Short Courses are offered as full- or half-day courses. The extended length of these courses allows attendees to obtain an in-depth understanding of the topic. These courses often integrate seminar lectures covering foundational concepts with hands-on lab sessions that allow users to implement these concepts into practice.
Matthew A. Psioda, Department of Biostatistics, Gillings School of Global Public Health, University of North Carolina at Chapel Hill
Joseph G. Ibrahim, Department of Biostatistics, Gillings School of Global Public Health, University of North Carolina at Chapel Hill
This short course is designed to give biostatisticians and data scientists a comprehensive overview of the use of Bayesian methods for clinical trial design and training on how these methods can be implemented using standard software. Specially, applications of methodology will be demonstrated using R, SAS or both. Part I will give a broad overview of Bayesian sample size determination with a focus on fixed sample size trials either in the phase II or the phase III setting. Focus is paid to four concepts that govern sample size determination: (1) the sampling prior that reflects knowledge about the parameter(s) in the data model, (2) the fitting prior used to analyze data once collected, (3) the criterion used as the basis of sample size determination, and (4) the strategy for monitoring, if the trial will include one or more interim analyses. For (3), a comprehensive review of Bayesian criterion for sample size determination will be given, covering such topics as Bayesian type I error rate control and power, average coverage criterion, average length criterion, and worst outcome criterion. For (4) multiple strategies will be discussed for monitoring accumulating data, including using predictive probability of success and sequential methods. Part II will focus broadly on advanced Bayesian trial designs that incorporate information borrowing. The types of designs considered fall into two broad categories: (1) designs that borrow information through the use of an informative prior specified a priori based on one or more historical datasets, and (2) designs that seek to borrow information across subgroups within a trial. Example designs of type (1) include trials where the goal may be to show that a next-generation medical device (e.g., a coronary stent) is non-inferior or superior to a previous generation of the same device, and designs that seek to extrapolate information on treatment efficacy from adult trials to the pediatric setting. Example designs of type (2) include basket trials where the goal is to make inferences regarding treatment activity for different tumor types in patients whose tumor has a genetic marker targeted by the investigational treatment.
Matthew Psioda is an Assistant Professor of Biostatistics at the University of North Carolina at Chapel Hill (UNC) and Associate Director of Clinical Trials Research at the Collaborative Studies Coordinating Center. He currently serves as co-investigator for the Data Integration, Algorithm Development and Operations Management Center (DAC) within the NIH Back Pain Consortium (BACPAC) as well as Executive Committee Chair for the Consortium. He serves as co-Investigator for the Comprehensive Post-Acute Stroke Services Study (COMPASS Study), co-investigator for the Adolescent Medicine Trials Network (ATN) for HIV/AIDS Interventions, co-investigator for the UNC Laboratory for Innovative Clinical Trials, and as a statistical advisor for the Center for Drug Evaluation and Research at the United States Food and Drug Administration (FDA). He received his PHD in biostatistics from UNC in 2016 and worked in the pharmaceutical product development industry for 7 years prior to obtaining his PHD. His primary methodological research focus is the design of innovative clinical trials using Bayesian methods but he is also interested in methodological research related to Bayesian computation and estimation of heterogeneous treatment effects. He also teaches graduate courses in statistical computing and longitudinal data analysis methods.
