A0681
Title: An approach for long-term survival data with dependent censoring
Authors: Silvana Schneider - Federal University of Rio Grande do Sul (Brazil) [presenting]
Abstract: In long-term studies, some causes of censoring are generally falsely assumed to be independent, leading to bias being neglected. Therefore, a likelihood-based approach is proposed for long-term clustered survival data, which is suitable to accommodate the dependent censoring. The association between lifetimes and dependent censoring is accommodated through the conditional approach of the frailty models. The marginal distributions can be adjusted assuming Weibull or piecewise exponential distributions, respectively. A Monte Carlo Expectation-Maximization algorithm is developed to estimate the proposed estimators. The simulation study results show a small relative bias and coverage probability near the nominal level, indicating that the proposed approach works well. Moreover, the model identifiability is assured once data has a cluster structure. Finally, the survival times of free-ranging dogs from West Bengal, India, collected between 2010 to 2015, are analyzed, and it is concluded that survival time (death due to natural cause) is negatively correlated to dependent censoring (missing cause).
