Nordic Cancer Registries – an overview of their procedures and data comparability

Figures & data

Table 1. Nordic cancer registries, administrative facts and quality issues. For questions on how to contact the Cancer Registry; how to apply information from the Cancer Registry (with links to forms and instructions); links to documents on data protection principles and to publications on data quality; please see the more detailed web version of this table.

Variable	Denmark	Finland	Iceland	Norway	Sweden
Official name	The Danish Cancer Registry	Cancer Society of Finland, Finnish Cancer Registry — Institute for Statisticaland EpidemiologicalCancer Research	The Icelandic Cancer Registry, the Icelandic Cancer Society	Cancer Registry of Norway. Institute of Population-based Cancer Research	Swedish Cancer Registry, the Swedish National Board of Health and Welfare
Founded	1942	1952	1954	1952	1958
Earliest/first complete year of cancer registration	1943/1943	1930/1953	1911/1955	1952/1953	1958
Percentage of death certificate only (DCO) cases (NORDCAN data 2009–2013)	0.3%	1.2% in 2009–2011 (tracing of DCI data for 2012+ delayed)	0.3%	1.1%	—
Percentage of microscopically verified (MV) cases (NORDCAN data 2009–2013)	95%	93%	95%	94%	98%

Table 2. Data sources of the Nordic cancer registries (routine data collection).

	Denmark	Finland	Iceland	Norway	Sweden
Public hospitals	Yes (received automatically from Danish Patient Register 2004+)	Yes (automatic reporting from some hospital registers since late 1980s, still mostly manual)	Registrars have direct access to electronic medical records for completing registration and information on stage	Yes (manual reporting, mostly electronically)	Yes
Private clinicians	Yes	Yes	Yes, only for prostatecancer, manual reported on paper	Yes (manual reporting, mostly on paper)	Yes
Dentists	Yes	Yes	No	No	No
Inpatient registry	Yes	No^a	2000+ (automatic reporting)	1998+ (in- and out-patients)	No
Laboratories, pathological reports	Yes (for cases coded 2004+)	Yes (automatic reporting from almost all laboratories since late 1980s)	Yes (automatic reporting)	Yes (automatic reporting, both paper and electronically)	Yes
Laboratories, hematological reports	No	Yes	Yes (automatic reporting)	No (only samples sent to pathological laboratories)	Yes
Laboratories, cytological reports	No	Yes	Yes (automatic reporting)	Yes (automatic reporting, both paper and electronically)	Yes
Death certificates	Yes	Yes (automatic reporting from Statistics Finland 1981+)	Yes (automatic reporting from Statistics Iceland until 2009 and from Directorate of Health 2010+)	Yes (automatic reporting, data file and picture files of death certificates)	No
Radiotherapy data (from all machines)	No	No	No	1997+^b	No

^a Since 2015 the hospital discharge data has been available for cross linkage with the Cancer Registry from year 1996 and can be used to complement possibly missing cases and to confirm DCI-cases.

^b Data items: region being treated; diagnosis (ICD-10); intention of treatment; date for start of treatment; number of days for treatment; date for end of treatment; total dose; total number of fractions.

Figure 1. Time trends of incidence of leukemia in men in four Nordic countries. Three-year floating averages of age-standardized rates (World) 1943–2104. In 1978, 18% of leukemia cases were missing from the Swedish Cancer Registry [2]; the proportion now may be smaller.

Figure 2. Incidence of pancreatic cancer in men in in the Nordic countries in 2009–2013, by age. The proportion of malignancies detected via the Cause of Death Register is particularly high for cancers not diagnosed or treated in hospital, e.g., cancers with a poor prognosis in the elderly such as pancreatic cancer.

Table 3. Numbers of cancers and other disease entities collected by the Nordic cancer registries, 2009–2013.

Disease/disease group	Number of registered cases in 2009–2013
	Denmark	Finland	Iceland	Norway	Sweden
1. NORDCAN cancer entities (NORDCAN 7.3; [11])	186,141	150,441	7360	146,007	272,658
Disease entities registered in the national registries but not included in NORDCAN tabulations^a
2. Cancers registered in the national registries, but not included in NORDCAN
2.1 Multiple cancers in same organ group^b	1823	2089	[147]	[4309]	24,714
2.1.1 Breast	555	1261	[81]	[1240]	6983
2.1.2 Skin, melanoma	0	106	0	[687]	1181
2.1.3 Skin, non-melanoma, excluding basal cell carcinoma	0	576	0	[165]	7846
2.1.4 Colon	390	31	[21]	[814]	1728
2.1.5 Bladder, etc.	749	91	[26]	[423]	820
2.2 Polycythaemia vera	678	376	[14]	[349]	793
2.3 Skin, basal cell carcinoma	58,208	40,866	[1671]	[40,606]	208,985
2.4 Other excluded categories	2608	43	0	89
3. Premalignant or borderline diseases registered in the national registries but not included in NORDCAN
3.1 Ovary, borderline cancer	[880]	767	[39]	[788]	810
3.2 Cervix uteri, precancerous lesions/in situ	26,835/2648	5872	[25]	[60,674]	14,568 (CIN III)
3.3 In situ cancers, excluding cervix uteri	2516 (only breast)	2732 (only breast) [423]	[618] [breast:100;melanoma: 96;other skin: 422]	[23,699]	[20,872]
3.4 Other	363			16
4. Other registrations	0	[636]	[23]	[305]

^a The numbers in brackets indicate that these numbers are not shown in the national routine statistics publication.

^b See IARC/IACR multiple coding rules, organ grouping in multiple coding (http://www.iacr.com.fr/images/doc/MPrules_july2004.pdf).

Figure 3. Time trends of incidence of non-melanoma skin cancer in men in the Nordic countries as reported by NORDCAN. Age-standardized rates (World) 1943–2014. Before 1978, Danish Cancer Registry could not separate basal cell carcinoma from the other non-melanoma skin cancers.

Figure 4. Time trends of incidence of bladder cancer in men in the Nordic countries. Five-year floating averages of age-standardized rates (World) 1943–2014. Part of the differences is explained by registration: Denmark has more often than the others included urothelial tumors of grades 1–4, unknown grade and ‘papilloma’ in the bladder in the incidence.

Figure 5. Time trends of incidence of prostate cancer in the Nordic countries. Age-standardized rates (World) 1943–2014. There was a steep increase in the incidence after global introduction of the PSA screening test in the early 1990s in all Nordic countries but in Denmark. In Denmark, the increase in incidence after 2004 partly reflects inclusion of pathology register information.

Figure 6. Geographical variation of incidence of thyroid cancer among women in the Nordic countries in 2004–2010 [39]. The red area in mid Finland is consequence of excessive testing with ultrasound.

Figure 7. Time trends of cervical cancer in the Nordic countries in the screened age categories 30–69 years. Truncated age- standardized rates (World) 1943–2014. The downward trend in incidence started at different times depending on the start of national organized cervical cancer screening program.

Figure 8. Time trends of female breast cancer in age 50–59 years in four Nordic countries. Truncated age- standardized rates (World) 1943–2014. Start of countrywide organized mammography screening varied from 1986 (Finland) to 2007–2009 (Denmark); this caused major increases in the incidence rates of the screened age category.

Figure 9. Time trends of incidence of basal cell carcinoma (BCC) among men in the Nordic countries in 1953–2015. Age-standardized rates (World), only the first BCC counted for each person. Registration practices of BCC – which has normally not been counted as true cancer in the routine statistics – have varied between countries and time periods.

Table 4. Variables on cancer patients registered by the Nordic cancer registries.

Variable	Denmark	Finland	Iceland	Norway	Sweden
Personal identity code (PIC)	For persons resident 1968+	For persons resident 1967+	For persons resident 1 Dec 1965+	For persons resident 1964+	Yes
Date of birth	Yes	Yes	Yes	Yes	Yes
Sex	Yes	Yes	Yes	Yes	Yes
Municipality of residence	1968+	Yes	Yes	1971+	Yes
Occupation	Until 2003	Yes	No	No	No
Education	No	Yes	No	No	No
Socio-economic position	No	Yes	No	No	No
Date of immigration	No	Yes	No	2015+	Yes
Country of origin	No	No	No	No	No
Country of birth	No	Yes	Yes	No	Yes
Date of emigration	Yes	Yes	No	Yes	Yes
Vital status	Yes	Yes	Yes	Yes	Yes
Date of death	Yes	Yes	Yes	Yes	Yes
Underlying cause of death	No	Yes	Yes (only for those who die from cancer)	Yes	Yes
Other causes of death	No	Yes	2003+ (only if the cause is cancer)	Yes	No
Incidental Autopsy	Yes	Yes	Yes	Yes (only for medical autopsies)	Yes

Table 5. Variables related to each cancer registered by the Nordic cancer registries.

Variable	Denmark	Finland	Iceland	Norway	Sweden
Date of diagnosis	Before 2004 only month and year	Before 2014 only month and year	Yes	Yes	Yes
Topography/primary site	Yes as ICD-10	Yes	Yes	Yes	Yes
Morphology/histology	Yes	Yes	Yes	Yes	Yes, with limitations before 1993
Behavior/malignancy	Yes	Yes	Yes	Yes	Yes
Method of confirmation	Yes	Yes	Yes	Yes	Yes
Stage at diagnosis	Before 2004	Yes	No	Yes	No
TNM	2004+	Incomplete	2010+ for selected sites	1993+ for selected sites; incomplete	2004+
Later metastases, recurrences	No	No	No	No	No
Treatment	Before 2004 (2004+ available via linkage to the Hospital Discharge Registry)	Yes, not complete	1999+, only for prostate cancer	Yes	No
Symptoms	No	No	No	Until 2011, incomplete; for melanomas, symptoms are still registered	No
Other	More detailed information may be available from the registries of 22 clinical cancer groups (www.DMCG.dk)	Code for evaluating cancer mortality (died of this cancer; died of other causes)		More detailed information available from 9 quality registries of the Cancer Registry 1985+ (WebTable 6)	More detailed information available from >25 clinical INCA registries (WebTable 7)

Yellow background: This information is not directly available (in a qualitatively accurate manner) but may be derived with a modest extra effort.

Mattsson B, Wallgren A. Completeness of the Swedish Cancer Register. Non-notified cancer cases recorded on death certificates in 1978 . Acta Radiol Oncol. 1984;23:305–313.

 PubMed Web of Science ®Google Scholar

Engholm G, Ferlay J, Christensen N, et al. NORDCAN: cancer incidence, mortality, prevalence and survival in the Nordic countries, Version 7.3 (08.07.2016). Association of the Nordic Cancer Registries. Danish Cancer Society. 2016. Available from http://www.ancr.nu, (accessed September 2017).

 Google Scholar

Patama T, Pukkala E. Small-area based smoothing method for cancer risk mapping. Spat Spatiotemporal Epidemiol. 2016;19:1–9.

 PubMed Web of Science ®Google Scholar