗效果与协变量的效果混淆时(即“偶然偏差”),可能会出现这种情况。<ref name="SchulzGrimes2002"/><ref name="Buyse-1989">{{Cite journal | author = Buyse ME | title = Analysis of clinical trial outcomes: some comments on subgroup analyses | journal = [[Controlled Clinical Trials]] | volume = 10 | issue = 4 Suppl | pages = 187S–194S | year = 1989 | pmid = 2605967 |  doi = 10.1016/0197-2456(89)90057-3 }}</ref>如果随机化程序导致与各组结果相关的协变量失衡,如果不对协变量进行调整,效果估计可能会有偏差(这可能无法测量,因此无法调整)。
      第136行: 第136行:       −
* '''置换区组随机化'''或'''区组随机化''':规定了“区组大小”和“分配比例”(一组受试者相对于另一组受试者的数量),受试者在每个区组内随机分配。<ref name="Schulz-2002"/>例如,块大小为6,分配比例为2:1,将导致4个受试者随机分配到一个组,2个分配到另一个组。这种类型的随机化可以与“分层随机化”相结合,例如通过多中心试验中的中心,以“确保每个组中参与者特征的良好平衡。”<ref name="Moher-2010"/>置换块随机化的一个特殊情况是随机分配,其中整个样本被视为一个块。<ref name="Schulz-2002"/>置换块随机化的主要缺点是,即使块大小很大且随机变化,该过程也会导致选择偏差。<ref name="Lachin-1988a"/>另一个缺点是,对置换区组随机对照试验数据的“适当”分析需要按区组分层。<ref name="Lachin-1988b"/>
+
* '''置换区组随机化'''或'''区组随机化''':规定了“区组大小”和“分配比例”(一组受试者相对于另一组受试者的数量),受试者在每个区组内随机分配。<ref name="Schulz-2002"/>例如,块大小为6,分配比例为2:1,将导致4个受试者随机分配到一个组,2个分配到另一个组。这种类型的随机化可以与“分层随机化”相结合,例如通过多中心试验中的中心,以“确保每个组中参与者特征的良好平衡。”<ref name="Moher-2010"/>置换块随机化的一个特殊情况是随机分配,其中整个样本被视为一个块。<ref name="Schulz-2002"/>置换块随机化的主要缺点是,即使块大小很大且随机变化,该过程也会导致选择偏差。<ref name="Lachin-1988a"/>另一个缺点是,对置换区组随机对照试验数据的“适当”分析需要按区组分层。<ref name="Lachin-1988b"/>
    
* '''自适应有偏-硬币随机化方法'''(其中瓮随机化是最广为人知的类型):在这些相对不常见的方法中,如果一个组的代表人数过多,被分配到该组的概率会降低,如果该组的代表人数不足,被分配到该组的概率会增加。这些方法被认为比置换块随机化受选择偏差的影响更小。<ref name="Lachin-1988b"/>
 
* '''自适应有偏-硬币随机化方法'''(其中瓮随机化是最广为人知的类型):在这些相对不常见的方法中,如果一个组的代表人数过多,被分配到该组的概率会降低,如果该组的代表人数不足,被分配到该组的概率会增加。这些方法被认为比置换块随机化受选择偏差的影响更小。<ref name="Lachin-1988b"/>
第152行: 第152行:  
=== 分配隐藏 ===
 
