Background: Monitoring population-based cancer survival is an essential component in the evaluation of cancer control, but subject to an inherent delay in the reporting of the most recent survival estimates with traditional techniques of analysis.
Methods: We examined survival trends between the years 2000 and 2004 for 20 common cancers based on follow-up data from 12 cancer registries from diverse areas of Europe using model-based period analysis techniques.
Results: Between 2000 and 2004, marked rises were seen in 5-year relative survival amongst patients with prostate, breast and colorectal cancer, which were statistically significant in 10, 8 and 7 of the 12 participating cancer registries, respectively. For cancer sites amenable to effective early detection and treatment, major geographical differences in patient prognosis still persisted, with a lower survival generally observed in Eastern European countries.
Conclusion: Model-based period analysis enables the timely monitoring of recent trends in population-based cancer survival. For colorectal and breast cancers, the identified rises in survival are probably (at least partly) explained by the improvements in clinical care and the management of the disease. Nevertheless, persisting geographic differences do point to the potential for a further reduction in the burden of cancer throughout Europe, towards which improvements in diverse areas of care, including secondary prevention, access to advances in treatment as well as subspecialisation and regionalisation of oncologic care may all contribute.