ENAR 2025 Educational Program | SHORT COURSES

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.

 

Sunday, March 23 | 8:00 am – 5:00 pm
SC1 | Introduction to Bayesian Meta-Analysis and Network Meta-Analysis

Instructors:
Ming-Hui Chen, University of Connecticut
Joseph G. Ibrahim, University of North Carolina at Chapel Hill

Course Description:

This full-day short course is designed to give biostatisticians and data scientists a comprehensive overview of Bayesian meta-analysis (MA) and network meta-analysis (NMA) to assess the efficacy of treatments with focus on cardiovascular disease, Crohn's disease, and other diseases. Meta-analysis (MA) combines evidence from relevant studies using appropriate statistical methods to allow inference on the population of interest. Network Meta-Analysis (NMA) is a meta-analysis in which multiple treatments are being compared using both direct comparisons of interventions within randomized controlled trials and indirect comparisons across trials based on a common comparator. This course covers six major aspects of Bayesian MA and NMA: (i) flexible modeling of univariate and multivariate continuous and ordinal responses; (ii) heavy-tailed multivariate random effects and covariance modeling in NMA; (iii) model assessments and model diagnostics; (iv) efficient computational algorithms via collapsed Gibbs sampling; (v) treatment ranking; and (vi) quantifying evidence of homogeneity and consistency. The course consists 4 parts: (1) Bayesian MA/NMA for univariate and multivariate aggregate data (AD) and individual level patient data (IPD) with applications to evaluation of cholesterol lowering Drugs; (2) Bayesian NMA for aggregate ordinal outcomes to assess the efficacy of treatments for Crohn's disease; (3) Assessing homogeneity and consistency in NMA; and (4) Demonstration of R package metapack for carrying out Bayesian MA and NMA.

The intended audience for this course include biostatisticians and data scientists who hold at least a masters-level degree in biostatistics or a related field. The primary learning objectives for this course are to (1) provide practitioners with a sound understanding of core, cross-cutting concepts for Bayesian MA and NMA, (2) help practitioners understand the benefits and challenges of applying Bayesian methods in fitting and analyzing univariate and multivariate AD and IPD using realistic case studies, and (3) teach practitioners about R software that can be used to implement and evaluate Bayesian MA and NMA in practice. By providing applied practitioners with a sound understanding of core concepts related to MA and NMA, they will better equipped to have discussions with internal and external colleagues regarding the most effective approaches to synthesize evidence from relevant studies.

Drs. Chen and Ibrahim will co-teach this short course. Both have significant expertise in both frequentist and Bayesian approaches in MA and NMA.

Statistical/Programming Knowledge Required:
Introductory Bayesian statistics, regression models, R

Instructor Biographies:

They co-authored two advanced graduate-level books on Bayesian survival analysis and Monte Carlo methods in Bayesian computation. They have been collaborating together for over 28 years and have written more than 100 research papers together.

 

Sunday, March 23 | 8:00 am – 5:00 pm
SC2 | Semiparametric Regression Analysis of Interval-Censored Data

Instructors:
Danyu Lin, The University of North Carolina at Chapel Hill
Donglin Zeng, University of Michigan
Yulia Marchenko, StataCorp

Course Description:

: In clinical and epidemiological studies, the onset of an asymptomatic disease (e.g., diabetes, hypertension, chronic obstructive pulmonary disease, HIV infection, SARS-CoV-2 infection, cancer, or dementia) cannot be observed directly but rather is known to occur sometime between two consecutive clinical examinations. The two examinations bookend a time interval, such that the event time is “interval-censored.” It is highly challenging to analyze interval-censored data because none of the event times is exactly known; therefore, investigators have resorted to statistical methods that are unreliable or even invalid. Recent theoretical and numerical advances, as well as software implementation in R, SAS, and Stata, have made semiparametric regression analysis of interval-censored data a practical reality. The goal of this short course is to present these recent developments to a broad audience in an accessible manner. Specifically, we formulate the effects of potentially time-dependent covariates on the interval-censored event time through semiparametric regression models, such as the Cox proportional hazards model. We perform nonparametric maximum likelihood estimation with an arbitrary number of monitoring times for each study subject. We describe an EM algorithm that involves remarkably simple calculations and converges stably for any dataset, even in the presence of time-dependent covariates. The resulting estimators for the regression parameters are consistent, asymptotically normal, and asymptotically efficient with an easily estimated covariance matrix. In addition, we will describe extensions to competing risks, multivariate failure time data, panel count data, as well as to big data involving potentially millions of subjects. We will also describe graphical and numerical techniques for checking model adequacy and improving goodness of fit. We will illustrate the numerical and inferential procedures through applications to a variety of medical studies, together with software demonstration. This short course targets both theoretical and applied statisticians. Participants will learn cutting-edge research in survival analysis, as well as practical skills to analyze interval-censored data, which include right-censored data as a special case. Prior knowledge in survival analysis is not required. We will provide an overview of basic survival analysis methods at the beginning of this short course. Prior experience with R, SAS or Stata is not required either.