=== 分配隐藏 ===
   −
“分配隐藏”(定义为“保护随机化过程的程序,以便在病人进入研究之前不知道要分配的治疗”)在随机对照试验中很重要。.<ref name="Forder-2005">{{Cite journal |vauthors=Forder PM, Gebski VJ, Keech AC | title = Allocation concealment and blinding: when ignorance is bliss | journal = Med J Aust | volume = 182 | issue = 2 | pages = 87–9 | year = 2005 | url = http://www.mja.com.au/public/issues/182_02_170105/for10877_fm.html | pmid = 15651970 | doi = 10.5694/j.1326-5377.2005.tb06584.x | s2cid = 202149 }}</ref>在实践中,临床研究人员在随机对照试验中常常发现难以保持公正性。关于调查人员将密封的信封举到灯光下或者搜查办公室来决定群组分配,以便指定下一个病人的分配的故事比比皆是。<ref name="Schulz-2002">{{Cite journal | title = Allocation concealment in randomised trials: defending against deciphering | journal = Lancet | volume = 359 | issue = 9306 | pages = 614–8 | year = 2002 | doi = 10.1016/S0140-6736(02)07750-4 | url =https://www.who.int/entity/rhl/LANCET_614-618.pdf | pmid = 11867132 |vauthors=Schulz KF, Grimes DA | s2cid = 12902486 }}</ref>这种做法引入了选择偏差和混杂因素(这两者都应该通过随机化来减少) ,可能会扭曲研究结果。<ref name="Schulz-2002"/>一旦研究开始并在研究结束后,充分的分配隐藏应该会阻止患者和研究者发现治疗分配。与治疗相关的副作用或不良事件可能足够具体,足以向研究者或患者揭示分配情况,从而引入偏差或影响研究者收集的或受试者要求的任何主观参数。
+
“分配隐藏”(定义为“保护随机化过程的程序,以便在病人进入研究之前不知道要分配的治疗”)在随机对照试验中很重要。.<ref name="Forder-2005">{{Cite journal |vauthors=Forder PM, Gebski VJ, Keech AC | title = Allocation concealment and blinding: when ignorance is bliss | journal = Med J Aust | volume = 182 | issue = 2 | pages = 87–9 | year = 2005 | url = http://www.mja.com.au/public/issues/182_02_170105/for10877_fm.html | pmid = 15651970 | doi = 10.5694/j.1326-5377.2005.tb06584.x | s2cid = 202149 }}</ref>在实践中,临床研究人员在随机对照试验中常常发现难以保持公正性。关于调查人员将密封的信封举到灯光下或者搜查办公室来决定群组分配,以便指定下一个病人的分配的故事比比皆是。<ref name="Schulz-2002">{{Cite journal | title = Allocation concealment in randomised trials: defending against deciphering | journal = Lancet | volume = 359 | issue = 9306 | pages = 614–8 | year = 2002 | doi = 10.1016/S0140-6736(02)07750-4 | url =https://www.who.int/entity/rhl/LANCET_614-618.pdf | pmid = 11867132 |vauthors=Schulz KF, Grimes DA | s2cid = 12902486 }}</ref>这种做法引入了选择偏差和混杂因素(这两者都应该通过随机化来减少) ,可能会扭曲研究结果。<ref name="Schulz-2002"/>一旦研究开始并在研究结束后,充分的分配隐藏应该会阻止患者和研究者发现治疗分配。与治疗相关的副作用或不良事件可能足够具体,足以向研究者或患者揭示分配情况,从而引入偏差或影响研究者收集的或受试者要求的任何主观参数。
      −
一些确保分配隐藏的标准方法包括顺序编号、不透明、密封信封(SNOSE);顺序编号的容器;药学控制的随机化;和中心随机化。分配隐藏方法被建议纳入RCT议定书,并在RCT结果的出版物中详细报告分配隐藏方法;然而,2005年的一项研究发现,大多数随机对照试验在其方案、出版物或两者中都有不清楚的分配隐藏。另一方面,2008年的一项对146项元分析的研究得出结论,分配隐瞒不充分或不明确的随机对照试验的结果往往只有在随机对照试验的结果是主观的而不是客观的情况下才会偏向于有益的结果。
+
一些确保分配隐藏的标准方法包括顺序编号、不透明、密封信封(SNOSE);顺序编号的容器;药学控制的随机化;和中心随机化。分配隐藏方法被建议纳入RCT议定书,并在RCT结果的出版物中详细报告分配隐藏方法;然而,2005年的一项研究发现,大多数随机对照试验在其方案、出版物或两者中都有不清楚的分配隐藏。另一方面,2008年的一项对146项元分析的研究得出结论,分配隐瞒不充分或不明确的随机对照试验的结果往往只有在随机对照试验的结果是主观的而不是客观的情况下才会偏向于有益的结果。
       
=== 样本量 ===
 
=== 样本量 ===
   −
分配给控制组和治疗组的治疗单位(受试者或受试者组)的样本量影响 RCT 的可靠性。如果治疗的效果很小,任何一组的治疗单位的样本量都可能不足以在各自的统计检验中拒绝零假设。拒绝无效假设的失败意味着在给定的试验中,治疗对被治疗者没有统计学上的显著影响。但是随着样本量的增加,同样的RCT可能能够证明治疗的显著效果,即使这种效果很小。
+
分配给控制组和治疗组的治疗单位(受试者或受试者组)的样本量影响 RCT 的可靠性。如果治疗的效果很小,任何一组的治疗单位的样本量都可能不足以在各自的统计检验中拒绝零假设。拒绝无效假设的失败意味着在给定的试验中,治疗对被治疗者没有统计学上的显著影响。但是随着样本量的增加,同样的RCT可能能够证明治疗的显著效果,即使这种效果很小。
       
