We will develop a general "extended hazard" spatial survival model to predict geographical effects and geo- graphically varying effects in cancer survival. The proposed model can include the proportional hazards spatial survival model and the accelerated failure time spatial survival model as its special cases. Furthermore, the new model can correctly identify the geographical effects and geographically varying effects in cancer survival. The performance of the proposed model will be evaluated by a comprehensive simulation study. To demonstrate the usage of the proposed model, we will apply the proposed method to analyze prostate cancer within Louisiana from the Surveillance, Epidemiology, and End Results program, and prostate cancer data set from South Carolina Central Cancer Registry (SCCCR). The software development will solve the computational issue in practice and will enable the practitioners and researchers apply the proposed method easily.
We will develop a general "extended hazard" spatial survival model, which includes current spatial survival models as its special cases, to predict geographical effects and geographically varying effects in cancer survival. Therefore, the proposed model can be used when the proportional hazards assumption is not satisfied. We will conduct a comprehensive simulation study to compare its performances to other existing spatial survival models and apply it to investigate the spatial patterns and racial disparities of prostate cancer in Louisiana and in South Carolina.
|Li, Li; Hanson, Timothy; Zhang, Jiajia (2015) Spatial extended hazard model with application to prostate cancer survival. Biometrics 71:313-22|
|Zhou, Haiming; Hanson, Timothy; Jara, Alejandro et al. (2015) MODELLING COUNTY LEVEL BREAST CANCER SURVIVAL DATA USING A COVARIATE-ADJUSTED FRAILTY PROPORTIONAL HAZARDS MODEL. Ann Appl Stat 9:43-68|
|Chen, Yuhui; Hanson, Timothy; Zhang, Jiajia (2014) Accelerated hazards model based on parametric families generalized with Bernstein polynomials. Biometrics 70:192-201|