Dr. Joseph G. Ibrahim, Alumni Distinguished Professor of Biostatistics at the University of North Carolina at Chapel Hill, is principal investigator of two National Institutes of Health (NIH) grants for developing statistical methodology related to cancer, imaging, and genomics research. Dr. Ibrahim is the Director of the Biostatistics Core at UNC Lineberger Comprehensive Cancer Center. He is the biostatistical core leader of a Specialized Program of Research Excellence in breast cancer from NIH. Dr. Ibrahim's areas of research focus are Bayesian inference, missing data problems, cancer, and genomics. He received his PHD in statistics from the University of Minnesota in 1988. With over 30 years of experience working in cancer clinical trials, Dr. Ibrahim directs the UNC Laboratory for Innovative Clinical Trials (LICT). He is also the Director of Graduate Studies in UNC’s Department o Biostatistics, as well as the Program Director of the cancer genomics training grant in the department. Dr. Ibrahim has published over 350 research papers, most in top statistical journals. He has published graduate-level books on Bayesian survival analysis and Bayesian computation. He teaches courses in Bayesian Statistics, Advanced Statistical Inference, Theory and Applications of Linear and Generalized Linear Models, and Statistical Analysis with Missing Data.
Beth Ann Griffin, RAND Corporation
Daniel F. McCaffrey, ETS
Estimation of causal effects is a primary activity of many studies. Examples include testing whether a substance abuse treatment program is effective, whether an intervention improves the quality of mental health care, or whether new medicines cure a disease. Controlled, random-assignment experiments are the gold standard for estimating such effects. However, experiments are often infeasible, forcing analysts to rely on observational data in which treatment assignments are out of the control of the researchers. This short course will provide an introduction to causal modeling using the potential outcomes framework and the use of propensity scores and weighting (i.e., propensity score or inverse probability of treatment weights) to estimate causal effects from observational data. The goals of the course are to increase attendee’s understanding of: (i) how to define and estimate causal effects using the potential outcomes framework; (ii) how to use propensity scores and inverse probability of treatment weights when estimating causal effects; and (iii) how to assess the validity of key assumptions of the proposed methods. We will also provide attendees with step-by-step instructions for analyses involving binary treatments, >2 treatments, and time-varying treatments. Attendees will gain hands-on experience estimating propensity score weights using boosted models in R, Shiny, SAS and Stata; evaluating the quality of those weights; and using them to estimate intervention effects. Additional topics will include methods for conducting sensitivity analyses for unobserved confounding and estimation of the effects of time-varying treatments. Attendees should be familiar with linear and logistic regression; no knowledge of propensity scores is expected.
Beth Ann Griffin is a senior statistician at the RAND Corporation. Her research has largely focused on causal effects estimation when using observational data. Her substantive research has primarily fallen into three areas: (1) substance use treatment for adolescents, (2) the impact of nongenetic factors on Huntington's disease, and (3) the effects of gun and opioid state policies on outcomes. She codirected the RAND Center for Causal Inference between 2013 and 2018 and is currently codirector of the RAND/USC Opioid Policy Tools and Information Center (OPTIC) whose goal is to foster innovative research, tools, and methods for tackling the opioid epidemic. She has served as the principal investigator on four grants sponsored by the National Institute of Drug Abuse (NIDA), the latest devoted to developing new tools and methods to understand causal mediation and moderation and assess the sensitivity of effect estimates to omitted variables (www.rand.org/statistics/twang ). Griffin's research has appeared in leading journals such as in Journal of the American Statistical Association, Annals of Applied Statistics, Statistics in Medicine, and Journal of Causal Inference. Dr. Griffin also serves as an editor for the Annals of Applied Statistics. She received her Ph.D. in biostatistics from Harvard University.
Daniel F. McCaffrey is the Associate Vice President of Psychometric Analysis and Research in the Research and Measurement Science unit of the Research and Development division at ETS. He has been studying causal modeling for over 20 years with numerous publications on the topic including studies on the use of machine learning in causal modeling and analysis in the presence of measurement error. He is one of the developers of the Toolkit for Waiting Analysis for Nonequivalent Group (TWANG) package in R, Stata, and SAS and continues to develop tools causal modeling. Dan is an experienced instructor on causal modeling having taught over ten short courses and workshops on the topic to diverse audiences. Dan received his Ph.D. in Statistics from North Carolina State University in 1991. He joined ETS in 2013 after over 20 years at RAND. Dan is a Fellow of the American Statistical Association.