== 盲法 ==
 
== 盲法 ==
一项RCT的盲法是指阻止研究参与者、照顾者或结果评估者知道哪些干预措施,这一程序也称“蒙面”。<ref name="Wood-2008">{{Cite journal |vauthors=Wood L, Egger M, Gluud LL, Schulz KF, Jüni P, Altman DG, Gluud C, Martin RM, Wood AJ, Sterne JA | title = Empirical evidence of bias in treatment effect estimates in controlled trials with different interventions and outcomes: meta-epidemiological study | journal = BMJ | volume = 336 | issue = 7644 | pages = 601–5 | year = 2008 | doi = 10.1136/bmj.39465.451748.AD | pmid = 18316340 | pmc = 2267990 }}</ref>与分配隐藏不同,在RCT,致盲有时是不合适的或不可能的;例如,如果RCT涉及需要患者积极参与的治疗(例如物理治疗),参与者不能对干预视而不见。
+
一项RCT的盲法是指阻止研究参与者、照顾者或结果评估者知道哪些干预措施,这一程序也称“蒙面”。<ref name="Wood-2008">{{Cite journal |vauthors=Wood L, Egger M, Gluud LL, Schulz KF, Jüni P, Altman DG, Gluud C, Martin RM, Wood AJ, Sterne JA | title = Empirical evidence of bias in treatment effect estimates in controlled trials with different interventions and outcomes: meta-epidemiological study | journal = BMJ | volume = 336 | issue = 7644 | pages = 601–5 | year = 2008 | doi = 10.1136/bmj.39465.451748.AD | pmid = 18316340 | pmc = 2267990 }}</ref>与分配隐藏不同,在RCT,致盲有时是不合适的或不可能的;例如,如果RCT涉及需要患者积极参与的治疗(例如物理治疗),参与者不能对干预视而不见。
      −
传统上,盲法随机对照试验分为“单盲”、“双盲”或“三盲”;然而,2001年和2006年的两项研究表明,这些术语对不同的人有不同的含义。<ref name="Devereaux-2001">{{Cite journal |vauthors=Devereaux PJ, Manns BJ, Ghali WA, Quan H, Lacchetti C, Montori VM, Bhandari M, Guyatt GH | title = Physician interpretations and textbook definitions of blinding terminology in randomized controlled trials | journal = [[J Am Med Assoc]] | volume = 285 | issue = 15 | pages = 2000–3 | year = 2001 | doi =  10.1001/jama.285.15.2000| pmid = 11308438 | doi-access = free }}</ref><ref name="Haahr-2006">{{Cite journal |vauthors=Haahr MT, Hróbjartsson A | title = Who is blinded in randomized clinical trials? A study of 200 trials and a survey of authors | journal = Clin Trials | volume = 3 | issue = 4 | pages = 360–5 | year = 2006 | doi = 10.1177/1740774506069153 | pmid = 17060210 | s2cid = 23818514 }}</ref>2010年CONSORT 声明明确指出,作者和编辑不应使用”单盲”、”双盲”和”三盲”等术语;相反,关于盲法 RCT 的报告应讨论”如果完成,干预分配后谁被“蒙面”了(例如,参与者、护理提供者、评估结果的人员)以及其原因”。<ref name="Moher-2010"/>
+
传统上,盲法随机对照试验分为“单盲”、“双盲”或“三盲”;然而,2001年和2006年的两项研究表明,这些术语对不同的人有不同的含义。<ref name="Devereaux-2001">{{Cite journal |vauthors=Devereaux PJ, Manns BJ, Ghali WA, Quan H, Lacchetti C, Montori VM, Bhandari M, Guyatt GH | title = Physician interpretations and textbook definitions of blinding terminology in randomized controlled trials | journal = [[J Am Med Assoc]] | volume = 285 | issue = 15 | pages = 2000–3 | year = 2001 | doi =  10.1001/jama.285.15.2000| pmid = 11308438 | doi-access = free }}</ref><ref name="Haahr-2006">{{Cite journal |vauthors=Haahr MT, Hróbjartsson A | title = Who is blinded in randomized clinical trials? A study of 200 trials and a survey of authors | journal = Clin Trials | volume = 3 | issue = 4 | pages = 360–5 | year = 2006 | doi = 10.1177/1740774506069153 | pmid = 17060210 | s2cid = 23818514 }}</ref>2010年CONSORT 声明明确指出,作者和编辑不应使用”单盲”、”双盲”和”三盲”等术语;相反,关于盲法 RCT 的报告应讨论”如果完成,干预分配后谁被“蒙面”了(例如,参与者、护理提供者、评估结果的人员)以及其原因”。<ref name="Moher-2010"/>
      −
没有盲法的随机对照试验被称为“未盲法”,<ref name="Marson-2007">{{Cite journal  |vauthors=Marson AG, Al-Kharusi AM, Alwaidh M, Appleton R, Baker GA, Chadwick DW, etal | title = The SANAD study of effectiveness of valproate, lamotrigine, or topiramate for generalised and unclassifiable epilepsy: an unblinded randomised controlled trial | journal = [[The Lancet|Lancet]] | volume = 369 | issue = 9566 | pages = 1016–26 | year = 2007 | doi = 10.1016/S0140-6736(07)60461-9 | pmid = 17382828 | pmc = 2039891 }}</ref>也称“开放”,<ref name="Chan-1995">{{Cite journal |vauthors=Chan R, Hemeryck L, O'Regan M, Clancy L, Feely J | title = Oral versus intravenous antibiotics for community acquired lower respiratory tract infection in a general hospital: open, randomised controlled trial | journal = [[BMJ]] | volume = 310 | issue = 6991 | pages = 1360–2 | year = 1995 | pmid = 7787537 | pmc = 2549744 | doi=10.1136/bmj.310.6991.1360}}</ref>或者(如果干预是一种药物)“开放标签”。