BACKGROUND: Despite use of the North American Association of Central Cancer Registries' indicator for assessing completeness of case ascertainment in population-based cancer registries, little has been published about its methodology, usefulness and accuracy in Canada. DATA AND METHODS: Canadian cancer incidence, cancer mortality, and population census data were used to quantify case completeness in 2007. Two indicators (I₁ and I₂) that expressed the observed age-standardized incidence rate relative to the expected rate were calculated. The assumption of stable age-standardized sex- and cancer-site-specific incidence-to-mortality rate ratios across regions was assessed. Associations between I₁, I₂ and simpler indicators of completeness were examined. RESULTS: The assumption of stable age-standardized sex- and cancer-site-specific incidence-to-mortality rate ratios across regions was not consistently supported—substantial regional differences emerged. I₁ was strongly correlated with I₂ (r=0.93, n=315, p<0.0001), and both were most strongly and consistently associated with the age-standardized incidence-to-mortality rate ratio. The frequency of undercoverage did not increase consistently with expected case-finding difficulty. INTERPRETATION: The age-standardized incidence-to-mortality rate ratio may provide a less complicated method of identifying undercoverage.
BACKGROUND: Despite use of the North American Association of Central Cancer Registries' indicator for assessing completeness of case ascertainment in population-based cancer registries, little has been published about its methodology, usefulness and accuracy in Canada. DATA AND METHODS: Canadian cancer incidence, cancer mortality, and population census data were used to quantify case completeness in 2007. Two indicators (I₁ and I₂) that expressed the observed age-standardized incidence rate relative to the expected rate were calculated. The assumption of stable age-standardized sex- and cancer-site-specific incidence-to-mortality rate ratios across regions was assessed. Associations between I₁, I₂ and simpler indicators of completeness were examined. RESULTS: The assumption of stable age-standardized sex- and cancer-site-specific incidence-to-mortality rate ratios across regions was not consistently supported—substantial regional differences emerged. I₁ was strongly correlated with I₂ (r=0.93, n=315, p<0.0001), and both were most strongly and consistently associated with the age-standardized incidence-to-mortality rate ratio. The frequency of undercoverage did not increase consistently with expected case-finding difficulty. INTERPRETATION: The age-standardized incidence-to-mortality rate ratio may provide a less complicated method of identifying undercoverage.
Entities:
Keywords:
SEER Program; cancer; data collection; incidence; mortality; neoplasms; registries
Authors: Rachel S D Lane; Ladislav Tomášek; Lydia B Zablotska; Estelle Rage; Franco Momoli; Julian Little Journal: Int Arch Occup Environ Health Date: 2019-02-08 Impact factor: 3.015