The Etzioni Lab focuses on innovative statistical and computer modeling to project the comparative outcomes of cancer control interventions. Our projects “go beyond the data” to develop a deeper, more mechanistic understanding of cancer progression which we use to quantify the relative benefits and harms of candidate interventions and policies.
Recent projects have used modeling to:
Our work on cancer modeling is part of a broader research program that includes methodologic as well as applied research. A major area of methodologic interest is the estimation of lead time and overdiagnosis. Our group was the first to quantify the extent of overdiagnosis associated with prostate cancer screening in the US and we are also active in development of best practices for estimating overdiagnosis in breast cancer screening. Other methods work includes the development of Bayesian models and methods to study longitudinal biomarker trajectories and their associations with disease transitions in the pre- and post-diagnosis settings. We are using these these methods to study disease progression following PSA recurrence as well as in the active surveillance setting.
In addition to these projects, the Etzioni Lab leads the Biostatistics Core for the Pacific Northwest Prostate Cancer SPORE and serves as a central consulting resource for the prostate cancer investigators at the FHCRC, the SCCA, and the University of Washington. Our investigators are statisticians with broad expertise in data analysis, simulation modeling, Bayesian methods, statistical programming, and advanced graphics.
Dr. Etzioni will present part 3 of NIH 3-part series on disease prevention screening
Friday, November 18, 2016
1:00 p.m. to 2:00 p.m. Eastern time
Presented via NIH Videocast.
Registration is encouraged for planning purposes.
The problem of overdiagnosis, the detection by screening of latent cancers that would never have surfaced, has been much in the news lately. What is overdiagnosis, and how significant is the problem? In this presentation, Dr. Etzioni will examine how overdiagnosis arises and will discuss what it takes to validly estimate its frequency. Different types of study designs and estimation methods have been used, but many of these probably yield biased results. Consumers of overdiagnosis studies need to carefully navigate the published literature to properly understand the extent of the problem. Dr. Etzioni will provide a guide using examples from breast and prostate cancer screening to identify the types of studies that are most likely to produce reliable results.
Dr. Etzioni is a biostatistician and full member in the Division of Public Health Sciences at the Fred Hutchinson Cancer Research Center. Her research program focuses on the development of innovative statistical and computer models to learn about the latent process of cancer progression from observed data on disease incidence and mortality.
Medicine: Mind the Gap is a seminar series that explores research design, measurement, intervention, data analysis, and other methods of interest to prevention science.
Dr. Etzioni will accept questions about her presentation via email at firstname.lastname@example.org and on Twitter with #NIHMtG.