<ref name="Fukase-2008">{{Cite journal | author = Fukase K, Kato M, Kikuchi S, Inoue K, Uemura N, Okamoto S, Terao S, Amagai K, Hayashi S, Asaka M; Japan Gast Study Group | title = Effect of eradication of Helicobacter pylori on incidence of metachronous gastric carcinoma after endoscopic resection of early gastric cancer: an open-label, randomised controlled trial | journal = Lancet | volume = 372 | issue = 9636 | pages = 392–7 | year = 2008 | doi = 10.1016/S0140-6736(08)61159-9 | pmid = 18675689 | hdl = 2115/34681 | s2cid = 13741892 | url = https://eprints.lib.hokudai.ac.jp/dspace/bitstream/2115/34681/1/asaka.pdf | hdl-access = free }}</ref>2008年的一项研究得出结论,只有当随机对照试验的结果是主观的而不是客观的时候,非盲法随机对照试验的结果往往偏向于有益的结果;<ref name="Wood-2008"/>例如,在RCT多发性硬化症的治疗中,未盲的神经学家认为治疗是有益的。<ref name="Noseworthy-1994">{{Cite journal |vauthors=Noseworthy JH, Ebers GC, Vandervoort MK, Farquhar RE, Yetisir E, Roberts R | author-link1=John H. Noseworthy|title = The impact of blinding on the results of a randomized, placebo-controlled multiple sclerosis clinical trial | journal = Neurology | volume = 44 | issue = 1 | pages = 16–20 | year = 1994 | url = http://www.neurology.org/cgi/content/abstract/44/1/16 | pmid = 8290055 | doi=10.1212/wnl.44.1.16| s2cid=2663997}}</ref>在实用的随机对照试验中,尽管参与者和提供者往往是非盲的,但是”仍然需要并且往往可能使评估者“蒙面”,以获得评估结果的客观数据来源”。<ref name="Zwarenstein-2008"/>
+
没有盲法的随机对照试验被称为“未盲法”,<ref name="Marson-2007">{{Cite journal  |vauthors=Marson AG, Al-Kharusi AM, Alwaidh M, Appleton R, Baker GA, Chadwick DW, etal | title = The SANAD study of effectiveness of valproate, lamotrigine, or topiramate for generalised and unclassifiable epilepsy: an unblinded randomised controlled trial | journal = [[The Lancet|Lancet]] | volume = 369 | issue = 9566 | pages = 1016–26 | year = 2007 | doi = 10.1016/S0140-6736(07)60461-9 | pmid = 17382828 | pmc = 2039891 }}</ref>也称“开放”,<ref name="Chan-1995">{{Cite journal |vauthors=Chan R, Hemeryck L, O'Regan M, Clancy L, Feely J | title = Oral versus intravenous antibiotics for community acquired lower respiratory tract infection in a general hospital: open, randomised controlled trial | journal = [[BMJ]] | volume = 310 | issue = 6991 | pages = 1360–2 | year = 1995 | pmid = 7787537 | pmc = 2549744 | doi=10.1136/bmj.310.6991.1360}}</ref>或者(如果干预是一种药物)“开放标签”。<ref name="Fukase-2008">{{Cite journal | author = Fukase K, Kato M, Kikuchi S, Inoue K, Uemura N, Okamoto S, Terao S, Amagai K, Hayashi S, Asaka M; Japan Gast Study Group | title = Effect of eradication of Helicobacter pylori on incidence of metachronous gastric carcinoma after endoscopic resection of early gastric cancer: an open-label, randomised controlled trial | journal = Lancet | volume = 372 | issue = 9636 | pages = 392–7 | year = 2008 | doi = 10.1016/S0140-6736(08)61159-9 | pmid = 18675689 | hdl = 2115/34681 | s2cid = 13741892 | url = https://eprints.lib.hokudai.ac.jp/dspace/bitstream/2115/34681/1/asaka.pdf | hdl-access = free }}</ref>2008年的一项研究得出结论,只有当随机对照试验的结果是主观的而不是客观的时候,非盲法随机对照试验的结果往往偏向于有益的结果;<ref name="Wood-2008"/>例如,在RCT多发性硬化症的治疗中,未盲的神经学家认为治疗是有益的。<ref name="Noseworthy-1994">{{Cite journal |vauthors=Noseworthy JH, Ebers GC, Vandervoort MK, Farquhar RE, Yetisir E, Roberts R | author-link1=John H. Noseworthy|title = The impact of blinding on the results of a randomized, placebo-controlled multiple sclerosis clinical trial | journal = Neurology | volume = 44 | issue = 1 | pages = 16–20 | year = 1994 | url = http://www.neurology.org/cgi/content/abstract/44/1/16 | pmid = 8290055 | doi=10.1212/wnl.44.1.16| s2cid=2663997}}</ref>在实用的随机对照试验中,尽管参与者和提供者往往是非盲的,但是”仍然需要并且往往可能使评估者“蒙面”,以获得评估结果的客观数据来源”。<ref name="Zwarenstein-2008"/>
      第177行: 第177行:  
随机对照试验中使用的统计方法类型取决于数据的特征,包括:
 
