Modelling the recurrence of bladder cancer

The risk of a new tumor recurrence after transurethral resection (surgical operation) in patients with primary superficial bladder carcinoma is evaluated. As the exact time of the event is not known, it is suggested to consider interval-censored survival data methodology. We use the model of Farrington to perform a prognostic model for predicting a new recurrence and to determine the clinicopathologic factors associated within interval-censored data framework. The implementation of this method is easy using standard statistical packages. Anyway, we use diagnostic methods for the validation of the model developed by the same author for interval-censored data. The obtained model is easily interpretable and let us to individualize the course of follow-up and treatment of each patient in order to improve the patient's quality of life.