Background: In this paper we investigated whether the capture-recapture method is useful for a cancer registry to monitor its completeness of case ascertainment on a routine basis.
Methods: The capture-recapture method was used to estimate the completeness of case ascertainment in three regional cancer registries in the Netherlands, which are based on case finding by pathology laboratories and hospitals.
Results: Completeness was estimated to be 98.3%. The estimate of completeness was dependent on age and cancer site, with lower estimates of completeness for skin cancer and lymphatic and haematopoietic malignancies and for the age group > or = 75 years.
Conclusions: A major drawback of the capture-recapture method is its inability to estimate the number of cases that are not routinely notified to the registry by one or both notification sources. Another limitation is the lack of statistical power to detect incompleteness in an early stage. It is concluded that the capture-recapture method is not useful for everyday surveillance of completeness in cancer registration.