Edited by Feifei Wei, Biometrics Section Publications Officer

The Biometrics Section would like to introduce you to the current members of the section’s executive committee and offer a well-deserved thank you to the outgoing officers.

The section chair for 2014 is Mike Daniels, professor in the Section of Integrative Biology and Division of Statistics and Scientific Computation at The University of Texas at Austin. His research interests include Bayesian methods for incomplete longitudinal data and causal inference and priors and estimators for dependence. He earned his ScD in biostatistics from Harvard University in 1995. He currently collaborates on projects involving new interventions for weight management and imaging to monitor progression of Duchenne’s muscular dystrophy.

Our chair-elect for 2013 is Diana Miglioretti, dean’s professor of biostatistics in the department of public health sciences at the University of California Davis School of Medicine. She earned her PhD in biostatistics from the Johns Hopkins Bloomberg School of Public Health in 2000. Her current research interests include clustered and longitudinal data analysis, screening and diagnostic test evaluation, and risk prediction modeling.

Yu Shen continues as our secretary/treasurer. Yu is a professor of biostatistics in the department of biostatistics at The University of Texas MD Anderson Cancer Center and an adjunct professor at Rice University. She earned her PhD in biostatistics from the University of Washington. Her areas of research interest include survival analysis with biased sampling, design and analysis of cancer clinical trials and observational studies, primary and secondary cancer prevention, and cost-effectiveness analysis of cancer screening strategies.

Jonathan Schildcrout is the section program chair to the 2014 JSM Program Committee. Jonathan is an associate professor in the biostatistics department at the Vanderbilt University School of Medicine. He earned his PhD in biostatistics from the University of Washington in 2004. His methodological interests relate to longitudinal data analysis and efficient study designs with data that are measured repeatedly over time.

Jason Roy is the section representative to the 2014 ENAR Meeting Program. He is associate professor in the department of biostatistics and epidemiology at the University of Pennsylvania. He earned his doctorate from the University of Michigan in 2000. His methodological research interests include causal inference, missing data, and multiple outcome models.

Rebecca Hubbard is the section program chair to the 2015 JSM Program Committee. She is an associate investigator at the Group Health Research Institute and an affiliate assistant professor at the University of Washington. She earned her PhD in biostatistics from the University of Washington in 2007. Rebecca’s research interests include statistical methods for longitudinal observational data, including multi-state models, with a focus on methods for data from electronic health records.

Joe Hogan is our 2014–2016 representative to the Council of Sections. He is professor of biostatistics at Brown University School of Public Health. His current research interests are in causal inference for observational studies and methods for analyzing data from electronic health records.

LiHong Qi is the section program chair for the 2015 ENAR Meeting Program. She is an associate professor of biostatistics in the department of public health sciences at the University of California Davis School of Medicine. She earned her PhD in biostatistics from the University of Washington in 2003 and had her postdoctoral training at the Fred Hutchinson Cancer Research Center. Her research interests include survival analysis, missing data, and identifying genetic and environmental factors for complex diseases/disorders including cancer, autism, and birth defects.

Scarlett L. Bellamy continues as our 2012–2014 representative to the Council of Sections. She has been a faculty member at the University of Pennsylvania Department of Biostatistics and Epidemiology since completing her doctoral studies at Harvard in 2001. Her research interests are focused on the methodological issues related to the design and analysis of cluster-randomized trials. She is particularly interested in applying this methodology to community-based research projects and projects that address health disparities for a variety of clinical and behavioral health outcomes.

Limin Clegg continues as our 2013–2015 representative to the Council of Sections. She is a senior-level federal government employee, serving as the director for the Division of Biostatistics and Program Evaluations at the Office of Inspector General in the U.S. Department of Veterans Affairs. She earned her doctorate in biostatistics from The University of North Carolina at Chapel Hill. Her research interests are in survival analysis, statistical methods for epidemiology and public health, and building population-based administrative data files across federal government health care systems to evaluate health care issues across the continuum of care systems.

Donglin Zeng continues as the continuing education chair. He is a professor in the department of biostatistics and the co-director of the Carolina Survey Research Lab at The University of North Carolina. He earned his PhD in statistics from the University of Michigan. His research interests include semiparametric models, high-dimensional data analysis, personalized medicine, survival analysis, clinical trials, and survey sampling.

Roslyn Stone is our co-chair of the Strategic Initiatives Committee. She is a professor in the department of biostatistics at the University of Pittsburgh Graduate School of Public Health. She earned her doctorate in biomathematics from the University of Washington. Her research interests are in generalized linear models, survival analysis, multi-level models, statistical methods for occupational and environmental epidemiology, guideline implementation, and cluster-randomized studies.

