Abstract
The incidence data (1991. 7. 1$\sim$1992. 6. 30) from the Implementation Study of Seoul Cancer Registry (ISSCR) were evaluated in terms of its completeness and validity. Two indicators for the completeness, Mortality/Incidence ratio (M/I ratio) and Age-specific Incidence Curve, showed fairy good registration throughout the age-sex specific strata, except the strata aged over 75 years old. The strata had very high M/I ratio (over 100%) and decreasing pattern of incidence, which suggested incomplete registration of cancer in this group. The active surveillance by a ISSCR staff improved the registration rate especially among elderlies. From the site specific M/I ratio, we found that liver cancer had oddly high M/I ratio. Since this high M/I ratio of liver cancer appears consistently in other reliable cancer registries, it is more like to be due to the high fatality of it rather than incomplete registration. The validity of the incidence data was assessed by three indicators; Histological Verification (HV%), Primary Site Unknown (PSU%), and Age Unknown (Age UNK%). The average HV% were 77% for men and 85% for women, which were slightly lower than those of other reliable cancer registries. This low HV% might be due to the considerable size of relative frequency of liver cancer in Korea, regarding the fact that the diagnosis of liver cancer is made mostly by non-biopsical radiologic methods (CT, Ultrasono, Angiography, MRI etc.). The level of PSU% and Age UNK% were in acceptable range, but not low enough, especially in terms of Age UNK%. Although ISSCR data had acceptable quality in general, it is needed to have more hospitals participate in the registry surveillance, to make registery data merged with death certificate data regulary, and educate the registration stans to be more competent and dedicated.
(Department of Preventive Medicine, Seoul National University College of Medicine)