Aim:
Cancer Australia’s STaR program aims to bring together national data on stage at diagnosis with other national data to enhance our understanding of population-level patterns of stage, treatment, survival and recurrence to help inform policy and practice and ultimately improve cancer outcomes.
Given the recognised lack of national data on cancer stage at diagnosis, the development and validation of a methodology for the national standardised collection of cancer stage at diagnosis was required.
Methods:
In collaboration with all the jurisdictional Population Based-Cancer Registries (PBCRs), Business Rules (BRs) were developed for the top 5 adult incident cancers (prostate, breast, lung, colorectal, and melanoma) to provide clear guidelines for the abstraction of T-, N- and M-Stage values from commonly existing data sources available to the PBCRs. These values were then used to generate a nationally standardised stage measure - ‘Registry-derived stage’ (RD-Stage). The BRs were then applied to a sample set of 6,000 tumour records. For a subset of records, RD Stage was compared to stage derived from medical record review.
Results:
Derivation of RD-Stage using the BRs was achievable for very high proportion (86-96%) of four of the five tumour types. The concordance between RD-Stage stage and stage derived from medical record review was at least 80% across cancer types studied. The BRs have now been endorsed by the Australasian Association of Cancer Registries as a national standard and Cancer Australia is working with all the PBCR to collect RD-Stage for the top 5 incidence cancers in 2011.
Conclusion:
RD-stage can be derived using nationally endorsed BRs. This provides the capability to collect standardised national data on cancer stage at diagnosis. The first full year of RD-Stage derivation is expected to be completed in early 2018 and the linkage of stage to treatment and survival data is expected in 2019.