对纵向心血管健康研究数据应用不同的多重填补方法，结果有所不同。
Results differ by applying distinctive multiple imputation approaches on the longitudinal cardiovascular health study data.
作者信息
Ning Yuming, McAvay Gail, Chaudhry Sarwat I, Arnold Alice M, Allore Heather G
机构信息
Section of Geriatrics, Department of Internal Medicine, Yale University School of Medicine, New Haven, CT 06511, USA.
出版信息
Exp Aging Res. 2013;39(1):27-43. doi: 10.1080/0361073X.2013.741968.
Abstract
UNLABELLED
BACKGROUND/STUDY CONTEXT: The objective of this study was to examine sequential and simultaneous approaches to multiple imputation of missing data in a longitudinal data set where losses due to death were common.
METHODS
Comparison of results from analyses and simulations of time to incident difficulty of activities of daily living (ADL) in the Cardiovascular Health Study when missing data were imputed simultaneously or sequentially.
RESULTS
Results differed with imputation methods. The largest proportional differences in 12 risk factor parameter estimates were heart failure by 106%, social support by 33%, and arthritis by 27%.
CONCLUSION
Decedents' final characteristics were influential on future imputations of those with missing values.
摘要
未加标注
背景/研究背景：本研究的目的是检验在纵向数据集中对缺失数据进行多次插补的顺序法和同步法，在该数据集中因死亡导致的数据缺失很常见。
方法
在心血管健康研究中，当对缺失数据进行同步或顺序插补时，比较分析结果和对日常生活活动（ADL）发生困难时间的模拟结果。
结果
插补方法不同，结果也不同。12个风险因素参数估计值的最大比例差异为心力衰竭106%、社会支持33%、关节炎27%。
结论
死者的最终特征对缺失值者未来的插补有影响。
相似文献
1
Results differ by applying distinctive multiple imputation approaches on the longitudinal cardiovascular health study data.对纵向心血管健康研究数据应用不同的多重填补方法，结果有所不同。
Exp Aging Res. 2013;39(1):27-43. doi: 10.1080/0361073X.2013.741968.
2
Multiple imputation of baseline data in the cardiovascular health study.心血管健康研究中基线数据的多重填补
Am J Epidemiol. 2003 Jan 1;157(1):74-84. doi: 10.1093/aje/kwf156.
3
Multiple imputation of missing values was not necessary before performing a longitudinal mixed-model analysis.在进行纵向混合模型分析之前，不需要对缺失值进行多重插补。
J Clin Epidemiol. 2013 Sep;66(9):1022-8. doi: 10.1016/j.jclinepi.2013.03.017. Epub 2013 Jun 21.
4
Imputation of missing longitudinal data: a comparison of methods.缺失纵向数据的插补：方法比较
J Clin Epidemiol. 2003 Oct;56(10):968-76. doi: 10.1016/s0895-4356(03)00170-7.
5
Missing data in longitudinal studies: cross-sectional multiple imputation provides similar estimates to full-information maximum likelihood.纵向研究中的缺失数据：横断面多项插补提供的估计值与完全信息极大似然法相似。
Ann Epidemiol. 2014 Jan;24(1):75-7. doi: 10.1016/j.annepidem.2013.10.007. Epub 2013 Oct 18.
6
Diagnosing problems with imputation models using the Kolmogorov-Smirnov test: a simulation study.使用柯尔莫哥洛夫-斯米尔诺夫检验诊断插补模型中的问题：一项模拟研究。
BMC Med Res Methodol. 2013 Nov 20;13:144. doi: 10.1186/1471-2288-13-144.
7
The performance of prognostic models depended on the choice of missing value imputation algorithm: a simulation study.预后模型的性能取决于缺失值插补算法的选择：一项模拟研究。
J Clin Epidemiol. 2024 Dec;176:111539. doi: 10.1016/j.jclinepi.2024.111539. Epub 2024 Sep 24.
8
Multiple imputation for non-response when estimating HIV prevalence using survey data.使用调查数据估计艾滋病毒流行率时对无应答情况的多重填补法
BMC Public Health. 2015 Oct 16;15:1059. doi: 10.1186/s12889-015-2390-1.
9
Development and Validation of a Large Synthetic Cohort for the Study of Cardiovascular Health Across the Life Span.开发和验证一个大型综合队列，以研究整个生命周期的心血管健康。
Am J Epidemiol. 2021 Oct 1;190(10):2208-2219. doi: 10.1093/aje/kwab137.
10
Trajectories of activities of daily living/ instrumental activities of daily living and the risk of cardiovascular diseases.日常生活活动/工具性日常生活活动轨迹与心血管疾病风险
Nutr Metab Cardiovasc Dis. 2025 Jun;35(6):103969. doi: 10.1016/j.numecd.2025.103969. Epub 2025 Mar 6.
引用本文的文献
1
Recovery from Severe Disability that Develops Progressively Versus Catastrophically: Incidence, Risk Factors, and Intervening Events.从逐渐发展和灾难性发展的严重残疾中恢复：发生率、风险因素和干预事件。
J Am Geriatr Soc. 2020 Sep;68(9):2067-2073. doi: 10.1111/jgs.16567. Epub 2020 Jun 3.
