Long-term trends in bladder cancer incidence using a harmonized staging variable – A SEER-based study

Abstract Background: Coding changes in disease definitions have influenced trends in bladder cancer epidemiologic outcomes in registries. The Surveillance, Epidemiology, and End Results (SEER) Program introduced a harmonized staging variable (available by request to the SEER Program) for long-term trend analysis. This study analyzes trends in bladder cancer incidence using the harmonized staging variable. Methods: Using SEER 12 registry data, we compared trends in the incidence of bladder cancer cases diagnosed during 1992-2019 using the revised (or harmonized) staging variable against the original staging variables (SEER modified AJCC 3rd edition for 1992-2003, Derived AJCC 6th edition for 2004-15, Derived SEER Combined for 2016-17, Derived EOD 2018 for 2018-2019). We used JoinPoint regression to analyze changes in trends. Results: The data availability has improved with the revised staging system as the proportion of cases with missing N (lymph node), and M (metastasis) stages was substantially reduced. However, the trends varied by T stages between the two systems. There were generally more discontinuities in trends with the original system than with the revised system. Unlike the trend observed with the original staging system, the harmonized staging system has shown a 4% annual decrease in Tis incidence since 1992. Conclusions: With the revised variable, we observed a consistent decrease in the incidence of Tis cases, and the trends appear smoother. Impact: Our study highlights the benefits of using revised staging variables to reveal previously hidden patterns, supporting the use of new variables for a more nuanced understanding of temporal trends in epidemiology.