随机对照试验中使用的统计方法类型取决于数据的特征,包括:
   −
* 对于二元结果数据,可以使用逻辑回归(例如,预测接受聚乙二醇干扰素α-2a治疗丙型肝炎后的持续病毒学应答<ref name="Manns-2001">{{Cite journal |vauthors=Manns MP, McHutchison JG, Gordon SC, Rustgi VK, Shiffman M, Reindollar R, Goodman ZD, Koury K, Ling M, Albrecht JK | title = Peginterferon alfa-2b plus ribavirin compared with interferon alfa-2b plus ribavirin for initial treatment of chronic hepatitis C: a randomised trial | journal = [[The Lancet|Lancet]] | volume = 358 | issue = 9286 | pages = 958–65 | year = 2001 | doi = 10.1016/S0140-6736(01)06102-5 | pmid = 11583749 | s2cid = 14583372 }}</ref>)和其他方法。
+
* 对于二元结果数据,可以使用逻辑回归(例如,预测接受聚乙二醇干扰素α-2a治疗丙型肝炎后的持续病毒学应答<ref name="Manns-2001">{{Cite journal |vauthors=Manns MP, McHutchison JG, Gordon SC, Rustgi VK, Shiffman M, Reindollar R, Goodman ZD, Koury K, Ling M, Albrecht JK | title = Peginterferon alfa-2b plus ribavirin compared with interferon alfa-2b plus ribavirin for initial treatment of chronic hepatitis C: a randomised trial | journal = [[The Lancet|Lancet]] | volume = 358 | issue = 9286 | pages = 958–65 | year = 2001 | doi = 10.1016/S0140-6736(01)06102-5 | pmid = 11583749 | s2cid = 14583372 }}</ref>)和其他方法。
   −
* 对于连续的结果数据,协方差分析(例如,急性冠状动脉综合征后接受阿托伐他汀后血脂水平的变化<ref name="Schwartz-2001">{{Cite journal | author = Schwartz GG, Olsson AG, Ezekowitz MD, Ganz P, Oliver MF, Waters D, Zeiher A, Chaitman BR, Leslie S, Stern T; Myocardial Ischemia Reduction with Aggressive Cholesterol Lowering (MIRACL) Study Investigators | title = Effects of atorvastatin on early recurrent ischemic events in acute coronary syndromes: the MIRACL study: a randomized controlled trial | journal = [[J Am Med Assoc]] | volume = 285 | issue = 13 | pages = 1711–8 | year = 2001 | pmid = 11277825 | doi = 10.1001/jama.285.13.1711 | doi-access = free }}</ref>)可以用于检测预测变量效果。
+
* 对于连续的结果数据,协方差分析(例如,急性冠状动脉综合征后接受阿托伐他汀后血脂水平的变化<ref name="Schwartz-2001">{{Cite journal | author = Schwartz GG, Olsson AG, Ezekowitz MD, Ganz P, Oliver MF, Waters D, Zeiher A, Chaitman BR, Leslie S, Stern T; Myocardial Ischemia Reduction with Aggressive Cholesterol Lowering (MIRACL) Study Investigators | title = Effects of atorvastatin on early recurrent ischemic events in acute coronary syndromes: the MIRACL study: a randomized controlled trial | journal = [[J Am Med Assoc]] | volume = 285 | issue = 13 | pages = 1711–8 | year = 2001 | pmid = 11277825 | doi = 10.1001/jama.285.13.1711 | doi-access = free }}</ref>)可以用于检测预测变量效果。
   −
* 对于可能删失的时间到事件结果数据,生存分析(如绝经后接受激素替代治疗后冠心病发生时间的卡普兰-迈耶估计值 Kaplan–Meier estimator和考克斯比例风险模型 Cox proportional hazards model<ref name="Rossouw-2002">{{Cite journal | author = Rossouw JE, Anderson GL, [[Ross Prentice|Prentice RL]], LaCroix AZ, Kooperberg C, Stefanick ML, Jackson RD, Beresford SA, Howard BV, Johnson KC, Kotchen JM, Ockene J; Writing Group for the Women's Health Initiative Investigators | title = Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial | journal = [[J Am Med Assoc]] | volume = 288 | issue = 3 | pages = 321–33 | year = 2002 | pmid = 12117397 | doi = 10.1001/jama.288.3.321 | s2cid = 20149703 | url = https://escholarship.org/content/qt3mr6f93p/qt3mr6f93p.pdf?t=prll4c | doi-access = free }}</ref>)是合适的。
+
* 对于可能删失的时间到事件结果数据,生存分析(如绝经后接受激素替代治疗后冠心病发生时间的卡普兰-迈耶估计值 Kaplan–Meier estimator和考克斯比例风险模型 Cox proportional hazards model<ref name="Rossouw-2002">{{Cite journal | author = Rossouw JE, Anderson GL, [[Ross Prentice|Prentice RL]], LaCroix AZ, Kooperberg C, Stefanick ML, Jackson RD, Beresford SA, Howard BV, Johnson KC, Kotchen JM, Ockene J; Writing Group for the Women's Health Initiative Investigators | title = Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial | journal = [[J Am Med Assoc]] | volume = 288 | issue = 3 | pages = 321–33 | year = 2002 | pmid = 12117397 | doi = 10.1001/jama.288.3.321 | s2cid = 20149703 | url = https://escholarship.org/content/qt3mr6f93p/qt3mr6f93p.pdf?t=prll4c | doi-access = free }}</ref>)是合适的。
      第188行: 第188行:  
* 由于中期结果,是否应该提前停止RCT。例如,如果干预产生“大于预期的益处或危害”,或者如果“研究者发现实验干预和对照干预之间没有重要区别的证据”,则可能会提前停止随机对照试验。<ref name="Moher-2010"/>
 
