The impact of selection bias on test decisions in randomized clinical trials

Selection bias affects the evaluation of clinical trials, for example, by elevating type I error rate. We investigated the effect of selection bias on type I error rate considering permuted block randomization. We also considered stratified randomization in general and for the special case of multicenter clinical trials, where we incorporated preferences of the investigator in our approach. Finally, the effect of underrunning is modeled, that is, where the randomization list exceeds the actual number of patients taking part in the trial. For all situations, we illustrate and discuss the impact of selection bias on type I error rate.

[1]  V. Berger.  Selection Bias and Covariate Imbalances in Randomized Clinical Trials: Berger/Selection Bias and Covariate Imbalances in Randomized Clinical Trials , 2005 .

[2]  C. Viscoli,et al.  Stratified randomization for clinical trials. , 1999, Journal of clinical epidemiology.

[3]  J. Matthews,et al.  Randomization in Clinical Trials: Theory and Practice; , 2003 .

[4]  J M Lachin,et al.  Statistical properties of randomization in clinical trials. , 1988, Controlled clinical trials.

[5]  J. Lewis,et al.  Statistical principles for clinical trials (ICH E9): an introductory note on an international guideline. , 1999, Statistics in medicine.

[6]  V W Berger,et al.  Detecting selection bias in randomized clinical trials. , 1999, Controlled clinical trials.

[7]  Varying the block size does not conceal the allocation. , 2006, Journal of critical care.

[8]  H Merabet,et al.  The design and analysis of sequential clinical trials , 2013 .

[9]  Kenneth F Schulz,et al.  For Personal Use. Only Reproduce with Permission from the Lancet Publishing Group. Trialists Rely on the Security of Unpredictability. in the Forcing Cosmetic Credibility Unequal Group Sizes in Randomised Trials: Guarding against Guessing , 2022 .

[10]  Vance W Berger,et al.  Quantifying the Magnitude of Baseline Covariate Imbalances Resulting from Selection Bias in Randomized Clinical Trials , 2005, Biometrical journal. Biometrische Zeitschrift.

[11]  Conditional Markov chain design for accrual clinical trials , 1983 .

[12]  William C. Grant,et al.  Sensitivity designs for preventing bias replication in randomized clinical trials , 2010, Statistical methods in medical research.

[13]  Anastasia Ivanova,et al.  Minimizing predictability while retaining balance through the use of less restrictive randomization procedures , 2003, Statistics in medicine.

[14]  V. Gebski,et al.  Randomisation in clinical trials , 2002, The Medical journal of Australia.

[15]  Misguided Precedent is not a Reason to Use Permuted Blocks , 2006 .

[16]  Do not test for baseline imbalances unless they are known to be present? , 2009, Quality of Life Research.

[17]  The effect of estimation and biasing strategies on selection bias in clinical trials with permuted blocks , 1994 .

[18]  Fiona Simpson,et al.  Randomization and allocation concealment: a practical guide for researchers. , 2005, Journal of critical care.

[19]  Douglas G Altman,et al.  Treatment allocation in controlled trials: why randomise? , 1999, BMJ.

[20]  Douglas G Altman,et al.  How to randomise , 1999, BMJ.

[21]  Randomization, permuted blocks, masking, allocation concealment, and selection bias in the Tobacco Quit Line Study. , 2010, Contemporary clinical trials.