Gary Sullivan, Espirer Consulting, LLC
The need for leadership from statisticians and data scientists is greater than ever. Regardless of your sector of employment, data and quantitative approaches are being increasingly leveraged to influence strategic direction, improve the quality & speed of decisions, inform & optimize policy and create competitive business advantages. These activities require more leadership of statisticians to direct the thinking, set the strategy and drive the appropriate methods.
In this course, statisticians and data scientists will learn the skills and concepts to better influence and persuade decision-makers, cross functional colleagues, and team members to act on their ideas. Topics covered include leadership fundamentals, effective communication, building trust, and developing business/organizational acumen with a focus on the role these play in influencing and leading. The course will also feature examples, group discussions, and exercises to gain insights into leadership and to prepare for opportunities to influence. As part of the training, each participant will develop a plan of action to improve their leadership skills and ability.
This course was created specifically for statisticians and data scientists, and explores and teaches leadership from their perspective. The target audience for this course is statisticians and data scientists with at least 3 years of professional experience. The course is designed for any statistician or data scientist whether you are an individual technical contributor or have supervisory responsibilities as the skills required to lead and influence in these roles are very similar.
Dr. Gary Sullivan is the founder of Espirer Consulting, LLC where he provides leadership training, coaching, and advising to statisticians and statistical organizations. He is the co-developer and primary instructor for The Effective Statistician Leadership Program, an on-line leadership program for statisticians and data scientist in the pharmaceutical industry. He is also the primary developer and instructor for leadership training within the ASA, having taught courses on leadership, cultural competence, and executive presence at conferences and chapters. He has provided leadership training to over 500 professionals within the pharmaceutical industry and the ASA.
Dr. Sullivan retired from Eli Lilly and Company in 2017 as the Senior Director for Non-Clinical Statistics. He joined Eli Lilly in 1989 and held various technical and administrative roles over his 28 years there. While at Eli Lilly, he led the development and administration of a leadership program for the statistics function from 2009–2017.
Dr. Sullivan has organized and participated on leadership panels, given numerous leadership presentations, and has authored many articles on leadership development for statisticians. He holds a Bachelor’s degree in Statistics from the University of Pittsburgh, and both a Master’s and Doctorate in Statistics from Iowa State University.
Lu Mao, University of Wisconsin
This course provides an overview of the emergent statistical methodology for the analysis of composite time-to-event outcomes. These outcomes combine death and (possibly recurrent) nonfatal events, such as hospitalization, tumor progression, or infection, and are routinely used as the primary efficacy endpoint in modern phase-III clinical trials. The traditional approach to composite outcomes focuses on time to the first event, whichever type it is, using standard univariate survival analysis techniques. Recent years have seen a surge of more sophisticated and versatile methods, attracting the attention of both statisticians and practitioners. Examples of such methods include the win ratio (Pocock et al., 2012) and its various extensions, the restricted mean time in favor of treatment (a generalized restricted mean survival time), the event (or loss) rate ratio while alive, generalized semiparametric proportional odds regression models, and so on. They improve upon the traditional time-to-first-event analysis in (1) proper prioritization of death over nonfatal events; (2) fuller utilization of multiple/recurrent events; (3) clear and interpretable definition of effect-size estimands; and (4) flexible modeling of different outcome types. In the meantime, a number of user-friendly R-packages that implement the aforementioned methodology have become available. This short course will provide a survey of these methodological developments, along with some practical guidance on using the associated R-packages for real data analysis.