* 由于中期结果,是否应该提前停止RCT。例如,如果干预产生“大于预期的益处或危害”,或者如果“研究者发现实验干预和对照干预之间没有重要区别的证据”,则可能会提前停止随机对照试验。<ref name="Moher-2010"/>
   −
* 这些组在多大程度上可以完全按照随机化时的状态进行分析(即,是否使用了所谓的“意向治疗分析”)。一项“纯”意向治疗分析“只有在获得所有随机受试者的完整结果数据时才有可能”;<ref name="Hollis-1999">{{Cite journal |vauthors=Hollis S, Campbell F | title = What is meant by intention to treat analysis? Survey of published randomised controlled trials | journal = [[Br Med J]] | volume = 319 | issue = 7211 | pages = 670–4 | year = 1999 | pmid = 10480822 | pmc = 28218 | doi=10.1136/bmj.319.7211.670}}</ref>当一些结果数据缺失时,选项包括仅分析具有已知结果的病例和使用估算数据。<ref name="Moher-2010"/> 然而,分析越能包括他们被随机分组的所有参与者,RCT受到的偏见就越少。<ref name="Moher-2010"/>
+
* 这些组在多大程度上可以完全按照随机化时的状态进行分析(即,是否使用了所谓的“意向治疗分析”)。一项“纯”意向治疗分析“只有在获得所有随机受试者的完整结果数据时才有可能”;<ref name="Hollis-1999">{{Cite journal |vauthors=Hollis S, Campbell F | title = What is meant by intention to treat analysis? Survey of published randomised controlled trials | journal = [[Br Med J]] | volume = 319 | issue = 7211 | pages = 670–4 | year = 1999 | pmid = 10480822 | pmc = 28218 | doi=10.1136/bmj.319.7211.670}}</ref>当一些结果数据缺失时,选项包括仅分析具有已知结果的病例和使用估算数据。<ref name="Moher-2010"/> 然而,分析越能包括他们被随机分组的所有参与者,RCT受到的偏见就越少。<ref name="Moher-2010"/>
    
