Effectiveness of Delay treatment time for advanced oral cancer in Taiwan: Propensity analysis

Aim: To determine the effectiveness of delay treatment time (the interval from diagnosis to treatment ,in days), and influenced these survival for oral cancer patients in Taiwan. Methods: It was a retrospective follow up study of a cohort of patients registered in Taiwan Cancer Database (TCDB) linked Taiwan's Death Registries of new oral cancer patients diagnosed between 2002 and 2006. It considered the sample selection bias in observation data. This study was analysed using the Cox proportional hazards regression with propensity score matching model. Main results: 3,592 cases of oral cancer were analysed. In the time interval between the first diagnosis and treatment, a significantly shorter mean delay was found for patients with the control group((the interval from diagnosis to treatment less than 30 days) compared to the case group ( over 30 days )(14.9�7.9 v.s. 68.6�71.25 P<0.0001). As expected, the probability survival of case group was shorter than the control group after controlling other influence factors(HR��1.15, 95%CI: 1.04-1.28). Conclusions: The longer delays are associated with poorer survival. The study provides the a baseline for the effectiveness of delay treatment time association with outcome and should assess progress in reducing treatment delays in Taiwan.