A Multiple Imputation Approach For Semiparametric Cure Model With Interval Censored Data

Interval censored survival data, where the exact event time is only known to lie in an observed time interval, is commonly encountered in practice. Such data analysis may be conducted under the setting where a fraction of patients can be considered as fully recovered and will not experience the event of interest in the future; while the other patients who did not recover totally will have the outcome of interest. We proposed a semiparametric estimation method for the proportional hazard mixture cure model, which is easy to implement and computationally efficient. A multiple imputation approach based on the asymptotic normal data augmentation (ANDA) is used to obtain parameter and variance estimates for both the cure probability and survival probability for uncured patients. A simulation study is performed to evaluate the proposed method and the results are compared with a fully parametric approach. The proposed method is applied to 2000-2010 Greater Georgia breast cancer dataset from the Surveillance, Epidemiology, and End Results (SEER) Program