Accelerated Hazards Model: Method, Theory and Applications

Ying Qing Chen, Division of Biostatistics, School of Public Health, University of California, Berkeley
Nicholas P. Jewell, Division of Biostatistics, School of Public Health, University of California, Berkeley
Jingrong Yang, Division of Biostatistics, School of Public Health, University of California, Berkeley

Abstract

In an accelerated hazards model, the hazard functions of a failure time are related through the time scale-change, which is often a function of covariates and associated parameters. When the hazard functions have special properties, such as monotonicity in time, the parameters may be clinically meaningful in measuring a treatment effect. This paper reviews methodological and theoretical development of this model. Applications of the accelerated hazards model including sample size calculation in clinical trials, are also explored.