STATISTICS SEMINAR

Due To Illness This Seminar Has Been Cancelled

SPEAKER:

Ori Davidov

Department of Statistics

University of Haifa, Israel

TITLE:

Overdiagnosis in Early Detection Programs with Application to Prostate Cancer

ABSTRACT:

Evaluating the benefit of early detection programs is often obscured by the phenomena of overdiagnosis. Overdiagnosis refers to the situation where a screening exams detects a disease that would have otherwise been undetected because its onset would have been preceded by death due to another cause. Although the probability of over diagnosis is an important quantity in the design of early detection programs as well as personal utility, it has not been rigorously studied. We analyze an idealized, general, early detection program and derive the mathematical expression for the probability of overdiagnosis. This quantity which we call the overdiagnosis function is rigorously studied. Several monotonically results are proved. The overdiagnosis function is evaluated numerically for prostate cancer and applied to a variety of screening schedules. Our investigation reveals that the probability of overdiagnosis is unexpectedly high for most standard schedules. The implications for population screening programs are examined. Finally the age of the last exam is studied from a decision theory point of view.

This is joint work with Marvin Zelen of Harvard University

About the Speaker

Dr. Ori Davidov is a lecturer at the Department of Statistics, University of Haifa, Israel. His research interests are in the area of biostatistics and he has a number of publications in this general area. Dr. Davidov graduated with a Sc.D. from the Department of Biostatistics at Harvard University in 1996. After graduation he did a two-year postdoctoral fellowship at the Fred Hutchinson Cancer Research Center and the Department of Biostatistics University of Washington. He then spent a year working at Merck Research Laboratories in New Jersey before taking up his present position.

References

Some relevant background references will be posted here shortly.

Department of Mathematics and Statistics

Graduate Program in Statistics

This page is maintained by Angelo Canty,
Last updated on February 5, 2003