TITLE: Randomization and Inference in Clinical Trials
SPEAKERS: Diane Uschner, George Washington University
MODERATOR: Ivan S. F. Chan
Randomization in clinical trials is a design technique that tends to balance treatment groups with respect known and unknown covariates. It is therefore commonly regarded as the key component of clinical trials that provides comparability of treatment groups. Despite the favorable properties of randomization, a randomized clinical trial (RCT) can still suffer from a lack of comparability among the treatment groups. Bias may arise for example from unobserved time trends, or from predictability of treatment assignments. Comparability is of special importance in small population groups (SPG) that arise in pediatric trials and clinical trials of rare diseases. Trials in SPG are frequently conducted as multi-arm trials to achieve greater efficiency through a shared control group. An important feature of randomization that is often overlooked is that it provides a valid basis for inference. This course will be divided into two parts. First, we will cover the basics of randomization in clinical trials; goals and ethics; restricted and adaptive randomization; quantification of bias; and regulatory guidelines. The methods will be illustrated using available software packages and shiny apps. The second part of the workshop will cover randomization as a basis for inference. This section will cover statistical and practical properties including the efficient implementation and extensions to causal inference.
- Wang, W. F. Rosenberger, D. Uschner. (2019) Randomization Tests for Multi- armed Randomized Clinical Trials. Statistics in Medicine (accepted).
- Wang, W. F. Rosenberger, D. Uschner. (2019). Randomization-based inference and the choice of randomization procedures. Statistical papers. 60(2)
- F. Rosenberger, D. Uschner, Y. Wang (2018). The fifteenth Armitage Lecture: Randomization: The forgotten component of the randomized clinical trial. Statistics in Medicine. 38(1).
- Schindler, D. Uschner, R.-D. Hilgers and N. Heussen (2018). randomizeR: Randomization for Clinical Trials. R package version 1.4.
- Uschner, D. Schindler, R.-D. Hilgers, N. Heussen (2018). randomizeR: An R Package for the Assessment and Implementation of Randomization in Clinical Trials. Journal of Statistical Software. 85(8).
- Uschner, N. Heussen, R.-D. Hilgers (2018). The Impact of Selection Bias in Randomized Multi-Arm Parallel Group Clinical Trials. PLOS ONE 13(1): e0192065.
- -D. Hilgers, D. Uschner, W. F. Rosenberger and N. Heussen (2017). A general framework to assess randomization procedures in the presence of selection and chronological bias. BMC Medical Research Methodology.17:159
Diane Uschner, PhD., is an Assistant Research Professor in the Department of Biostatistics and Bioinformatics at the George Washington University School of Public Health. She joined the GW Biostatistics Center in April 2018 and currently serves as Co-Investigator of the Coordinating Center for the TODAY2 long-term follow-up study of the Treatment Options for T2D in Adolescents and Youth (TODAY) study. She served as a statistician for the C-Peptide Ancillary Study of the DCCT/EDIC (Diabetes Control and Complications Trial/Epidemiology of Interventions and Complications) study. Dr. Uschner has recently developed an interest in antibacterial resistance and infectious diseases, and is serving as a biostatistician in the Antibiotic Resistance Leadership Group (ARLG). She is the PI of the Data Coordinating Center of the North Carolina COVID-19 Community Research Partnership. Prior to joining GW, Dr. Uschner received her Ph.D. in Biostatistics from the RWTH Aachen University, Germany in 2018. Her PhD thesis focused on the choice of randomization procedures to mitigate bias in clinical trials and was funded by the EU FP7-project “Integrated Design and Analysis of Small Population group Trials” (IDeAl). Her research interests are focused on the design of clinical trials, randomization, control of bias, as well as survival analysis, penalized regression, and statistical learning. Dr. Uschner regularly serves as a reviewer for scientific journals, NIH grant reviews, and meeting abstracts.