2
Risk Factors and Precipitants of Severe Disability Among Community-Living Older Persons.社区居住老年人严重残疾的风险因素和促发因素。
JAMA Netw Open. 2020 Jun 1;3(6):e206021. doi: 10.1001/jamanetworkopen.2020.6021.
3
Refining the Association Between Body Mass Index and Atrial Fibrillation: G-Formula and Restricted Mean Survival Times.体重指数与心房颤动相关性的再分析：G 公式与限制平均生存时间。
J Am Heart Assoc. 2019 Aug 20;8(16):e013011. doi: 10.1161/JAHA.119.013011. Epub 2019 Aug 8.
4
A statistical analysis protocol for the time-differentiated target temperature management after out-of-hospital cardiac arrest (TTH48) clinical trial.院外心脏骤停后时间分辨目标温度管理（TTH48）临床试验的统计分析方案
Scand J Trauma Resusc Emerg Med. 2016 Nov 28;24(1):138. doi: 10.1186/s13049-016-0334-0.
5
Time-limited home-care reablement services for maintaining and improving the functional independence of older adults.限时居家护理康复服务以维持和改善老年人的功能独立性。
Cochrane Database Syst Rev. 2016 Oct 11;10(10):CD010825. doi: 10.1002/14651858.CD010825.pub2.
6
Sensitivity analysis in multiple imputation in effectiveness studies of psychotherapy.心理治疗有效性研究中多重填补法的敏感性分析
Front Psychol. 2015 Jul 27;6:1042. doi: 10.3389/fpsyg.2015.01042. eCollection 2015.
7
A longitudinal, observational study with many repeated measures demonstrated improved precision of individual survival curves using Bayesian joint modeling of disability and survival.一项采用多次重复测量的纵向观察性研究表明，使用残疾与生存的贝叶斯联合模型可提高个体生存曲线的精度。
Exp Aging Res. 2015;41(3):221-39. doi: 10.1080/0361073X.2015.1021640.
8
Using multiple imputation to deal with missing data and attrition in longitudinal studies with repeated measures of patient-reported outcomes.使用多重插补处理具有患者报告结局重复测量的纵向研究中的缺失数据和流失。
Clin Epidemiol. 2015 Jan 16;7:91-106. doi: 10.2147/CLEP.S72247. eCollection 2015.
9
Functional Status and Social Contact Among Older Adults.老年人的功能状态与社会交往
Res Aging. 2015 Nov;37(8):815-36. doi: 10.1177/0164027514566091. Epub 2015 Jan 8.
10
Respiratory impairment in older persons: when less means more.老年人的呼吸功能障碍：少即是多。
Am J Med. 2013 Jan;126(1):49-57. doi: 10.1016/j.amjmed.2012.07.016. Epub 2012 Nov 20.
本文引用的文献
1
Treatment of death in the analysis of longitudinal studies of gerontological outcomes.对老年结局纵向研究中死亡的处理。
J Gerontol A Biol Sci Med Sci. 2011 Jan;66(1):109-14. doi: 10.1093/gerona/glq188. Epub 2010 Oct 28.
2
Geriatric impairments and disability: the cardiovascular health study.老年功能障碍和残疾：心血管健康研究。
J Am Geriatr Soc. 2010 Sep;58(9):1686-92. doi: 10.1111/j.1532-5415.2010.03022.x.
3
Longitudinal Data with Follow-up Truncated by Death: Match the Analysis Method to Research Aims.随访因死亡而截断的纵向数据：使分析方法与研究目的相匹配。
Stat Sci. 2009;24(2):211. doi: 10.1214/09-STS293.
4
Missing data: a special challenge in aging research.缺失数据：衰老研究中的一项特殊挑战。
J Am Geriatr Soc. 2009 Apr;57(4):722-9. doi: 10.1111/j.1532-5415.2008.02168.x. Epub 2009 Feb 10.
5
The use of missingness screens in clinical epidemiologic research has implications for regression modeling.在临床流行病学研究中使用缺失值筛查对回归建模有影响。
J Clin Epidemiol. 2007 Dec;60(12):1239-45. doi: 10.1016/j.jclinepi.2007.03.006. Epub 2007 Jul 25.
6
Population inference with mortality and attrition in longitudinal studies on aging: a two-stage multiple imputation method.纵向衰老研究中考虑死亡率和失访情况的总体推断：一种两阶段多重填补方法
Exp Aging Res. 2007 Apr-Jun;33(2):187-203. doi: 10.1080/03610730701239004.
7
An index of local sensitivity to nonignorable drop-out in longitudinal modelling.纵向建模中对不可忽略缺失值的局部敏感性指标。
Stat Med. 2005 Jul 30;24(14):2129-50. doi: 10.1002/sim.2107.
8
A method for imputing missing data in longitudinal studies.一种在纵向研究中插补缺失数据的方法。
Ann Epidemiol. 2004 May;14(5):354-61. doi: 10.1016/j.annepidem.2003.09.010.
9
Imputation of missing longitudinal data: a comparison of methods.缺失纵向数据的插补：方法比较
J Clin Epidemiol. 2003 Oct;56(10):968-76. doi: 10.1016/s0895-4356(03)00170-7.
10
Multiple imputation of baseline data in the cardiovascular health study.心血管健康研究中基线数据的多重填补
Am J Epidemiol. 2003 Jan 1;157(1):74-84. doi: 10.1093/aje/kwf156.
Zotero 插件