Disclosing progress in cancer survival with less delay.

Area of Science:

• Epidemiology
• Public Health
• Biostatistics

Background:

• Cancer registration is crucial for monitoring cancer burden.
• Timeliness of cancer registry data is often delayed to ensure completeness.
• This delay impacts the ability to provide up-to-date survival estimates.

Purpose of the Study:

• To investigate if delayed cancer registry data is mandatory for survival analyses.
• To determine if earlier data access can yield accurate, up-to-date survival estimates.
• To compare survival estimates using datasets with varying delays in case ascertainment.

Main Methods:

• Compared prognostic factor distributions and period relative survival estimates.
• Utilized population-based cancer registry data from three German regions (Schleswig-Holstein, Rhineland-Palatinate, Saarland).
• Analyzed datasets extracted 1 to 5 years post-diagnosis (DY+1 to DY+5), focusing on seven cancer sites.

Main Results:

• Data completeness varied significantly across registries (e.g., 57% in DY+1 for SH, 2% for RP, 26% for SA).
• Five-year survival estimates using DY+1 data were comparable to reference (DY+5) data (mean absolute deviations 0.2-0.6%).
• Minor deviations (>1%) were observed for specific cancers (pancreatic, lung, leukemia) in certain registries; 1-year survival estimates required slightly longer delays (1-2 years).

Conclusions:

• Delayed cancer registry data is not strictly mandatory for reliable survival analyses.
• Earlier use of cancer registry data can provide more timely survival estimates.
• This facilitates quicker disclosure of progress in cancer survival, improving public health monitoring.