Gilsenan A, Harris DH, Midkiff K, Wright J, Andrews E. Linking with multiple state cancer registries for safety surveillance study – Is it feasible? Poster presented at the 27th ICPE International Conference on Pharmacoepidemiology & Therapeutic Risk Management; August 18, 2011. Chicago, IL.

BACKGROUND: In 2009, the voluntary Forteo Patient Registry was implemented as one method to characterize the possible risk of osteosarcoma in patients taking teriparatide. Patient information will be linked annually for 12 years with state cancer registry data. Due to the individualized approval process for each state, launching a data linkage study with multiple state cancer registries at the same time can be challenging.

OBJECTIVES: To describe the feasibility of performing a data linkage in the United States with multiple state cancer registries and the Forteo Patient Registry.

METHODS: Patients are invited to participate in the Forteo Patient Registry through multiple routes of communication and are registered once a 1-page registration form and signed informed consent form are returned. In May 2009, cancer registries in all 50 states and the District of Columbia were invited to participate in the first annual linkage. A database was developed to track the recruitment process. All necessary applications and agreements for study approval were submitted to cancer registries. Registries that completed all local approval requirements and attended training on a standard linkage algorithm developed for this study were included in the first annual linkage in September 2010.

RESULTS: Forty-two of 51 cancer registries expressed interest in participating. Of these 42 registries, 28 required local institutional review board (IRB) approval and 14 deferred to the central IRB. At least one additional approval was required at 36 of the 42 registries. Twenty-seven registries met all approval requirements in time to participate in the first annual data linkage. The average time from submission of the first application to the date that all necessary approvals and agreements were in place was 94 days (range, 10–195 days). These registries (covering 70% of the adult US population) were successful in using the standard linkage algorithm.

CONCLUSIONS: Although there are substantial challenges in conducting a data linkage with multiple state cancer registries, including developing a customized recruitment strategy for each cancer registry, the results of the first linkage indicate that it is feasible.

Share on: