Methodological Notes – Cancer Incidence Module

Updated: April 2012

Concepts and Definitions

Cancer is one of the few diseases among the chronic, non-communicable diseases for which population-based registries have been in existence for decades, at least among the developed countries. Cancer registration is complex and requires an adequate health care infrastructure to maintain and sustain it. Registries differ in their comprehensive coverage of cases and the quality of the data captured. CircHOB reports age-standardized incidence rates (ASIR) of cancer, both for all cancers combined, and several more common sites, including lung, colon and rectum (the last two tend to be combined in some countries) in both sexes, prostate in men, and breast and cervix in women. There are also very rare cancers (eg. nasopharyngeal and salivary glands) for which some circumpolar indigenous people such as the Inuit are particularly at high risk. For information specific to the circumpolar Inuit, please consult the report of the Circumpolar Inuit Cancer Review Working Group [Part I] and [Part II].

For age-standardization, a hypothetical standard population known as the “world standard population” is widely used in international comparisons of cancer incidence rates, developed by the International Agency for Research on Cancer (IARC). The age distribution in a population of 100, 000 people is as follows:

Age group Population
0-4 12, 000
5-9 10, 000
10-14 9, 000
15-19 9, 000
20-24 8, 000
25-29 8, 000
30-34 6, 000
35-39 6, 000
40-44 6, 000
45-49 6, 000
50-54 5, 000
55-59 4, 000
60-64 4, 000
65-69 3, 000
70-74 2, 000
75-79 1, 000
80-84 500
85+ 500
Total 100, 000

Note that in CircHOB, a different standard population, known as the “European standard population” developed by Eurostat is used to compare age-standardized mortality rates – see the Methodological Notes: Mortality Module. The method for age-standardization is discussed in the Methodological Notes for the mortality module. The age-standardized incidence rates (ASIR) presented can only be compared among themselves, or with any of the published rates in IARC or other scientific publications that use the same world standard population. They cannot be compared with published rates by some national agencies that use that country’s population from a specific year as the standard. Increasingly many national agencies do provide different sets of ASIR using their own national population and the IARC world standard population as standards.

Among circumpolar countries, the Nordic countries collaborate in the Association of Nordic Cancer Registries (NORDCAN) database. Globally, the IARC maintains the GLOBOCAN database and also publishes periodically Cancer Incidence in Five Continents from which comparative cancer incidence data from a larger group of countries and registries can be obtained.

Because of the large amount of data needed to compute ASIRs by cancer site, the raw data and the different steps in the computation are not shown in CircHOB.

Data Sources and Limitations

United States

United States national and Alaska state data were obtained from the National Program of Cancer Registries maintained by the Centers for Disease Control and available from the CDC Wonder website. Note that the Alaska data in CircHOB refer to all Alaskans in the state. Alaska also has a unique resource in its Alaska Native Tumour Registry, initiated under the United States Indian Health Service, and currently maintained by the Alaska Native Tribal Health Consortium, with data going back to 1969. A comprehensive 35-year review Cancer in Alaska Natives 1969-2003 is available here.

Canada

Statistics Canada maintains the Canadian Cancer Registry. Age-sex specific data for the whole of Canada are available from CANSIM Table 103-0550, from which ASIR using the world standard population can be calculated. However, for the small population of Yukon, Northwest Territories and Nunavut, many age-sex categories have fewer than 5 cases, which for confidentiality reasons are not reported to the public by Statistics Canada. Special requests need to be made to Statistics Canada, which calculates the ASIR standardized to the world population without releasing the intermediate steps.

Greenland, Faroe Islands, Iceland, and Denmark

Denmark and Iceland data are from NORDCAN on the web (v5.0). Faroe Islands and Greenland data are from the NOMESCO spreadsheet. Only data from the most recent 5-year period are posted on the website; earlier periods are not archived.

National data for the Nordic countries are available from NORDCAN on the web, v5.0. Regional data for these countries are available for the 2000-2004 period from PC-NORDCAN v2.4, updating for which has been discontinued. For the 2005-09 period, regional data have to be obtained from each country’s respective national cancer registry:

Norway

Table 14 in the report Cancer in Norway 2009.

Sweden

Socialstyrelsen’s statistical database.

Finland

The Finnish Cancer Registry. Both the Finnish Cancer Registry and NORDCAN report regional data by hospital districts. For CircHOB, data for Oulun lääni are obtained by combining the hospital districts of Pohjois-Pohjanmaa and Kainuu, while those of Lapin lääni are from the Lansi-Pohja and Lappi hospital districts.

No registry data are available for Russia nationally, although Public Health in Russia reports “morbidity” data for cancers, neither age-standardized nor age-specific rates are available for the country as a whole or the regions.