Data from Population-based Cancer Registration for Secondary Data Analysis: Methodological Challenges and Perspectives

Abstract

Population-based cancer registries have a long-standing role in cancer monitoring. Scientific use of cancer registry data is one important purpose of cancer registration, but use of cancer registry data is not restricted to cancer registries. Cancer registration in Germany is currently heading towards population-based collection of detailed clinical data. This development together with additional options for record linkage and long-term follow-up will offer new opportunities for health services and outcome research. Both regional population-based registries and the German Centre for Cancer Registry Data (ZfKD) at the Robert Koch-Institute as well as international cancer registries and consortia or organizations may provide external researchers access to individual or aggregate level data for secondary data analysis. In this review, we elaborate on the access to cancer registry data for research purposes, availability of specific data items, and options for data linkage with external data sources. We also discuss as well as on limitations in data availability and quality, and describe typical biases in design and analysis.

Small‐area analysis on socioeconomic inequalities in cancer survival for 25 cancer sites in Germany

“…Hence, data quality of German cancer registries has improved and is going to improve through the implementation of clinical cancer registries. 44,45 A strength of our study was the large cohort of the cancer patients of a population of 32 million German residents. Furthermore, we used data from population-based cancer registries with a completeness of more than 90% in 2014.…”

“…Furthermore, we used data from population-based cancer registries with a completeness of more than 90% in 2014. 44 It was possible to investigate survival differences in the 25 most common cancer sites. Area-based socioeconomic deprivation in our study region was assessed on a relatively small area-level (median population for included municipalities:…”

Incidence of Smoking-Related Second Primary Cancers After Lung Cancer in Germany: An Analysis of Nationwide Cancer Registry Data

“…Within the context of substantial legal and organizational changes in cancer registration in Germany since 2013, 13 it remains unclear whether follow-up time and the actual application of multiple primary definition play an important role in SIR estimations. We observed largely increased rates for less than 6 months follow-up time that might be due to increased patient surveillance during the treatment period or registration of nonindependent second cancers; however, we excluded this follow-up period from our main analysis (Supplementary Fig.…”

“…Incomplete record linkage of PBCR with death registries leads to immortal time bias. 13 Both effects increase follow-up time for the expected count, therefore underestimating the true SIR.…”

“…The ZfKD collected anonymous data comprising basic patient information, such as year and month of birth, sex, region of residence, vital status, year, month and data on type of diagnosis, location, morphology, staging, and grading of the tumor. 13 Data from 11 German states (Bavaria, Brandenburg, Bremen, Hamburg, Mecklenburg-Western Pomerania, Lower Saxony, Muenster district, Saarland, Saxony, Schleswig-Holstein, Thuringia) covering approximately 50% of the German population were pooled for this study (Supplementary Fig. 1).…”