Joint Statistical Meetings (JSM) 2022: Topic-Contributed Session on N-of-1
Last updated: Apr 02, 2022
I thank my co-organizer and co-host Professor Francisco Diaz for prompting me to submit a proposal this year (after I was unable to secure a spot for this same Topic-Contributed Session last year), and for re-writing the proposal text. I also thank Dr. Ziwei Liao from Columbia University for agreeing to chair our Session.
I’m particularly excited that while four of the speakers (including me) hail from biostatistics or statistics, our key speaker comes from psychology. Professor Mariola Moeyaert’s presence on our interdisciplinary panel helps advance our Stats-of-1 mission to build bridges across the various fields that all focus on esametric (i.e., quantitative idiographic) methodological approaches.
In addition, our session was selected as a “Topic-Contributed” session—a welcome bellwether of the rising popularity of n-of-1, single-subject, or within-individual approaches. (The JSM 2022 Topic is “Statistics: A Foundation for Innovation”.)
Here are the deets.
Session Number: 471
Date and Time: Wed, 8/10/2022, 2:00 PM - 3:50 PM (Eastern Time)
Description: The advent of personalized medicine and the availability of wireless devices for monitoring clinical responses have aroused a renewed interest in the utilization of N-of-1 clinical trials to individualize medical or pharmacological treatments in clinical or daily life settings. These trials are ones conducted on a single patient with a chronic, stable, or slowly progressing condition to decide which of two or more treatments is the most beneficial for the patient. The related counterpart in psychological research and practice is called a single-case experimental design (SCED). During the trial and by design, the patient crosses over from one treatment to another. The goal is not to find an optimal treatment sequence applicable to many individuals, but rather to discover the optimal individualized treatment that will be administered regularly to the single patient after the N-of-1 trial is completed. The authors of the statistical and medical literature on N-of-1 trials have investigated two types of methodological problems. One type of problem is how to combine N-of-1 trials conducted with different patients in order to compare treatments and draw statistical inferences about the patient population by identifying within-individual patterns that are common across patients in a “bottom-up” approach. This includes how to conduct meta-analyses. The other problem type is the development of statistical methods that assist the clinician in deciding which treatment is more appropriate for a specific patient. This session embraces these two types of problems. It will present modern approaches for decision-making in n-of-1 trials or their statistical analyses, including methods based on random effects models and/or causal inference. The session will consider both controlled trials and uncontrolled ones from observational studies.
Speakers and Talks:
- 2:05 PM Efficient Sequential N-Of-1 Trials and Observational Studies
- Nicholas Schork, The Translational Genomics Research Institute (TGen)
- 2:25 PM Using Population Crossover Trials to Improve the Decision Process Regarding Treatment Individualization in N-Of-1 Trials
- Francisco J. Diaz, The University of Kansas Medical Center
- 2:45 PM Experiences with Large Series of Individual Trials for Personalized Health
- Christopher Schmid, Brown University
- 3:05 PM Single Case Designs: MVA’: a Mobile Application for Conducting Single-Case Experimental Research Remotely
- Mariola Moeyaert, State University of New York at Albany
- 3:25 PM Individual Average Treatment Effect Estimation using Real-World Observational Data from Wearables
- Eric J. Daza, Evidation Health; Logan Schneider, Alphabet Inc.
- 3:45 PM Floor Discussion
This is a golden opportunity to showcase the power and potential of n-of-1 studies and single-case designs to truly personalize health and medicine: JSM is one of the oldest and largest professional conferences in the world. More on JSM from the main conference website:
JSM Is One of the Largest Statistical Events in the World
- More than 6,500 attendees from 52 countries
- 600+ sessions, including invited, topic-contributed, contributed, and poster
- More than 1,000 student attendees
- 75+ employers hiring for more than 200 positions
- 100+ exhibitors
- More than 40 professional development short courses and workshops
It is also one of the broadest, with topics ranging from statistical applications to methodology and theory to the expanding boundaries of statistics, such as analytics and data science.
JSM also offers a unique opportunity for statisticians in academia, industry, and government to exchange ideas and explore opportunities for collaboration. Beginning statisticians (including current students) can learn from and interact with senior members of the profession.