Page Moore is our co-chair of the Strategic Initiatives Committee. She earned her doctorate in statistics from Baylor University in 2006 and is now an associate professor in the department of biostatistics at the University of Arkansas for Medical Sciences. Her main research interests are in multiple imputation techniques, longitudinal data analysis, computational statistics, and clinical trial design.

Feifei Wei continues as the section’s publication officer through 2014. She is an associate professor in the department of biostatistics in the college of public health of the University of Arkansas for Medical Sciences. She earned her doctorate in statistics from The Ohio State University. Her research interests focus on using population-based health care data sets to identify disparities in immunization and predict heath-related events occurring in community settings.

Gerald Beck is our webmaster. He is a staff member in the department of quantitative health sciences at the Cleveland Clinic Foundation. His primary interest is in the design, conduct, and analysis of clinical trials. He serves as principal investigator of data coordinating centers for multi-center clinical studies supported by the National Institutes of Health, including the Frequent Hemodialysis Network Trials and the Hemodialysis Fistula Maturation Study.

Helping Out at JSM

Want to get more involved in JSM? Consider volunteering to chair a session. Chairing a session is an important responsibility and a great way to meet your colleagues. If you are interested, contact Schildcrout at jonathan.schildcrout@vanderbilt.edu.

Strategic Initiatives

The following three proposals have been funded as part of the section’s Strategic Initiative, “Developing the Next Generation of Biostatisticians”:

• TSHS Resources Portal Taxonomy, Dennis Pearl, The Ohio State University
• Biostatistics: High-Impact Discipline and Growing Career Field, Jane Monaco, The University of North Carolina at Chapel Hill
• TSHS Resources Portal Best Practices, Carol Bigelow, University of Massachusetts-Amherst

Continuing Education

The following continuing education proposals sponsored by our section were selected for JSM 2014:

Cure Models and Their Applications in Biomedical Research (1/2-day) Jeremy Taylor, University of Michigan • Yingwei Peng, Queen’s University

Cure models refer to a class of models for survival data with a cured fraction. The standard survival models often assume subjects in a study will experience the event of interest with sufficient follow-up. However, this assumption may not be appropriate in situations such as cancer studies where patients may be cured and will not experience relapse however long the follow-up. The last 15 years witnessed a rapid growth in extending survival models to accommodate potential cured subjects. New statistical methodologies were developed to extend the existing survival models, and the newly proposed cure models greatly expand the applicability of cure models to various types of survival data with a cured fraction and provide appealing ways to interpret the results of analysis, compared to standard survival analysis models. This course will cover the mixture and bounded cumulative hazard formulation of cure models, estimation methods, identifiability issues, and software and consider extensions to clustered data, population studies, and joint longitudinal and survival data. The instructors will introduce data sets from clinical studies, present necessary details of the cure models, and demonstrate software.

Adaptive Methods in Action: How to Improve Pharmaceutical Drug Development (1-day) Guosheng Yin, Hong Kong University • Byron Jone, Novartis Pharma • Frank Bretz, Novartis Pharma

Clinical trials play a critical role in pharmaceutical drug development. New trial designs often depend on historical data, which may not be accurate for the current study due to changes in study populations, patient heterogeneity, or different medical facilities. As a result, the original plan and study design may need to be adjusted, or even altered, to accommodate new findings and unexpected interim results. The goal of using adaptive methods in clinical trials is to enhance the flexibility of trial conduct and maintain the integrity of trial findings. Through carefully thought-out and planned adaptation, we can pinpoint the right dose faster, treat patients more effectively, identify treatment effects more efficiently, and expedite the drug-development process. From perspectives of practicality, this one-day short course will introduce various adaptive methods for phase I to phase III clinical trials. Accordingly, different types of adaptive designs will be introduced and illustrated with case studies. This includes dose escalation/de-escalation and dose insertion based on observed data; adaptive dose-finding studies using optimal designs to allocate new cohorts of patients based on the accumulated evidence; population enrichment designs; early stopping for toxicity, futility, or efficacy using group-sequential designs; blinded and unblinded sample size re-estimation; and adaptive for confirmatory trials with treatment or population selection at interim.