* 是否应进行亚组分析。这些“通常是不鼓励的”,因为多次比较可能会产生假阳性结果,而其他研究无法证实。<ref name="Moher-2010"/>
 
* 是否应进行亚组分析。这些“通常是不鼓励的”,因为多次比较可能会产生假阳性结果,而其他研究无法证实。<ref name="Moher-2010"/>
第194行: 第194行:     
== 结果报告 ==
 
== 结果报告 ==
CONSORT 2010声明是“一套基于证据的报告随机对照试验的最低建议。”CONSORT 2010核对表包含25个项目(许多带有子项目),重点关注“个体随机、两组、平行试验”,这是RCT最常见的类型。
+
CONSORT 2010声明是“一套基于证据的报告随机对照试验的最低建议。”CONSORT 2010核对表包含25个项目(许多带有子项目),重点关注“个体随机、两组、平行试验”,这是RCT最常见的类型。
    
对于其他RCT研究设计,“CONSORT扩展版”已经发布,一些例子是:
 
对于其他RCT研究设计,“CONSORT扩展版”已经发布,一些例子是:
第212行: 第212行:  
=== 统计结果解读 ===
 
=== 统计结果解读 ===
   −
与所有统计方法一样,随机对照试验同时存在ⅰ型(“假阳性”)和ⅱ型(“假阴性”)统计误差。关于第一类错误,典型的RCT将使用0.05(即20分之一)作为RCT错误地发现两种同等有效的治疗方法显著不同的概率。关于第二类错误,尽管1978年发表的一篇论文指出,许多“阴性”随机对照试验的样本量太小,无法对阴性结果做出明确的结论,但到2005-2006年,相当大比例的随机对照试验仍然有不准确或不完全报告的样本量计算。
+
与所有统计方法一样,随机对照试验同时存在ⅰ型(“假阳性”)和ⅱ型(“假阴性”)统计误差。关于第一类错误,典型的RCT将使用0.05(即20分之一)作为RCT错误地发现两种同等有效的治疗方法显著不同的概率。关于第二类错误,尽管1978年发表的一篇论文指出,许多“阴性”随机对照试验的样本量太小,无法对阴性结果做出明确的结论,但到2005-2006年,相当大比例的随机对照试验仍然有不准确或不完全报告的样本量计算。
       
=== 同行评审 ===
 
=== 同行评审 ===
   −
结果的同行评审是科学方法的重要组成部分。审查者检查研究结果是否存在可能导致不可靠结果的潜在设计问题(例如,通过产生系统偏差),在相关研究和其他证据的背景下评估研究,并评估是否可以合理地认为研究已经证明了其结论。为了强调同行评审的必要性和过度概括结论的危险,两位波士顿地区的医学研究人员进行了一项随机对照试验,他们随机给23名从双翼飞机或直升机上跳下的志愿者分配了一个降落伞或一个空背包。这项研究能够准确地报告,与空背包相比,降落伞不能减少伤害。限制这一结论普遍适用性的关键背景是,飞机停在地面上,参与者只跳了大约两英尺。
+
结果的同行评审是科学方法的重要组成部分。审查者检查研究结果是否存在可能导致不可靠结果的潜在设计问题(例如,通过产生系统偏差),在相关研究和其他证据的背景下评估研究,并评估是否可以合理地认为研究已经证明了其结论。为了强调同行评审的必要性和过度概括结论的危险,两位波士顿地区的医学研究人员进行了一项随机对照试验,他们随机给23名从双翼飞机或直升机上跳下的志愿者分配了一个降落伞或一个空背包。这项研究能够准确地报告,与空背包相比,降落伞不能减少伤害。限制这一结论普遍适用性的关键背景是,飞机停在地面上,参与者只跳了大约两英尺。
    
== 优势 ==
 
== 优势 ==
  −
RCTs are considered to be the most reliable form of [[scientific method|scientific evidence]] in the [[hierarchy of evidence]] that influences healthcare policy and practice because RCTs reduce spurious causality and bias. Results of RCTs may be combined in [[systematic review]]s which are increasingly being used in the conduct of [[evidence-based practice]]. Some examples of scientific organizations' considering RCTs or systematic reviews of RCTs to be the highest-quality evidence available are:
  −
* As of 1998, the [[National Health and Medical Research Council]] of Australia designated "Level I" evidence as that "obtained from a [[systematic review]] of all relevant randomised controlled trials" and "Level II" evidence as that "obtained from at least one properly designed randomised controlled trial."<ref>{{cite book |url= http://www.nhmrc.gov.au/_files_nhmrc/file/publications/synopses/cp30.pdf |title= A guide to the development, implementation and evaluation of clinical practice guidelines |author= National Health and Medical Research Council |date=1998-11-16 |publisher= Commonwealth of Australia |location = Canberra | isbn = 978-1-86496-048-8 |page= 56|access-date=2010-03-28}}</ref>
  −
* Since at least 2001, in making [[clinical practice guideline]] recommendations the [[United States Preventive Services Task Force]] has considered both a study's design and its [[internal validity]] as indicators of its quality.<ref name="Harris-2001">{{Cite journal | author = Harris RP, Helfand M, Woolf SH, Lohr KN, Mulrow CD, Teutsch SM, Atkins D; Methods Work Group, Third US Preventive Services Task Force | title = Current methods of the US Preventive Services Task Force: a review of the process | journal = Am J Prev Med | volume = 20 | issue = 3 Suppl | pages = 21–35 | year = 2001 | url = http://www.ahrq.gov/clinic/ajpmsuppl/review.pdf | pmid = 11306229 | doi = 10.1016/S0749-3797(01)00261-6 }}</ref> It has recognized "evidence obtained from at least one properly randomized controlled trial" with good [[internal validity]] (i.e., a rating of "I-good") as the highest quality evidence available to it.<ref name="Harris-2001"/>
  −
* The [[Evidence-based medicine#Grade Working Group|GRADE Working Group]] concluded in 2008 that "randomised trials without important limitations constitute high quality evidence."<ref name="Guyatt-2008">{{Cite journal | author = Guyatt GH, Oxman AD, Kunz R, Vist GE, Falck-Ytter Y, Schünemann HJ; GRADE Working Group | title = What is "quality of evidence" and why is it important to clinicians? | journal = BMJ | volume = 336 | issue = 7651 | pages = 995–8 |year = 2008 | doi = 10.1136/bmj.39490.551019.BE | pmc = 2364804 | pmid = 18456631 }}</ref>
  −
* For issues involving "Therapy/Prevention, Aetiology/Harm", the [[Oxford Centre for Evidence-based Medicine]] as of 2011 defined "Level 1a" evidence as a systematic review of RCTs that are consistent with each other, and "Level 1b" evidence as an "individual RCT (with narrow [[Confidence Interval]])."<ref>{{cite web |url= http://www.cebm.net/index.aspx?o=1025 |title= Levels of evidence |author= Oxford Centre for Evidence-based Medicine |date=2011-09-16 |access-date=2012-02-15}}</ref>
      
