临床试验中处理缺失数据和信息性删失的问题。
Problems in dealing with missing data and informative censoring in clinical trials.
作者信息
Shih Weichung
机构信息
Division of Biometrics, University of Medicine and Dentistry of New Jersey School of Public Health, New Brunswick, New Jersey, USA.
出版信息
Curr Control Trials Cardiovasc Med. 2002 Jan 8;3(1):4. doi: 10.1186/1468-6708-3-4.
Abstract
Acommon problem in clinical trials is the missing data that occurs when patients do not complete the study and drop out without further measurements. Missing data cause the usual statistical analysis of complete or all available data to be subject to bias. There are no universally applicable methods for handling missing data. We recommend the following: (1) Report reasons for dropouts and proportions for each treatment group; (2) Conduct sensitivity analyses to encompass different scenarios of assumptions and discuss consistency or discrepancy among them; (3) Pay attention to minimize the chance of dropouts at the design stage and during trial monitoring; (4) Collect post-dropout data on the primary endpoints, if at all possible; and (5) Consider the dropout event itself an important endpoint in studies with many.
摘要
临床试验中的一个常见问题是出现缺失数据，即患者未完成研究且未进行进一步测量就退出。缺失数据会导致对完整或所有可用数据进行的常规统计分析产生偏差。目前尚无普遍适用的处理缺失数据的方法。我们建议如下：（1）报告各治疗组的退出原因及比例；（2）进行敏感性分析以涵盖不同的假设情况，并讨论它们之间的一致性或差异；（3）在设计阶段和试验监测期间注意尽量减少退出的可能性；（4）如果可能，收集关于主要终点的退出后数据；以及（5）在有许多退出情况的研究中，将退出事件本身视为一个重要终点。
相似文献
1
Problems in dealing with missing data and informative censoring in clinical trials.临床试验中处理缺失数据和信息性删失的问题。
Curr Control Trials Cardiovasc Med. 2002 Jan 8;3(1):4. doi: 10.1186/1468-6708-3-4.
2
Modern statistical methods for handling missing repeated measurements in obesity trial data: beyond LOCF.处理肥胖试验数据中缺失重复测量值的现代统计方法：超越末次观测结转法。
Obes Rev. 2003 Aug;4(3):175-84. doi: 10.1046/j.1467-789x.2003.00109.x.
3
How should we deal with missing data in clinical trials involving Alzheimer's disease patients?在涉及阿尔茨海默病患者的临床试验中，我们应该如何处理缺失数据？
Curr Alzheimer Res. 2011 Jun;8(4):421-33. doi: 10.2174/156720511795745339.
4
Joint modelling with competing risks of dropout for longitudinal analysis of health-related quality of life in cancer clinical trials.用于癌症临床试验中健康相关生活质量纵向分析的具有失访竞争风险的联合建模。
Qual Life Res. 2022 May;31(5):1359-1370. doi: 10.1007/s11136-021-03040-8. Epub 2021 Nov 24.
5
Handling trial participants with missing outcome data when conducting a meta-analysis: a systematic survey of proposed approaches.进行荟萃分析时处理结局数据缺失的试验参与者：对提议方法的系统调查
Syst Rev. 2015 Jul 23;4:98. doi: 10.1186/s13643-015-0083-6.
6
HIV viral dynamic models with dropouts and missing covariates.具有失访和协变量缺失的HIV病毒动力学模型。
Stat Med. 2007 Jul 30;26(17):3342-57. doi: 10.1002/sim.2816.
7
Intention-to-treat: methods for dealing with missing values in clinical trials of progressively deteriorating diseases.意向性分析：进行性恶化疾病临床试验中处理缺失值的方法。
Stat Med. 2001 Dec 30;20(24):3931-46. doi: 10.1002/sim.1149.
8
Considerations of multiple imputation approaches for handling missing data in clinical trials.考虑在临床试验中采用多种插补方法处理缺失数据。
Contemp Clin Trials. 2018 Jul;70:62-71. doi: 10.1016/j.cct.2018.05.008. Epub 2018 May 17.
