SPEAKER: Ronghui (Lily) Xu, UC San Diego
TITLE: “Applying Additive Hazards Model to Learn from Electronic Health Data”
ABSTRACT: Our work was motivated by the analysis projects using the linked US SEER-Medicare database to studying treatment effects in men of age 65 years or older who were diagnosed with prostate cancer. Such data sets contain up to 100,000 human subjects and over 20,000 claim codes. The data are obviously not randomized with regard to the treatment of interest, for example, radical prostatectomy versus conservative treatment. Instrumental variable (IV) is an essential tool for addressing unmeasured confounding in this type of observational data. We develop an appropriate IV approach using the additive hazards model for patient survival, correcting the error in the literature using the Cox model as well as assuming a linear relationship between the binary treatment and the IV. Additionally, we have previously shown that the high dimensional claim codes contain rich information about the patients’ non-cancer mortality, possibly capturing some of the unmeasured confounding beyond the usual clinical variables. We describe our approaches to making inference about the treatment effect using the high dimensional claim codes.
DATE: Thursday, May 2nd, 4:10pm
LOCATION: MSB 1147, Colloquium Room
REFRESHMENTS: 3:30pm MSB 4110 (4th floor lounge)
STA 290 Seminar List: https://statistics.ucdavis.edu/seminars