Simpson's Paradox: Aggregating and Partitioning Populations in Health Care Disparities of Lung Cancer Patients

• Activity Number: 107
• Type: Invited
• Date/Time: Monday, August 2, 2010 : 8:30 AM to 10:20 AM
• Sponsor: Committee on Applied Statisticians
• Author(s): Pingfu Fu*+ and Mark Schluchter and Rom Leidner and Balazs Halmos
• Companies: Case Western Reserve University and Case Western Reserve University and Case Western Reserve University and Columbia University Medical Center
• Keywords: Simpson's paradox ; homogeneity ; lung cancer ; separation of logistic regression ; EGFR FISH ; EGFR mutation

Abstract

In the development of targeted therapeutic agents, attempts including novel designs and other strategies to better characterize tumors with the goal of finding more specific therapies that are tailored to a particular tumor type as determined by molecular and genetic signatures continue to play a big role. Statistical inferences based on aggregating and partitioning data may have contradicting conclusions. We give a real life example regarding genetic abnormalities of the EGFR pathway in patients with Non-Small Cell Lung Cancer and deal with it using the methods that solve the problem of separation in logistic regression.