Statistical/Programming Knowledge Required:
Statistical Inference; Linear Regression

Instructor Biographies:

 

Sunday, March 23 | 8:00 am – 5:00 pm
SC3 | Statistical Inference in Large Language Models for Biomedical Applications

Instructors:
Weijie Su, University of Pennsylvania
Qi Long, University of Pennsylvania
Inyoung Choi, University of Pennsylvania

Course Description:

Large Language Models (LLMs) have emerged as revolutionary AI tools with immense potential to reshape biomedical research and healthcare. However, when harnessing their potential for medical decision-making and research, it becomes essential to understand and mitigate the risks associated with their outputs. Evaluating and ensuring reliability in LLMs present both challenges and intriguing opportunities for today's statisticians and biomedical researchers. The aim of this one-day short course is to equip participants with the skills to integrate inferential concepts into the applications and advancement of LLMs in biomedicine. Course topics include: a brief introduction to the fundamentals of LLMs tailored for biomedical contexts, a primer on statistical inference techniques for text data using LLMs with a focus on biomedical literature and clinical notes, in-depth exploration of methods to address bias, enhance calibration, and reduce hallucinations in LLM alignment for medical applications, techniques for watermarking LLMs to ensure the authenticity and integrity of AI-generated medical content, and case studies applying these methods to real-world biomedical datasets and use cases. By the end of the course, attendees will possess the skills needed to develop and apply more reliable, trustworthy, and ethically aligned LLMs in biomedical research and healthcare settings. While this course promises a deep and enriching dive into the confluence of statistics, advanced AI, and biomedicine, no prior knowledge of LLMs is required.

Statistical/Programming Knowledge Required:
None

Instructor Biographies:

 

Sunday, March 23 | 8:00 am – 12:00 pm
SC4 | Improving Precision and Power in Randomized Trials by Leveraging Baseline Variables

Instructors:
Josh Betz, Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health
Kelly Van Lancker, Department of Applied Mathematics, Computer Science and Statistics, Ghent University
Michael Rosenblum, Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health

Course Description:

In May 2023, the U.S. Food and Drug Administration (FDA) released guidance for industry on “Adjustment for Covariates in Randomized Clinical Trials for Drugs and Biological Products”. Covariate adjustment is a statistical analysis method for improving precision and power in clinical trials by adjusting for pre-specified, prognostic baseline variables. Here, the term “covariates” refers to baseline variables, that is, variables that are measured before randomization such as age, gender, BMI, comorbidities. The resulting sample size reductions can lead to substantial cost savings and more ethical trials since they avoid exposing more participants than necessary to experimental treatments. Though covariate adjustment is recommended by the FDA and the European Medicines Agency (EMA), many trials do not exploit the available information in baseline variables or only make use of the baseline measurement of the outcome.

In Part 1, we introduce the concept of covariate adjustment. We explain what covariate adjustment is, how it works, when it may be useful to apply, and how to implement it in a preplanned way that is robust to model misspecification.

In Part 2, we present statistical methods that enable investigators to easily combine covariate adjustment with trial designs that allow for interim stopping for efficacy and futility, including information monitoring and group sequential designs. The result will be faster, more efficient trials for many disease areas, without sacrificing validity or power. This approach can lead to faster trials even when the experimental treatment is ineffective; this may be more ethical in settings where it is desirable to stop as early as possible to avoid unnecessary exposure to side effects.

In Part 3, we demonstrate how to implement covariate adjustment across the life cycle of a study using data and code, including planning new studies, monitoring ongoing studies, and performing pre-specified analyses.

Statistical/Programming Knowledge Required:

Instructor Biographies:

 

Sunday, March 23 | 1:00 pm – 5:00 pm
SC5 | Statistical Methods for Time-to-Event Data from Multiple Sources: A Causal Inference Perspective

Instructors:
Xiaofei Wang, Duke University School of Medicine
Shu Yang, North Carolina State University

Course Description:

The short course will review important statistical methods for survival data arising from multiple data sources, including randomized clinical trials and observational studies. It consists of two parts, all of which will be discussed in a unified causal inference framework. In each part, we will review the theoretical background and supplement it with data examples. We will emphasize the application of these methods and their implementation in freely available statistical software.