RCT被认为是影响医疗保健政策和实践的证据层次中最可靠的科学证据形式,因为RCT减少了虚假的因果关系和偏见。随机对照试验的结果可以在系统综述中结合使用,越来越多地用于循证实践。一些科学组织认为随机对照试验或随机对照试验的系统审查是现有的最高质量证据的例子有:
 
RCT被认为是影响医疗保健政策和实践的证据层次中最可靠的科学证据形式,因为RCT减少了虚假的因果关系和偏见。随机对照试验的结果可以在系统综述中结合使用,越来越多地用于循证实践。一些科学组织认为随机对照试验或随机对照试验的系统审查是现有的最高质量证据的例子有:
* 截至1998年,澳大利亚国家卫生和医学研究委员会将“一级”证据指定为“从所有相关随机对照试验的系统审查中获得的”,将“二级”证据指定为“从至少一项适当设计的随机对照试验中获得的”
+
* 截至1998年,澳大利亚国家卫生和医学研究委员会将“一级”证据指定为“从所有相关随机对照试验的系统审查中获得的”,将“二级”证据指定为“从至少一项适当设计的随机对照试验中获得的”<ref>{{cite book |url= http://www.nhmrc.gov.au/_files_nhmrc/file/publications/synopses/cp30.pdf |title= A guide to the development, implementation and evaluation of clinical practice guidelines |author= National Health and Medical Research Council |date=1998-11-16 |publisher= Commonwealth of Australia |location = Canberra | isbn = 978-1-86496-048-8 |page= 56|access-date=2010-03-28}}</ref>
   −
* 至少自2001年以来,美国预防服务工作组在提出临床实践指南建议时,将研究的设计及其内部有效性作为其质量的指标。它承认“从至少一个适当的随机对照试验中获得的证据”具有良好的内部有效性(即“良好”评级),是它所能获得的最高质量的证据。
+
* 至少自2001年以来,美国预防服务工作组在提出临床实践指南建议时,将研究的设计及其内部有效性作为其质量的指标。<ref name="Harris-2001">{{Cite journal | author = Harris RP, Helfand M, Woolf SH, Lohr KN, Mulrow CD, Teutsch SM, Atkins D; Methods Work Group, Third US Preventive Services Task Force | title = Current methods of the US Preventive Services Task Force: a review of the process | journal = Am J Prev Med | volume = 20 | issue = 3 Suppl | pages = 21–35 | year = 2001 | url = http://www.ahrq.gov/clinic/ajpmsuppl/review.pdf | pmid = 11306229 | doi = 10.1016/S0749-3797(01)00261-6 }}</ref>它承认“从至少一个适当的随机对照试验中获得的证据”具有良好的内部有效性(即“良好”评级),是它所能获得的最高质量的证据。<ref name="Harris-2001"/>
   −
* GRADE工作组在2008年得出结论,“没有重要限制的随机试验构成了高质量的证据。”
+
* GRADE工作组在2008年得出结论,“没有重要限制的随机试验构成了高质量的证据。”<ref name="Guyatt-2008">{{Cite journal | author = Guyatt GH, Oxman AD, Kunz R, Vist GE, Falck-Ytter Y, Schünemann HJ; GRADE Working Group | title = What is "quality of evidence" and why is it important to clinicians? | journal = BMJ | volume = 336 | issue = 7651 | pages = 995–8 |year = 2008 | doi = 10.1136/bmj.39490.551019.BE | pmc = 2364804 | pmid = 18456631 }}</ref>
   −
* 对于涉及“治疗/预防、病因学/危害”的问题,截至2011年,牛津循证医学中心将“1a级”证据定义为相互一致的随机对照试验的系统审查,“1b级”证据定义为“个体RCT(置信区间较窄)。”
+
* 对于涉及“治疗/预防、病因学/危害”的问题,截至2011年,牛津循证医学中心将“1a级”证据定义为相互一致的随机对照试验的系统审查,“1b级”证据定义为“个体RCT(置信区间较窄)。”<ref>{{cite web |url= http://www.cebm.net/index.aspx?o=1025 |title= Levels of evidence |author= Oxford Centre for Evidence-based Medicine |date=2011-09-16 |access-date=2012-02-15}}</ref>
     
7,129

个编辑

导航菜单