Methodological choices affect cancer incidence rates: a cohort study

Brooke, Hannah L.; Talbäck, Mats; Feychting, Maria; Ljung, Rickard

doi:10.1186/s12963-017-0120-x

Population Health Metrics

Table 1 Four definitions of the study population applied to hypothetical data from seven patients

From: Methodological choices affect cancer incidence rates: a cohort study

	Year of diagnosis			Number of breast cancer diagnoses counted for each patient for each different definition of the study population
	Prior to study period 1958–1999	Study period 2000–2010
Patient	1987	2002	2009	Aggr^a	Subtype^b	xNMSC^c	First ever^d
A^e	Breast	-	-	0	0	0	0
B^f	-	Breast	-	1	1	1	1
C^g	NMSC	Breast	-	1	1	1	0
D^h	Colon	Breast	-	1	1	0	0
Eⁱ	Breast	Breast	-	1	0	0	0
F^j	-	Breast	Colon	1	1	1	1
G^k	-	Breast	Breast	2	1	1	1

Cancer diagnoses prior to (1958–1999) and during (2000–2010) the study period, and number of breast cancer (BC) diagnoses counted using each definition of the study population (SP)
^aAggr, SP based on aggregate general population statistics
^bSubtype, SP excluding individuals with a previous diagnosis of the cancer subtype of interest
^cxNMSC, SP excluding individuals with any previous cancer diagnosis, except if the previous cancer was non-melanoma skin cancer
^dFirst ever, SP excluding individuals with any previous cancer diagnosis
^ePatient A: excluded from all methods as BC in 1987 was prior to the study period
^fPatient B: counted in all methods as there was no previous cancer diagnosis before BC in 2002
^gPatient C: BC in 2002 is counted in Aggr, Subtype and xNMSC, but not in First ever as the first ever cancer diagnosis was a non-melanoma skin cancer (NMSC) in 1987
^hPatient D: BC in 2002 is counted once in Aggr and once in Subtype, as it is the first subtype specific cancer. However, due to a previous colon cancer diagnosis in 1987 BC in 2002 is not counted in xNMSC or First ever
^IPatient E: BC in 2002 is counted once in Aggr. However, the patient is excluded from all other methods as the first BC occurred in 1987, which is not within the study period
^jPatient F: BC in 2002 is counted once in all methods (similar to patient B). The diagnosis of colon cancer in 2009 does not influence the incidence of BC
^kPatient G: BC in 2002 and 2009 are counted as two cancers in Aggr, as there are two records of new primary tumours in the study period. However, only the BC in 2002 is counted in the other methods as the first diagnosis

Back to article page

ISSN: 1478-7954

Contact us

General enquiries: journalsubmissions@springernature.com