Abstract Number:
1672
Submission Type:
Topic-Contributed Paper Session
Participants:
Michael Pennell (1), Michael Pennell (1), Garnet Anderson (2), Janet Tooze (3), Haley Hedlin (4), Ran Dai (5), Alexandra Binder (6)
Institutions:
(1) The Ohio State University, N/A, (2) Fred Hutchinson Cancer Research Center, N/A, (3) Wake Forest School of Medicine, N/A, (4) Stanford University, N/A, (5) University of Nebraska Medical Center, N/A, (6) University of Hawaii, N/A
Chair:
Session Organizer:
Speaker(s):
Ran Dai
University of Nebraska Medical Center
Session Description:
The Women's Health Initiative is an ongoing cohort study of post-menopausal women that initiated in 1993. A total of 161,808 women have been enrolled across 40 U.S. study centers making it valuable resource for studying women's health and risk factors for disease incidence. As with any large cohort study, the WHI has presented researchers with some interesting statistical challenges. The purpose of this session is to introduce the WHI to statistical researchers, present practical and sophisticated approaches to dealing with complications in the WHI data, describe new methods that have been developed in response to these challenges, and discuss opportunities for future statistical research in the WHI.
Our first speaker will be Dr. Garnet Anderson, Principal Investigator of the WHI Clinical Coordinating Center at Fred Hutchinson Cancer Center. Dr. Anderson will present an overview of the WHI including the data collected and challenges in working with the data. Our second speaker, Dr. Alexandra Binder, will discuss selection bias issues in studies of WHI cancer survivors and methods for remediating these problems. Our third speaker, Dr. Janet Tooze, will focus on measurement error in WHI dietary data and present ways to address these problems at the design and analysis phases of an epidemiologic study. Dr. Ran Dai, will present a novel high dimensional variable selection approach motivated by problems with missing data and measurement error in metabolomic data collected in the WHI. Finally, Dr. Haley Hedlin will discuss multiple testing problems in secondary analyses of large cohorts.
Although this session is centered on solutions to problems from a single (albeit large) cohort study, the issues that motivated these methods are common to many observational cohort studies. Thus, we expect the session to attract interest from a wide variety of data scientists, particularly those interested in epidemiological study design, causal inference, longitudinal data analysis, and solutions to messy data. Furthermore, since the WHI data are freely available, this session will be valuable to those looking for motivating or example data sets for their statistical research.
Sponsors:
Biometrics Section 2
ENAR 3
Section on Statistics in Epidemiology 1
Theme:
Statistics and Data Science: Informing Policy and Countering Misinformation
No
Applied
Yes
Estimated Audience Size
Medium (80-150)
I have read and understand that JSM participants must abide by the Participant Guidelines.
Yes
I understand and have communicated to my proposed speakers that JSM participants must register and pay the appropriate registration fee by June 1, 2024. The registration fee is nonrefundable.
I understand