Analysis of Genome-Wide Sequencing Association Studies (1-day) Xihong Lin, Harvard University • Mike Wu, The University of North Carolina

This short course will teach the current statistical methodology for designing and analyzing sequencing association studies to identify the genetic basis of common complex diseases. Rapid advances in next-generation sequencing technologies offer exciting opportunities to gain a better understanding of biological processes underlying complex disease and can lead to new approaches for prevention and treatment. Recently, investigators have exploited these advances and are conducting large-scale sequencing association studies, such as the whole exome sequencing studies, to identify new genetic variants that play important roles in complex diseases. Such data are becoming increasingly common in the hopes that results will not only better our understanding of disease etiology, but also improve treatment response. However, due to the massive number of variants and the rareness of many of these variants across the genome, combined with sequencing costs and the complexity of diseases, analysis of such studies remains challenging. To address the critical statistical gaps, significant effort has led to development of new efficient methods for designing and analyzing emerging sequencing studies. This short course provides an overview of statistical methods for analysis of genome-wide sequencing association studies. Topics include pipelines for low-level processing of whole exome sequencing data, QC methods, review of the 1000 Genome Project and the Whole Exome Sequencing Project, imputation methods for sequencing data, statistical methods for detecting rare variant effects, meta-analysis, interaction testing, and designs for whole genome-wide (exome) sequencing studies. Data examples will be provided and software will be discussed.

Quantile Regression (1-day) Roger Koenker, University of Illinois, Urbana-Champaign • Huixia Judy Wang, North Carolina State University

Quantile regression is a valuable alternative to classical least squares regression. Instead of modeling the conditional functions, quantile regression offers a variety of methods for estimating conditional quantile functions. In applications such as birth weight studies, survival analysis, climate studies and so on, the covariates may have different impacts on different tails of the response distribution. Quantile regression enables the researcher to explore more thoroughly heterogeneous covariate effects and study the tails of scientific interest. The course will offer a comprehensive introduction to quantile regression methods and briefly survey recent developments. The first part of the course will introduce basics of quantile regression, including its features, comparison with the least squares regression, estimation and computation, statistical properties, and inferential procedures. Methods will be demonstrated by using examples from birth weight, climate, and survival studies. The second part of the course will discuss more advanced topics, including nonparametric quantile regression, quantile regression for longitudinal and time series data, censored quantile regression and survival analysis, and Bayesian quantile regression. Course lectures will be complemented by a computationally oriented interlude designed to give students experience with application of the methods. This session will be conducted in the open-source R language and rely on the quantreg package.

Missing Data Methods for Regression Modeling (1-day) Joe Ibrahim, The University of North Carolina

This short course covers an important topic in statistical inference, namely missing data methods in regression models. Missing data is a major issue in many applied problems, especially in the biomedical sciences, including clinical trials, longitudinal studies, observational studies, and sample surveys. The short course on such a topic is timely, since much software has been recently developed to fit various types of regression models with missing covariate and/or response data. One unique and extremely strong feature in this short course is that it will focus on regression models and research problems encountered in actual practice and demonstrate a wide variety of statistical packages dealing with missing data, including SAS, logXact, and WinBUGS.

Regression models covered will include linear and generalized linear models, models for longitudinal data, and survival models. Missing responses and/or covariates will be examined as well as ignorable and nonignorable missing data mechanisms. This short course will be comprehensive in its coverage of the various methodologies for handling missing data, including detailed coverage of maximum likelihood, multiple imputation, fully Bayesian methods, and weighted estimating equations. We will examine several case studies with missing data and demonstrate the various missing data methodologies using these case studies.

Applied Longitudinal Analysis (1-day) Garrett Fitzmaurice, Harvard University • Nan Laird, Harvard University

The goal of this course is to provide a broad introduction to statistical methods for analyzing longitudinal data. The main emphasis is on the practical aspects of longitudinal analysis. The course begins with a review of established methods for longitudinal data analysis when the response of interest is continuous. A general introduction to linear mixed effects models for continuous responses is presented. Next, we discuss how smoothing and semiparametric regression allow greater flexibility for the form of the relationship between the mean response and covariates. We demonstrate how the mixed model representation of penalized splines makes this extension straightforward. When the response of interest is categorical (e.g., binary or count data), two main extensions of generalized linear models to longitudinal data have been proposed: marginal models and generalized linear mixed models. While both classes of models account for the within-subject correlation among the repeated measures, they differ in approach. In this course, we highlight the main distinctions between these models and discuss the types of scientific questions addressed by each.

Visit the Biometrics Section website for more information. For information about the continuing education courses visit the ASA’s Joint Statistical Meetings website.