Exploring contrary trends in bladder cancer incidence, mortality and survival: implications for research and cancer control

Aim: To investigate trends in bladder cancer incidence, mortality and survival, and cancer-control implications. Methods: South Australian Registry data were used to calculate age-standardized incidence and mortality rates from 1980 to 2004. Sociodemographic predictors of invasive as opposed to in situ disease were examined. Determinants of disease-specific survival were investigated using Kaplan-Meier estimates and proportional hazards regression. Results: Incidence rates for invasive cancers decreased by 21% between 1980-84 and 2000-04, similarly affecting men and women. Meanwhile increases occurred for combined in situ and invasive disease. While mortality rates decreased by approximately a third in men and women less than 70 years of age after the early 1990s, no changes were evident for older residents. The proportion of cancers found at an in situ stage was higher in younger ages and more recent diagnostic periods. Five-year survivals of invasive cases decreased from 64% for 1980-84 diagnoses to 58% for 1995-2004. Multivariable analysis showed that diagnostic period was not predictive of survival after age adjustment (P = 0.719), with lower survival relating to older age, transitional compared with papillary transitional cancers, female sex, indigenous status and a country as opposed to metropolitan residence. Conclusions: Reductions in invasive disease incidence may be due to increased detection at an in situ stage. The decline in survival from invasive disease in more recent periods is explained by increased age at diagnosis. Poorer outcomes of invasive cases remain for women after adjusting for age, histology, indigenous status and residential location.