A comparison of imputation strategies in cluster randomized trials with missing binary outcomes

被引:10
|
作者
Caille, Agnes [1 ,2 ,3 ,4 ]
Leyrat, Clemence [1 ,2 ,3 ]
Giraudeau, Bruno [1 ,2 ,3 ,4 ]
机构
[1] INSERM, U1153, Paris, France
[2] INSERM, CIC 1415, 2 Bd Tonnelle, F-37044 Tours 9, France
[3] CHRU Tours, Tours, France
[4] Univ Francois Rabelais Tours, PRES Ctr, Val de Loire Univ, Tours, France
关键词
cluster randomized trial; missing data; outcome; multiple imputation; MULTIPLE-IMPUTATION; CLINICAL-TRIALS; CORRELATION-COEFFICIENT; INTRACLASS CORRELATION; TREAT; INTENTION; DESIGN; MODELS;
D O I
10.1177/0962280214530030
中图分类号
R19 [保健组织与事业(卫生事业管理)];
学科分类号
摘要
In cluster randomized trials, clusters of subjects are randomized rather than subjects themselves, and missing outcomes are a concern as in individual randomized trials. We assessed strategies for handling missing data when analysing cluster randomized trials with a binary outcome; strategies included complete case, adjusted complete case, and simple and multiple imputation approaches. We performed a simulation study to assess bias and coverage rate of the population-averaged intervention-effect estimate. Both multiple imputation with a random-effects logistic regression model or classical logistic regression provided unbiased estimates of the intervention effect. Both strategies also showed good coverage properties, even slightly better for multiple imputation with a random-effects logistic regression approach. Finally, this latter approach led to a slightly negatively biased intracluster correlation coefficient estimate but less than that with a classical logistic regression model strategy. We applied these strategies to a real trial randomizing households and comparing ivermectin and malathion to treat head lice.
引用
收藏
页码:2650 / 2669
页数:20
相关论文
共 50 条
  • [41] Using Multiple Imputation with GEE with Non-monotone Missing Longitudinal Binary Outcomes
    Stuart R. Lipsitz
    Garrett M. Fitzmaurice
    Roger D. Weiss
    Psychometrika, 2020, 85 : 890 - 904
  • [42] The use of multiple imputation (MI) in cluster randomised trials with suspected missing not at random (MNAR) outcome
    Playle, Rebecca
    Coulman, Elinor
    Gallagher, Dunla
    Simpson, Sharon
    TRIALS, 2015, 16
  • [43] The use of multiple imputation (MI) in cluster randomised trials with suspected missing not at random (MNAR) outcome
    Rebecca Playle
    Elinor Coulman
    Dunla Gallagher
    Sharon Simpson
    Trials, 16
  • [44] Analysis of binary outcomes with missing data: missing = smoking, last observation carried forward, and a little multiple imputation
    Hedeker, Donald
    Mermelstein, Robin J.
    Demirtas, Hakan
    ADDICTION, 2007, 102 (10) : 1564 - 1573
  • [45] Qualitative Imputation of Missing Potential Outcomes
    Coppock, Alexander
    Kaur, Dipin
    AMERICAN JOURNAL OF POLITICAL SCIENCE, 2022, 66 (03) : 681 - 695
  • [46] Global sensitivity analysis of randomized trials with nonmonotone missing binary outcomes: Application to studies of substance use disorders
    Scharfstein, Daniel O.
    Steingrimsson, Jon
    McDermott, Aidan
    Wang, Chenguang
    Ray, Souvik
    Campbell, Aimee
    Nunes, Edward
    Matthews, Abigail
    BIOMETRICS, 2022, 78 (02) : 649 - 659
  • [47] A comparison of imputation techniques for handling missing predictor values in a risk model with a binary outcome
    Ambler, Gareth
    Omar, Rumana Z.
    Royston, Patrick
    STATISTICAL METHODS IN MEDICAL RESEARCH, 2007, 16 (03) : 277 - 298
  • [48] On small-sample inference in group randomized trials with binary outcomes and cluster-level covariates
    Westgate, Philip M.
    BIOMETRICAL JOURNAL, 2013, 55 (05) : 789 - 806
  • [49] Some design considerations for cluster randomized trials with binary responses
    Singh, Arpan
    Pal, Meghendar
    Singh, Satya Prakash
    COMMUNICATIONS IN STATISTICS-THEORY AND METHODS, 2023, 52 (12) : 4310 - 4328
  • [50] Is the R coefficient of interest in cluster randomized trials with a binary outcome?
    Yepnang, Ariane M. Mbekwe
    Caille, Agnes
    Eldridge, Sandra M.
    Giraudeau, Bruno
    STATISTICAL METHODS IN MEDICAL RESEARCH, 2020, 29 (09) : 2470 - 2480