Record Details

Title Summary Stage: Data Effects of the Changes in 2000
Author Phillips, JL
Secondary Authors
Publication Type (Help) book
Month December
Year 2003
Volume
Volume
Series
Number
Publisher North American Association of Central Cancer Registries
Address Springfield, IL
Note
URL http://www.naaccr.org/LinkClick.aspx?fileticket=SV_vXj1Od90%3d&tabid=94&mid=434
PubMed ID
NCI Id
EPub Date
Citation Phillips JL (ed.) Summary Stage: Data Effects of the Changes in 2000. Springfield, IL: North American Association of Central Cancer Registries, December 2003. [http://www.naaccr.org/LinkClick.aspx?fileticket=SV_vXj1Od90%3d&tabid=94&mid=434.]
  BibTex    


Abstract

The Surveillance, Epidemiology, and End Results (SEER) Program developed SEER Summary Stage 2000 (SSS2000) to address issues related to directly coded summary stage and discrepancies in summary stage classification being identified between direct and indirect coding procedures. All of the changes to the extent of disease coding system (EOD) that occurred between 1977 and 2000 were incorporated into the modified SSS2000. The Summary Stage: Data Effects of the Changes in 2000 was written for registries that coded SSS2000 directly. Selection of cancer sites to include in the study was based on: (1) changes between SSS1977 and SSS2000 (colon cancer had none and thus was excluded; (2) completeness of staging information (melanoma was excluded due to expected high frequency of missing data from records of diagnosis and treatment occurring outside the registry’s reporting infrastructure); and (3) absence of other external factors that might affect time trends (prostate cancer was excluded based on regional variation in screening affecting the stage trends). As a result, invasive carcinomas of the lung and female breast were selected for the study. Cases were randomly sampled from three volunteer central registries. The change from SSS1977 to SSS2000 is only one factor affecting longitudinal use of the two Summary Stage coding systems. Although the Kappa statistics showed good-to-excellent agreement beyond chance between the registry records and the reabstracted SSS1977, all three central registries did have some discrepancies. The abstractors in the study often were required to assign codes based on a strict interpretation of SSS1977 that was inconsistent with their own experience. Finally, few central registries provided extensive field training in the coding of SSS1977.



Keywords

Keyword
summary stage