D‐optimal designs for multiarm trials with dropouts

Multiarm trials with follow‐up on participants are commonly implemented to assess treatment effects on a population over the course of the studies. Dropout is an unavoidable issue especially when the duration of the multiarm study is long. Its impact is often ignored at the design stage, which may lead to less accurate statistical conclusions. We develop an optimal design framework for trials with repeated measurements, which takes potential dropouts into account, and we provide designs for linear mixed models where the presence of dropouts is noninformative and dependent on design variables. Our framework is illustrated through redesigning a clinical trial on Alzheimer's disease, whereby the benefits of our designs compared with standard designs are demonstrated through simulations.

[1]  M. Berger,et al.  Robust designs for linear mixed effects models , 2004 .

[2]  Joseph W Hogan,et al.  Handling drop‐out in longitudinal studies , 2004, Statistics in medicine.

[3]  Martin Knapp,et al.  Donepezil and memantine for moderate-to-severe Alzheimer's disease. , 2012, The New England journal of medicine.

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

[5]  Agnes M. Herzberg,et al.  Some Considerations in the Optimal Design of Experiments in Non‐Optimal Situations , 1976 .

[6]  Joseph G. Ibrahim,et al.  Missing covariates in generalized linear models when the missing data mechanism is non‐ignorable , 1999 .

[7]  P. Hackl.  Optimal Design for Experiments with Potentially Failing Trials , 1995 .

[8]  T. Schmelter.  Considerations on group-wise identical designs for linear mixed models , 2007 .

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

[10]  S. A. Ortega-Azurduy,et al.  The effect of dropout on the efficiency of D‐optimal designs of linear mixed models , 2008, Statistics in medicine.

[11]  M. Kenward,et al.  Informative Drop‐Out in Longitudinal Data Analysis , 1994 .

[12]  Jerome P. Reiter,et al.  Estimating propensity scores with missing covariate data using general location mixture models. , 2011, Statistics in medicine.

[13]  R. Kolamunnage-Dona,et al.  Statistical primer: performing repeated-measures analysis. , 2018, Interactive cardiovascular and thoracic surgery.

[14]  Subir Ghosh,et al.  ON ROBUSTNESS OF DESIGNS AGAINST INCOMPLETE DATA , 2016 .

[15]  Martijn P F Berger,et al.  Maximin D‐Optimal Designs for Longitudinal Mixed Effects Models , 2002, Biometrics.

[16]  Anthony C. Atkinson,et al.  Optimum Experimental Designs, with SAS , 2007 .

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

[18]  K. Chaloner,et al.  Bayesian Experimental Design: A Review , 1995 .

[19]  Kim May Lee,et al.  Optimal design when outcome values are not missing at random , 2018 .

[20]  Sally Galbraith,et al.  Guidelines for the design of clinical trials with longitudinal outcomes. , 2002, Controlled clinical trials.

[21]  Stefanie Biedermann,et al.  Optimal design for experiments with possibly incomplete observations , 2018 .

[22]  A. Hedayat,et al.  Resistant and Susceptible BIB Designs , 1974 .

[23]  Roderick J. A. Little,et al.  Modeling the Drop-Out Mechanism in Repeated-Measures Studies , 1995 .

[24]  S. Gilmour,et al.  Robustness of subset response surface designs to missing observations , 2010 .

[25]  M. Kenward,et al.  Informative dropout in longitudinal data analysis (with discussion) , 1994 .

[26]  Geert Molenberghs,et al.  Analyzing incomplete longitudinal clinical trial data. , 2004, Biostatistics.