SEMINAR: New tests to check the main assumptions of a flexible parametric excess hazard model
Coraline Danieli, PhD Postdoctoral Fellow – RIMUHC - McGill University New tests to check the main assumptions of a flexible parametric excess hazard model
Tuesday, 14 March 2017 3:30 pm – 4:30 pm - Purvis Hall, 1020 Pine Ave. West, Room 24 ALL ARE WELCOME
Abstract: In many clinical prognostic studies of patients diagnosed with a particular disease, individual causes of death remain unknown. Net survival, the one that would be observed if the disease under study were the only cause of death, is an important and increasingly used indicator in public health, especially in population-based studies. Estimates of net survival and of the effects of prognostic factors on net survival, can be obtained by excess hazard regression modeling. However, whereas various diagnostic tools were developed for overall survival analysis, few methods are available to check the assumptions of excess hazard models. We propose two formal tests to check the proportional hazards assumption and the validity of the functional forms of the covariate effects in the context of flexible parametric excess hazard modeling. These tests were adapted from martingale-residual-based tests for parametric modeling of overall survival to allow adding to the model a necessary element for net survival analysis: the population mortality hazard. We studied the size and the power of these tests through an extensive simulation study based on complex but realistic data. The new tests showed sizes close to the nominal values and satisfactory powers. The power of the proportionality test was similar or greater than that of other tests already available in the field of net survival. We illustrate the use of these tests with real data from French cancer registries.Â
SEE PDF FOR DETAILS
