Summary of the 2018 SEER*DMS Face-to-Face Meeting

February 15, 2019 External Website Policy

The Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer Institute (NCI) held its annual SEER Data Management System (SEER*DMS) Face-to-Face (F2F) Meeting on September 26-28, 2018. The meeting brought together members of the cancer surveillance community to discuss SEER*DMS enhancement projects, address relevant questions, and encourage collaboration. Representatives from 22 registries participated in person or online, and the event occurred at the NCI Shady Grove campus in Rockville, MD.

The F2F Meeting focused on updates from the NCI SEER Program and SEER*DMS workgroups, as well as discussions on cancer surveillance challenges and potential solutions forward. These challenges include: (1) the large amounts of data that must be converted from different formats; (2) the numerous information sources; and (3) the need to support new and complex infrastructure that integrates new types of data, including genomics, longitudinal treatment outcomes, and pharmacy data.

NCI SEER Program Updates

A few updates from the NCI SEER Program included the following:

SEER*DMS Workgroup Updates

At the F2F meeting, SEER*DMS workgroups shared updates on the status of their projects. The Meaningful Use 2 Workgroup has been analyzing how electronic health records (EHRs) can enhance cancer surveillance data. From tests, EHRs have been found to supplement existing surveillance data, although data duplication was an issue. Algorithms are being tested to enhance EHR utility and reduce duplicate reports.

The Auto-consolidation Workgroup aims to increase automation in SEER*DMS and decrease the burden of manual work for data processing at registries. The workgroup defined a process for defining and testing new auto-consolidation rules. Rules were defined and implemented for two data items (Type of Reporting Source and Diagnostic Confirmation). The group reviewed and prioritized data items and continues to define rules for the SEER-required data items. This process includes a review of coding instructions and analysis to evaluate the agreement of outcomes between auto-consolidation and manual consolidation.

The Claims Workgroup aims to standardize processes for working with claims across registries, maximize automation, facilitate research access to source claims, and expand capture. Currently, the group is considering automation with a focus on approaches for identifying missing treatment information.

The Data Linkages Workgroup is working on rules to reduce linkage burden and approaches for handling unlinked data, backing out of automation, and handling unmatched patients. This group is working on a linkage to increase the completeness of Social Security Number information in registry data. This will help obtain better information on race/ethnicity, biomarkers, treatment, and recurrence, as well as enhance case finding and automation.

Other Topics

A productive discussion at the F2F Meeting focused on the use of pathology report data. Participants were divided into four small groups; each small group later shared a presentation on their discussion with the larger group. The small groups discussed: (1) coding and consolidation of pathology report data, (2) efficient use of pathology reports for case findings, (3) classifying pathology reports, and (4) providing data back to source facilities. Last, meeting participants discussed coding missing information in pathology reports and reporting of veterans’ cancer data.

More About SEER*DMS

SEER*DMS is a system that provides support for all core cancer registry functions, including data importation, editing, linkage, consolidation, and reporting. More than half of the registries that participated in the 2018 F2F meeting currently use SEER*DMS, and several more are in the process of transitioning to it. Overall, SEER*DMS aims to improve cancer surveillance cost efficiency, data quality and consistency, and knowledge sharing among registries, as well as reduce duplication of work.