In Part 1, we will review key issues and methods in designing randomized clinical trials (RCTs). We will introduce the statistical framework of causal inference, and then shift the focus to the causal inference methods for survival data. We will review various methods that allow valid visualization and testing for confounder-adjusted survival curves and RMST differences, including G-Formula, Inverse Probability of Treatment Weighting, Propensity Score Matching, calibration weighting, Augmented Inverse Probability of Treatment Weighting. In Part 2, we will cover the objectives and methods that allow integrative data analyses from RCTs and observational studies. These methods exploit the complementing features of RCTs and observational studies to estimate the average treatment effect (ATE), heterogeneity of treatment effect (HTE), and individualized treatment rules (ITRs) over a target population. Firstly, we will review existing statistical methods for generalizing RCT findings to a target population, leveraging the representativeness of the observational studies. Due to population heterogeneity, the ATE and ITRs estimated from the RCTs lack external validity/generalizability to a target population. We will review the statistical methods for conducting generalizable RCT analysis for the targeted ATE and ITRs, including inverse probability sampling weighting, calibration weighting, outcome regression, and doubly robust estimators. Secondly, we will review existing statistical methods for integrating RCTs and observational studies for robust and efficient estimation of the HTE. RCTs have been regarded as the gold standard for treatment effect evaluation due to randomization of treatment, which may be Underpowered to detect HTEs due to practical limitations. On the other hand, large observational studies contain rich information on how patients respond to treatment, which may be confounded. We will review statistical methods for robust and efficient estimation of the HTE leveraging the treatment randomization in RCTs and rich information in observational studies, including test-based integrative analysis, selective borrowing, and confounding function modeling.

Statistical/Programming Knowledge Required:
Attendees are expected to have some familiarity with survival analysis and some concepts of causal inference, but a deep understanding of the general principles of causal inference is not required.

Instructor Biographies:

 

Sunday, March 24 | 1:00 pm – 5:00 pm
SC6 | Bayesian Borrowing Techniques for Rare Disease Clinical Research

Instructors:
Joseph S. Koopmeiners, University of Minnesota
Steffen Ventz, University of Minnesota

Course Description:

Randomized clinical trials (RCTs) are the gold standard for estimating the effect of a treatment on an outcome. However, RCTs are also resource-intensive and require large samples to estimate moderate or small effect sizes. The resource-intensive nature of RCTs poses particular challenges in the setting of rare diseases where limitations on the number of potential trial participants limit the overall sample size of RCTs. Given these limitations, the design of RCTs in the context of rare diseases places a premium on efficiency and leveraging all available information to evaluate the effect of a treatment on an outcome. One approach to improve the efficiency of RCTs is to leverage external information, in the form of supplemental trial data or real-world data sources (EHR, etc.), through dynamic borrowing. Recent advances in Bayesian methods for dynamic borrowing provide a powerful set of statistical tools to improve the efficiency of RCTs by leveraging data external to the trial. This short course will provide an introduction to Bayesian methods for dynamic borrowing in the setting of rare disease clinical research. Specific topics to be covered include the motivation for leveraging external data in RCTs, an overview of Bayesian methods for dynamic borrowing, including recent advancements in the use of real-world data to augment RCTs, computational tools for implementing these methods, and a general discussion of the strengths and weaknesses of implementing these methods in practice. Throughout the course, methods will be illustrated via case studies from rare disease clinical research.

Statistical/Programming Knowledge Required:
Students should have a basic understanding of Bayesian statistics and some familiarity with clinical trials.

Instructor Biographies:

 

Monday, March 24 | 8:00 am – 12:00 pm
SC7 | Futility Monitoring in Clinical Trials

Instructors:
Ana M. Ortega-Villa, National Institute of Allergy and Infectious Diseases
Michael Proschan, National Institute of Allergy and Infectious Diseases

Course Description:

At the beginning of a phase III clinical trial, there is great optimism. After all, the phase II trial results were encouraging. Then early data from the phase III trial trend the wrong way, but there is still the opportunity for the trend to reverse and become statistically significant at the end. At what point does optimism become denial of reality? How do we decide when a clinical trial is futile? What does futility even mean? This short course reviews different concepts of, and tools for evaluating, futility.

In this short course we will differentiate between operational and treatment futility, and learn tools to answer two important questions, 1) will the final result be null? And 2) will a null result be meaningful? Specific topics include conditional and predictive power, reverse conditional power, predicted intervals, predicted interval plots, revised unconditional power, beta spending functions, and an introduction to Bayesian futility monitoring.

During this short course, participants will get theoretical and hands-on practical knowledge to conduct futility analyses, interpret results, and make informed decisions. We will review case studies to illustrate the application of futility monitoring techniques in real world clinical trials. Prior to attending the short course, participants should have a general understanding of design and conduct of clinical trials, basic statistical concepts and distributions, and basic knowledge of R. Bring your laptop (with R and RStudio already installed) and join us to learn, discuss, and implement futility monitoring techniques. We look forward to your participation!

Statistical/Programming Knowledge Required:
~Basic knowledge of the design and conduct of clinical trials. ~Understanding of basic statistical concepts and distributions. ~Basic knowledge of R.

Instructor Biographies: