Webinar Series

LiDS Webinar and Short Course Series

The following webinars are sponsored by the Lifetime Data Science Section.  These are part of the web-based lectures hosted by the ASA.

Upcoming

Short course: Dynamic Prediction Methods
Instructor: Alessandro Gasparini, Ph.D., Red Door Analytics, Stockholm, Sweden
Date and Time: February 12, 2025, 11:00 a.m. – 1:00 p.m. ET (This webinar will be taught via Zoom)
Sponsor: Lifetime Data Science Section
Registration Deadline: TBA

Webinar Description:
Prediction models in clinical settings are routinely developed using traditional, prospective study designs that define a baseline (origin) at which predictors are measured and from which to predict future risk. However, the increased availability and use of electronic health records and data registers for research purposes provide a large wealth of dynamic information collected over time, information that is directly related to disease status, progression, cure, and relapse. The hope is that such information can be used to inform and individualize predictions based on a dynamic assessment of a patient's characteristics: for instance, biomarker values and their dynamics could be predictive of future risk. Therefore, accommodating these time-varying features within a prediction model can enable dynamic predictions for updating the prognosis of a patient whenever new data is available. Several estimators have been proposed for the task of dynamic prediction, mainly from two approaches: joint modeling and landmarking. These approaches differ in terms of what information is used and how, underlying modeling assumptions, and computational complexity. In this short workshop, we will introduce the joint modeling and landmarking approaches for dynamic prediction, including clear definitions of risk estimators, various modeling strategies, and performance metrics. The two approaches are illustrated in practice using openly available observational data on heart function after surgery. Finally, state-of-the-art developments in the field are introduced and discussed as well.

Registration: https://amstat.users.membersuite.com/events/2e04bc98-0078-cb60-b2cc-0b4788873142/details (please note that this is the direct link; event registration is also accessible through Membership Portal on the ASA website under the "Events" tab: https://amstat.users.membersuite.com/events/browse)

Registration Fees:
Lifetime Data Science Section Members: $20
ASA Members: $30
Student ASA Member: $25
Nonmembers: $45

Access Information
After registering, you will receive a confirmation email. In the body of the confirmation email, you will receive the Zoom link for the webinar. 

Past

Title:  Semi-Competing Risks: Accounting for Death as a Competing Risk when the Outcome of Interest is Non-terminal

Presenter: Sebastien Haneuse, Professor of Biostatistics, Harvard University

Date and Time: June 28, 2024, 12:00 p.m. – 2:00 p.m. ET (This webinar was taught via Zoom)

Registration: https://amstat.users.membersuite.com/events/2e04bc98-0078-cc10-cd06-0b46dc93b4a4/details

Webinar Description: The short-course will provide an overview of semi-competing risks data analysis. Briefly, semi-competing risks corresponds to the setting where primary interest lies in some non-terminal event, the occurrence of which is subject to a terminal event. Although not as well-known as standard competing risks, semi-competing risks arise in any study of any event that is not mortality but where the force of mortality is strong. Examples include: Alzheimer’s disease in the elderly; quality of end-of-life care among patients with a terminal cancer diagnosis; graft-versus-host disease among bone marrow transplant recipients; and, developmental outcomes among infants admitted to a NICU. Semi-competing risks also arise in some settings where the terminal event is not mortality. In studies of preeclampsia, for example, “delivery” is a competing risk but not vice-versa. In this workshop, we will cover basic concepts of semi-competing risks, various modeling strategies, methods for prediction, and software. In addition, we will apply and illustrate the methods to a study of preeclampsia using data from the Beth Israel Deaconess Medical Center, in Boston, MA, specifically with the goals of quantifying risk factor associations and the joint prediction of preeclampsia and delivery.

Registration Fees:
Lifetime Data Science Section Members: $20
ASA Members: $30
Student ASA Member: $25
Nonmembers: $45

Title: Cure Models in Survival Analysis
Presenter: Ingrid Van Keilegom, Professor of Statistics, KU Leuven, Belgium

Date and Time: March 20, 2024, 12:00 p.m. – 2:00 p.m. ET (This webinar was taught via Zoom)

Registration: https://amstat.users.membersuite.com/events/2e04bc98-0078-cf52-7eee-0b4668a66f56/details

Webinar Description: When analyzing time-to-event data, it often happens that a certain fraction of the data corresponds to subjects who will never experience the event of interest. These event times are considered as infinite and the subjects are said to be cured. Survival models that take this feature into account are commonly referred to as cure models. This short course aims to review the literature on cure (regression) models in which the event time is subject to random right censoring and has a positive probability to be equal to infinity.  We will focus on the existing models, methodologies for estimating the cure fraction and the survival function of the uncured subjects, appropriate tests in these models, modern software packages for cure models, as well as a number of practical examples. 

Registration Fees:
Lifetime Data Science Section Members: $20
ASA Members: $30
Student ASA Member: $25
Nonmembers: $45

Title: Survival Models for Spatial Data
Presenter: Benjamin M. Taylor, Lancaster University, UK
Date and Time: Tuesday, October 18, 2022, 11:00 a.m. – 1:00 p.m. ET (This course will be taught via Zoom)
Sponsor: Lifetime Data Science Section

Registration Deadline: Tuesday, October 18, at 10:30 a.m. Eastern time

Course Description:
Statistical methods for the analysis of survival data are not only applicable in the medical context, but also in many other areas of science and engineering. When survival times are spatially-referenced, some evidence of clustering of high or low times might be apparent on a visual inspection of the data. The question naturally arises as to whether these observed spatial survival patterns can be explained by incorporating appropriate covariates into the model or whether, in order to obtain reliable inferences for model parameters of interest, it is necessary to explicitly model the unexplained spatial variation. In this short course, we give an overview of spatial survival analysis: models, inference, and software and cover three applications of these techniques to real-world challenges.

Part 1 - Background and Review of Methodology
---------------------------------------------
- An introduction to the modeling of spatial stochastic processes
- Combining spatial modeling with survival analyses
- Model selection
- Flexible parametric survival models
- Inference and software for fitting spatial survival models
- Extensions

Part 2 - Applications
---------------------------------------------
Application 1: Modelling Survival in HIV Cohorts in Malawi
Application 2: Modelling Survival from Colorectal Cancer in Malaysia
Application 3: Spatial Modelling of Emergency Service Response Times

Title: Analysis of Composite Time-to-event Outcomes

Presenter: Lu Mao, Department of Biostatistics and Medical Informatics, University of Wisconsin - Madison

Date and Time: June 8, 2022, 12:00 p.m. – 2:00 p.m. ET (This course will be taught via Zoom)

Sponsor: ASA Lifetime Data Science Section

Course Description:
This course provides an overview of the new 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.

Short course: Joint Modeling of Longitudinal and Survival Data
Instructors: Dimitris Rizopoulos, Professor of Biostatistics, Erasmus University Medical Center, Rotterdam, The Netherlands
Sponsor: ASA Lifetime Data Science Section
Date and Time:  Thursday, March 24, 2022, 11:00 a.m. – 2:00 p.m. ET (This course will be taught via Zoom)

Course Description:  In follow-up studies, different types of outcomes are typically collected for each subject. These include longitudinally measured responses (e.g., biomarkers) and the time until an event of interest occurs (e.g., death, dropout). These outcomes are often separately analyzed, but on many occasions, it is of scientific interest to study their association. This research question has given rise to the class of joint models for longitudinal and time-to-event data. These models constitute an attractive paradigm for the analysis of follow-up data that is mainly applicable in two settings: First when the focus is on a survival outcome, and we wish to account for the effect of endogenous time-dependent covariates measured with error, and second when the focus is on the longitudinal outcome, and we wish to correct for non-random dropout. This short course will introduce when these models should be used in practice, which are the key assumptions behind them, and how they can be utilized to extract relevant information from the data.

Time permitting, the following topics will be covered:

• Review of Relative Risk and Mixed Models: Definitions, Cox and linear mixed models, how to fit them in R
• The Basic Joint Model: Definition of joint models, assumptions, estimation, comparison with time-dependent Cox model, connection with missing data
• Extensions of the Basic Joint Model: Functional form, Multiple longitudinal outcomes, multiple failure times
• Special topics: Dynamic predictions for the survival and longitudinal outcomes

This course assumes knowledge of basic statistical concepts, such as standard statistical inference using maximum likelihood and regression models. Also, a basic knowledge of R would be beneficial but is not required.

Registration: www.amstat.org/education/web-based-lectures#JMLSD
Registration Deadline: Wednesday, March 23, at 12:00 p.m. ET
Fees:
 LiDS Section Members: $20
 Regular ASA Members: $30
 Students ASA Member: $25
 Nonmembers $45

Each registration is allowed one web connection. Sound is received via audio streaming from your computer's speakers.

Access Information

After registering, an automated email confirmation will be sent to you with instructions on how to access the webinar.

Title: Short course: Causal Time-to-Event Analysis
Instructors: Stijn Vansteelandt (Ghent University)
                     Oliver Dukes (Ghent University, University of Pennsylvania)
                     Torben Martinussen (University of Copenhagen)
Dates and Times (This course will be taught via Zoom on 2 consecutive days): 
      Part 1: November 29, 2021: 12:00 - 2:30 p.m. ET
      Part 2: November 30, 2021: 12:00 - 2:30 p.m. ET

Course Description:  Evaluating treatment effects using observational data increasingly requires adjustment for high-dimensional set of covariates in order to control confounding. This is the result of a lack of comparability between treated and untreated subjects in possibly many (pre-treatment) factors that are also related to outcome. Such adjustment is routinely achieved via parametric modelling in a manner that is often not directly targeting the causal question of interest. This can lead to bias (e.g., due to model misspecification or poor detection of confounders), inefficiency and invalid inference (e.g., as a result of ignoring the uncertainty induced by variable selection). In this course, we will review targeted strategies for inferring the causal effect of an exposure on a time-to-event endpoint subject to censoring. On the first day, we will review efficient analysis strategies for randomised experiments, as well as robust methods for analysing observational data. Special attention will be given to the problem of selection and modelling of variables that are sufficient to adjust for confounding and to render censoring non-informative; we will in particular introduce double/triple variable selection strategies, as well as targeted learning techniques. On the second day, we will introduce instrumental variable methods for time-to-event data, as well as methods for inferring causal mechanisms based on the mediational g-formula as well as natural effect models. Software demonstrations in R will be used throughout the course.

The course is aimed to be accessible to applied statisticians, epidemiologists, and other quantitative researchers already familiar with standard survival analysis methods as well as with the language of potential/counterfactual outcomes and with identification (i.e., exchangeability or ignorability assumptions), for example for the Average treatment effect.

Title: ASA Lifetime Data Science Section Career Development Webinar
Cost: Free for LiDS members
Date and Time: Wednesday, October 20, 2021, 12:00 p.m. - 1:30 p.m. Eastern Time

Do you want to increase your chances of success in publishing your statistics papers? Learn how to obtain grant funding for your research program? Then you won’t want to miss this Career Development webinar sponsored by the ASA Lifetime Data Science (LiDS) section!

You will be able to hear valuable advice and tips from two leading statisticians in the field about how to get your methodological papers accepted by the top statistics journals and write competitive grant proposals for the NIH and other funding agencies.  There will also be break-out sessions in which you will have opportunities to interact directly with the speakers and participants to share experiences and ask questions.  The webinar is free and only for LiDS members, but space is limited so please register early to secure your spot!

Speakers:

Malka Gorfine, PhD., is a Professor in the Department of Statistics and Operations Research at Tel Aviv University, Israel. Her research interests are in disease risk estimation and prediction using candidate genes and environmental risk factors from both family and case-control survival data; statistical genetics; non-parametric statistics; biostatistics; and machine learning. She was Co-Editor of Biometrics from 2017-2019; Associate Editor of Biometrics from 2009-2016, and a member of the Editorial Board of Lifetime Data Analysis from 2013-2016. 

Jeremy Taylor, PhD., is a Professor of Biostatistics and of Radiation Oncology at the University of Michigan. He is a Fellow of the American Statistical Association and recipient of the Mortimer Spiegelman Award from the American Public Health Association, the Michael Fry Award from the Radiation Research Society, and the Jerome Sacks Award from the National Institute of Statistical Sciences. His research interests include longitudinal and survival analysis, cure models, missing data, smoothing methods, clinical trial design, and surrogate and auxiliary variables. He was Editor of Biometrics from 2012-2014, a permanent member of the NIH Biostatistical Methods and Research Design Study Section (BMRD) from 2007-2012, and Chair of BMRD from 2010-2012. 

Doug Schaubel, Ph.D. (Moderator) is a Professor of Biostatistics at the University of Pennsylvania Perelman School of Medicine. His methodologic research interests include survival analysis, recurrent event data, causal inference, and evaluation of time-dependent treatments.  He is a Fellow of the American Statistical Association and received the Excellence in Research Award in 2015 from the University of Michigan School of Public Health. He is currently Associate Editor for Biometrics, Statistics in Medicine, Statistics in Biosciences, and Lifetime Data Analysis. He is also a Statistical Reviewer for JAMA Network Open.

Title: Survival Outcome Data with High-Dimensional Predictors: Methods and Applications
Presenter: Dr. Yi Li, Professor of Biostatistics, University of Michigan
Date and Time: Thursday, June 24, 2021, 1:00 p.m. - 5:00 p.m. Eastern Time

Description:
In the era of precision medicine, survival outcome data with high-throughput covariates and predictors are often collected. These high dimensional data defy classical survival regression models, which are either infeasible to fit or likely to incur low predictability because of overfitting. This short course will introduce various cutting-edge methods that handle survival outcome data with high dimensional predictors. I will cover statistical principles and concepts behind the methods, and will also discuss their applications to real medical examples.

Time permitting, the following topics will be covered:

1.  Survival analysis overview: basic concepts and models, e.g. Cox, Accelerated Failure Time (AFT), and Censored Quantile Regression (CQR) Models;
2. Survival models with high dimensional predictors (p>n): Regularized methods and Dantzig selector;
3. Survival analysis with ultra-high dimensional predictors (p>>n): Screening Methods, e.g, Principled sure independent screening (PSIS), Conditional screening, IPOD, Forward selection, etc;
4. Inference for survival models with high dimensional predictors (p>n).

The audience only needs to have some basic knowledge of regression analysis and survival analysis. The relevant papers and software for this short course can be found at: www-personal.umich.edu/~yili/resindex.html

Title: Competing Frameworks and Methods for Competing Risks Data
Presenter: Dr. Douglas Schaubel, Professor at the University of Pennsylvania Perelman School of Medicine
                   Chair-Elect, ASA LiDS Section
Date and Time: Friday, April 30, 2021, 12:00 p.m. - 2:00 p.m. Eastern Time

Description:
Competing risks data arise frequently in clinical and epidemiologic studies. Such data are characterized by a survival time that terminates due to one of several mutually exclusive causes. This webinar will cover the following: the two most commonly adopted frameworks for competing risk data; relevant estimands and estimators within each framework; the role of censoring as a competing risk; available modeling strategies; and causal inference in the competing risk setting. The main ideas will be illustrated through several real-data examples.

Title: Statistical Learning with Time-to-Event Outcome
Presenter: Dr. Noah Simon, University of Washington
Date and Time: Tuesday, January 26, 2021, 12:00 p.m. - 2:00 p.m. Eastern Time

Description:
We will discuss contemporary methodologies for statistical learning with time-to-event outcomes. We will discuss techniques for engaging with high-dimensional data, as well as methods appropriate for data of more modest dimensions with larger numbers of observations. We will look at modern software packages for statistical learning in these contexts, as well as the validation of these predictive models. Time permitting, we will also touch on deep learning with time-to-event data.

During this session we aim to engage with loss-based estimation for the time-to-event outcome; penalized regression; tree-based-survival models; as well as uses of cross-validation with kernel-weighted Kaplan-Meier estimation to evaluate/calibrate a statistical-learning-based model. We will engage with these methodologies using a number of practical examples.