Lu Mao joined the Department of Biostatistics and Medical Informatics at University of Wisconsin (UW)-Madison as an Assistant Professor after obtaining his doctoral degree in Biostatistics from UNC Chapel Hill in 2016. His research interests include survival analysis (particularly composite outcomes), causal inference, semiparametric theory, and clinical trials. He is currently the PI of an NIH R01 grant on statistical methodology for composite time-to-event outcomes in cardiovascular trials and an NSF grant on causal inference in randomized trials with noncompliance. Besides methodological studies, he also collaborates with medical researchers in cardiology, radiology, cancer, and health behavioral interventions, where time-to-event and longitudinal data are routinely collected and analyzed. At UW-Madison, he has been serving as the instructor for a graduate-level survival analysis course since 2017. Off campus, he has taught several short courses on statistical methods for composite outcomes to general audiences, including a recent one at the 2021 ASA Biopharmaceutical Section Regulatory-Industry Statistics Workshop.
Emily Butler, Director of Statistics, ProKidney
Diversity, Equity and Inclusion are vital components of every team, department and organization. We all benefit from a profession and society that recognizes that our differences are our strengths and uplifts our colleagues by acknowledging all components of one’s identity. As leaders and change agents, none of us “know it all”. None of us are immune from making faux pas. The difficult truth is that most of us don’t know what we don’t know. This short course serves as a groundbreaking opportunity to learn about the impact diversity, equity and inclusion has on our profession, how we can be better statisticians and barriers we all face together. The course will not be taught as a lecture; instead, it will be a series of guest speakers and round table discussions.
Emily Butler is the Director of Biostatistics at ProKidney, a cell therapy start up. She is responsible for all analytical projects across the company, including 5 clinical trials and investor relations. While she now works in nephrology, she previously supported early oncology cell therapy at a large pharmaceutical company. As a statistician, her true passion is around engaging with her clinical colleagues to ensure the statistical validity of all trial designs, interpretations and publications. Emily graduated from the University of North Carolina at Chapel Hill in 2016 with a PhD in biostatistics and Carnegie Mellon University in 2011 with a bachelors in statistics.
Mine Çetinkaya-Rundel, Duke University and RStudio
This workshop is all about the art and science of visualizing data with R. Learn about the what (types of visualizations, tools to produce them), the how (start with a design, pre-process the data, map it to graphical attributes, make strategic decisions about visual encoding, post-process for readability and visual appeal), and the why (the theory behind the grammar of graphics). Do it all in R, reproducibly, and using a variety of modern data visualization packages, primarily ggplot2.
Mine Çetinkaya-Rundel is Professor of the Practice at the Department of Statistical Science at Duke University and Data Scientist and Professional Educator at RStudio. Mine’s work focuses on innovation in statistics and data science pedagogy, with an emphasis on computing, reproducible research, student-centered learning, and open-source education as well as pedagogical approaches for enhancing retention of women and under-represented minorities in STEM. Mine works on integrating computation into the undergraduate statistics curriculum, using reproducible research methodologies and analysis of real and complex datasets. She also organizes ASA DataFest, an annual two-day competition in which teams of undergraduate students work to reveal insights into a rich and complex dataset. Mine has been working on the OpenIntro project since its founding and as part of this project she co-authored four open-source introductory statistics textbooks, including the newly published Introduction to Modern Statistics. She is also the creator and maintainer of datasciencebox.org and she teaches the popular Statistics with R MOOC on Coursera.
Eric Polley, The University of Chicago, Department of Public Health Sciences
This course will offer an overview of machine learning methods, with an emphasis on how the methods can be used to develop and evaluate predictive models. Commonly used methods like random forests, generalized additive models, and deep neural networks will be discussed along with code examples for implementation. The different methods will be compared across a variety of scenarios with advice on selection of an algorithm and how to tune the associated hyperparameters. Along with an overview of different methods, topics will include a review of loss functions and optimization algorithm, how to evaluate a machine learning model for fairness, model interpretability, and how to monitor a machine learning model in practice.
Dr. Eric Polley is an Associate Professor in the Department of Public Health Sciences at the University of Chicago and the faculty director of the data science concentration in public health program. He has a PhD in Biostatistics from the University of California, Berkeley and a post-doctoral fellowship in Computational Biology at the National Cancer Institute.