9
Handling missing data in vaccine clinical trials for immunogenicity and safety evaluation.疫苗临床试验中免疫原性和安全性评估的缺失数据处理
J Biopharm Stat. 2011 Mar;21(2):294-310. doi: 10.1080/10543406.2011.550111.
10
The treatment of missing data in a large cardiovascular clinical outcomes study.一项大型心血管临床结局研究中缺失数据的处理
Clin Trials. 2016 Jun;13(3):344-51. doi: 10.1177/1740774515626411. Epub 2016 Feb 17.
引用本文的文献
1
Prioritizing attributes of approaches to analyzing patient-centered outcomes that are truncated due to death in critical care clinical trials: a Delphi study.对重症监护临床试验中因死亡而被截断的以患者为中心结局分析方法的属性进行优先级排序：一项德尔菲研究。
Trials. 2025 Jan 10;26(1):15. doi: 10.1186/s13063-024-08673-x.
2
Assessing the Feasibility and Efficacy of Virtual Reality Navigational Training for Older Adults.评估虚拟现实导航训练对老年人的可行性和有效性。
Innov Aging. 2024 Dec 12;9(1):igae099. doi: 10.1093/geroni/igae099. eCollection 2025.
3
Informative censoring in externally controlled clinical trials: a potential source of bias.外部对照临床试验中的信息删失：偏倚的一个潜在来源。
ESMO Open. 2025 Jan;10(1):104094. doi: 10.1016/j.esmoop.2024.104094. Epub 2025 Jan 3.
4
The OHStat Guidelines for Reporting Observational Studies and Clinical Trials in Oral Health Research: Explanation and Elaboration.《口腔健康研究中观察性研究和临床试验报告的OHStat指南：解释与阐述》
JDR Clin Trans Res. 2024 Jul 12:23800844241247029. doi: 10.1177/23800844241247029.
5
Imputation of Missing Data for Time-to-Event Endpoints Using Retrieved Dropouts.利用追回的脱落数据对事件时间终点缺失数据进行插补。
Ther Innov Regul Sci. 2024 Jan;58(1):114-126. doi: 10.1007/s43441-023-00575-5. Epub 2023 Oct 7.
6
A three-arm randomized controlled trial using ecological momentary intervention, community health workers, and video feedback at family meals to improve child cardiovascular health: the Family Matters study design.采用生态瞬时干预、社区卫生工作者和家庭餐视频反馈的三臂随机对照试验改善儿童心血管健康：家庭事务研究设计。
BMC Public Health. 2023 Apr 19;23(1):708. doi: 10.1186/s12889-023-15504-2.
7
A Three-Arm Randomized Controlled Trial Using Ecological Momentary Intervention, Community Health Workers, and Video Feedback at Family Meals to Improve Child Cardiovascular Health: The Family Matters Study Design.一项三臂随机对照试验，采用生态瞬时干预、社区卫生工作者和家庭用餐时的视频反馈来改善儿童心血管健康：家庭重要性研究设计。
Res Sq. 2023 Mar 22:rs.3.rs-2662682. doi: 10.21203/rs.3.rs-2662682/v1.
8
An Expectation-Maximization Algorithm for Including Oncological COVID-19 Deaths in Survival Analysis.一种在生存分析中纳入癌症 COVID-19 死亡的期望最大化算法。
Curr Oncol. 2023 Feb 8;30(2):2105-2126. doi: 10.3390/curroncol30020163.
9
Post-covid medical complaints following infection with SARS-CoV-2 Omicron vs Delta variants.感染 SARS-CoV-2 奥密克戎和德尔塔变异株后的新冠后医疗投诉。
Nat Commun. 2022 Nov 30;13(1):7363. doi: 10.1038/s41467-022-35240-2.
