Missing binary outcomes under covariate‐dependent missingness in cluster randomised trials

Missing outcomes are a commonly occurring problem for cluster randomised trials, which can lead to biased and inefficient inference if ignored or handled inappropriately. Two approaches for analysing such trials are cluster‐level analysis and individual‐level analysis. In this study, we assessed the performance of unadjusted cluster‐level analysis, baseline covariate‐adjusted cluster‐level analysis, random effects logistic regression and generalised estimating equations when binary outcomes are missing under a baseline covariate‐dependent missingness mechanism. Missing outcomes were handled using complete records analysis and multilevel multiple imputation. We analytically show that cluster‐level analyses for estimating risk ratio using complete records are valid if the true data generating model has log link and the intervention groups have the same missingness mechanism and the same covariate effect in the outcome model. We performed a simulation study considering four different scenarios, depending on whether the missingness mechanisms are the same or different between the intervention groups and whether there is an interaction between intervention group and baseline covariate in the outcome model. On the basis of the simulation study and analytical results, we give guidance on the conditions under which each approach is valid. © 2017 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd.

[1]  Michael G. Kenward,et al.  Multiple Imputation and its Application: Carpenter/Multiple Imputation and its Application , 2013 .

[2]  Michael G Kenward,et al.  Are missing data adequately handled in cluster randomised trials? A systematic review and guidelines , 2014, Clinical trials.

[3]  David M. Murray,et al.  Design and Analysis of Group- Randomized Trials , 1998 .

[4]  Harvey Goldstein,et al.  REALCOM-IMPUTE Software for Multilevel Multiple Imputation with Mixed Response Types , 2011 .

[5]  Michael G. Kenward,et al.  Multiple Imputation and its Application , 2013 .

[6]  A Donner,et al.  Developments in cluster randomized trials and Statistics in Medicine , 2007, Statistics in medicine.

[7]  D. Rubin INFERENCE AND MISSING DATA , 1975 .

[8]  M. H. Gail,et al.  Tests for no treatment e?ect in randomized clinical trials , 1988 .

[9]  Russell V. Lenth,et al.  Statistical Analysis With Missing Data (2nd ed.) (Book) , 2004 .

[10]  Parminder Raina,et al.  Comparison of population-averaged and cluster-specific models for the analysis of cluster randomized trials with missing binary outcomes: a simulation study , 2013, BMC Medical Research Methodology.

[11]  J. Bartlett,et al.  Missing continuous outcomes under covariate dependent missingness in cluster randomised trials , 2016, Statistical methods in medical research.

[12]  Allan Donner,et al.  Design and Analysis of Cluster Randomization Trials in Health Research , 2001 .

[13]  Rebecca R Andridge,et al.  Quantifying the impact of fixed effects modeling of clusters in multiple imputation for cluster randomized trials , 2011, Biometrical journal. Biometrische Zeitschrift.

[14]  D. Redden,et al.  Comparing denominator degrees of freedom approximations for the generalized linear mixed model in analyzing binary outcome in small sample cluster-randomized trials , 2015, BMC Medical Research Methodology.

[15]  P. Raina,et al.  Comparing the performance of different multiple imputation strategies for missing binary outcomes in cluster randomized trials: a simulation study , 2012 .

[16]  Allan Donner,et al.  Imputation Strategies for Missing Continuous Outcomes in Cluster Randomized Trials , 2008, Biometrical journal. Biometrische Zeitschrift.

[17]  Lisa Dolovich,et al.  Imputation strategies for missing binary outcomes in cluster randomized trials , 2011, BMC medical research methodology.

[18]  Lena Osterhagen,et al.  Multiple Imputation For Nonresponse In Surveys , 2016 .

[19]  G. Molenberghs,et al.  Linear Mixed Models for Longitudinal Data , 2001 .

[20]  M G Kenward,et al.  Multiple imputation methods for bivariate outcomes in cluster randomised trials , 2016, Statistics in medicine.

[21]  H. Davies,et al.  When can odds ratios mislead? , 1998, BMJ.

[22]  John B. Carlin,et al.  Bias and efficiency of multiple imputation compared with complete‐case analysis for missing covariate values , 2010, Statistics in medicine.

[23]  Jonathan L. Blitstein,et al.  Design and analysis of group-randomized trials: a review of recent methodological developments. , 2004, American journal of public health.

[24]  Xiao-Li Meng,et al.  Multiple-Imputation Inferences with Uncongenial Sources of Input , 1994 .

[25]  Allan Donner,et al.  Cluster randomization trials in epidemiology: theory and application , 1994 .

[26]  B. Giraudeau,et al.  A comparison of imputation strategies in cluster randomized trials with missing binary outcomes , 2016, Statistical methods in medical research.

[27]  Ian R White,et al.  Are missing outcome data adequately handled? A review of published randomized controlled trials in major medical journals , 2004, Clinical trials.

[28]  Julian J. Faraway,et al.  Extending the Linear Model with R , 2004 .

[29]  F. E. Satterthwaite Synthesis of variance , 1941 .

[30]  A Donner,et al.  A methodological review of non-therapeutic intervention trials employing cluster randomization, 1979-1989. , 1990, International journal of epidemiology.

[31]  David M. Murray,et al.  Methods To Reduce The Impact Of Intraclass Correlation In Group-Randomized Trials , 2003, Evaluation review.

[32]  Ewout W Steyerberg,et al.  Covariate adjustment in randomized controlled trials with dichotomous outcomes increases statistical power and reduces sample size requirements. , 2004, Journal of clinical epidemiology.

[33]  Nicole A. Lazar,et al.  Statistical Analysis With Missing Data , 2003, Technometrics.

[34]  M. Kenward,et al.  Multiple imputation for missing data in epidemiological and clinical research: potential and pitfalls , 2009, BMJ : British Medical Journal.

[35]  S. Chinn,et al.  Intraclass correlation coefficient and outcome prevalence are associated in clustered binary data. , 2005, Journal of clinical epidemiology.

[36]  Roderick J. A. Little,et al.  Statistical Analysis with Missing Data: Little/Statistical Analysis with Missing Data , 2002 .

[37]  Matteo Quartagno,et al.  jomo: Multilevel Joint Modelling Multiple Imputation , 2016 .

[38]  J. Carlin,et al.  A simulation study of odds ratio estimation for binary outcomes from cluster randomized trials , 2006, Statistics in medicine.

[39]  Thomas H. Davenport,et al.  How to Design , 2009 .

[40]  D. Rubin,et al.  Multiple Imputation for Interval Estimation from Simple Random Samples with Ignorable Nonresponse , 1986 .

[41]  A Rogier T Donders,et al.  Dealing with missing outcome data in randomized trials and observational studies. , 2012, American journal of epidemiology.

[42]  T. Derouen,et al.  A Covariance Estimator for GEE with Improved Small‐Sample Properties , 2001, Biometrics.

[43]  Michael J. Campbell,et al.  How to Design, Analyse and Report Cluster Randomised Trials in Medicine and Health Related Research , 2014 .

[44]  D. Rubin,et al.  Small-sample degrees of freedom with multiple imputation , 1999 .

[45]  Elizabeth L. Turner,et al.  Impact of Intermittent Screening and Treatment for Malaria among School Children in Kenya: A Cluster Randomised Trial , 2014, PLoS medicine.