Flexible Modelling of the Cumulative Effects of Time-varying Covariates on the Hazard

Speaker: Michal Abrahamowicz

Affiliation: McGill University, Canada

Time: Monday 08/04/2013 from 14:00 to 15:00

Venue: Access Grid UWS. Presented from Parramatta (EB.1.32), accessible from Campbelltown (26.1.50) and Penrith (Y239).

Abstract: A flexible method for estimating the cumulative effects of the past values of a time-varying prognostic factor, within the Cox PH regression analyses of survival data, is proposed. Joint effect of the past values of a time-varying covariate $x(t)$, until time $\tau[x(t): 0 < t < \tau]$, on the hazard at time $\tau$ is modeled as the weighed cumulative exposure (WCE): $\text{WCE}(\tau ~|~ x(t), t<\tau) = {\sum w(\tau - t)*x(t)}$. The weight function $w(\tau - t)$, that assigns importance weights to past values as the function of time elapsed since exposure $(\tau - t)$ is modelled with flexible cubic regression splines. The estimated WCE is then included as a time-dependent covariate in the Cox's proportional hazards model. Estimation and inference will be discussed, and the method's will be assessed through simulations. Real-life examples will illustrate the applications to explore the risks of adverse effects of various treatments.

Biography: Dr. Michal Abrahamowicz is a James McGill Professor of Biostatistics at the Department of Epidemiology, Biostatistics and Occupational Health of McGill University. He obtained his PhD in Statistics & Econometrics from Krakow, Poland. His research involves both development of new, flexible statistical methods for survival analyses and collaborative research on applications of these methods in health research. He is the Principal Investigator on 2 major grants from the Canadian Institutes for Health Research (CIHR) that focus on the development, validation and applications of new statistical methods for pharmaco-epidemiology. He is a member of the Executive Committee of the International Society for Clinical Biostatistics (ISCB). His statistical research is supported by NSERC. He published more than 250 peer-reviewed papers, and has been the primary supervisor of more than 25 graduate students, in both Biostatistics and Epidemiology.