10
Field trial investigating the efficacy of a long-acting imidacloprid 10%/flumethrin 4.5% polymer matrix collar (Seresto®, Elanco) compared to monthly topical fipronil for the chemoprevention of canine tick-borne pathogens in Cambodia.在柬埔寨进行的一项田间试验，比较长效吡虫啉10%/氟氯氰菊酯4.5%聚合物基质项圈（Seresto®，伊兰科公司）与每月一次外用氟虫腈对犬蜱传播病原体的化学预防效果。
Curr Res Parasitol Vector Borne Dis. 2022 Jun 28;2:100095. doi: 10.1016/j.crpvbd.2022.100095. eCollection 2022.
本文引用的文献
1
Efficacy and safety of omapatrilat with hydrochlorothiazide for the treatment of hypertension in subjects nonresponsive to hydrochlorothiazide alone.奥美帕替利与氢氯噻嗪联合治疗对单独使用氢氯噻嗪无反应的高血压患者的疗效和安全性。
Am J Hypertens. 2001 Aug;14(8 Pt 1):788-93. doi: 10.1016/s0895-7061(01)01294-8.
2
Effect of niacin and atorvastatin on lipoprotein subclasses in patients with atherogenic dyslipidemia.烟酸和阿托伐他汀对致动脉粥样硬化血脂异常患者脂蛋白亚类的影响。
Am J Cardiol. 2001 Aug 1;88(3):270-4. doi: 10.1016/s0002-9149(01)01639-3.
3
Air Force/Texas Coronary Atherosclerosis Prevention Study (AFCAPS/TEXCAPS): additional perspectives on tolerability of long-term treatment with lovastatin.空军/德克萨斯冠状动脉粥样硬化预防研究（AFCAPS/TEXCAPS）：关于洛伐他汀长期治疗耐受性的更多观点。
Am J Cardiol. 2001 May 1;87(9):1074-9. doi: 10.1016/s0002-9149(01)01464-3.
4
Partial imputation approach to analysis of repeated measurements with dependent drop-outs.用于分析存在相依性失访的重复测量数据的部分插补方法。
Stat Med. 2001 Apr 30;20(8):1197-214. doi: 10.1002/sim.778.
5
Comparison of two calcium blockers on hemodynamics, left ventricular mass, and coronary vasodilatory in advanced hypertension.两种钙通道阻滞剂对重度高血压患者血流动力学、左心室质量和冠状动脉舒张功能的比较
Am J Hypertens. 2001 Mar;14(3):231-40. doi: 10.1016/s0895-7061(00)01267-x.
6
Comparison of candesartan versus enalapril in essential hypertension. Italian Candesartan Study Group.坎地沙坦与依那普利治疗原发性高血压的比较。意大利坎地沙坦研究组。
Am J Hypertens. 2001 Feb;14(2):129-34. doi: 10.1016/s0895-7061(00)01240-1.
7
Adjusting for drop-out in clinical trials with repeated measures: design and analysis issues.在重复测量的临床试验中对失访进行调整：设计与分析问题
Stat Med. 2001 Jan 15;20(1):93-108. doi: 10.1002/1097-0258(20010115)20:1<93::aid-sim655>3.0.co;2-2.
8
Randomised comparison of losartan vs. captopril on quality of life in elderly patients with symptomatic heart failure: the losartan heart failure ELITE quality of life substudy.氯沙坦与卡托普利对有症状心力衰竭老年患者生活质量影响的随机对照研究：氯沙坦心力衰竭ELITE生活质量子研究
Qual Life Res. 2000;9(4):377-84. doi: 10.1023/a:1008948930206.
9
Worst-rank score analysis with informatively missing observations in clinical trials.在临床试验中对存在信息性缺失观测值的最差排名分数分析。
Control Clin Trials. 1999 Oct;20(5):408-22. doi: 10.1016/s0197-2456(99)00022-7.
10
Combining mortality and longitudinal measures in clinical trials.在临床试验中结合死亡率和纵向测量指标。
Stat Med. 1999 Jun 15;18(11):1341-54. doi: 10.1002/(sici)1097-0258(19990615)18:11<1341::aid-sim129>3.0.co;2-7.
Zotero 插件