breast-cancer-research.com

Table 3
Studies assessing the association of birth order and the risk of breast cancer
Type of study	Ref.	Year	Design	Cases	Controls (or cohort)	Country/place of study	Birth order	OR (95% CI)	Comments

Case-control studies	[51]	1967	LCC-D	229	229	USA	1	-	The authors measured the mean value of birth weight instead of providing ORs (95% CIs). The mean difference between cases and matched controls was not significant (P > 0.2). They provided the frequency of each case and control in the tables and we calculated crude ORs
							2
							3
							4
							5
							6
							≥ 7
	[50]^a	1980	MCC	4339	12,760	USA, Japan, Slovenia, Athens, Taipei	1	Referent	The risks (point estimates) only by birth order were shown in the figure in the original article.
							2	0.93
							3	1.08
							4	0.99
							5	1.05
							6	1.07
							7	1.18
							≥ 8	1.02
	[42]	1988	PCC	153	461	USA	1	Referent	P for trend = 0.16
							2	0.92 (0.55–1.54)
							3	0.98 (0.58–1.72)
							4	0.69 (0.36–1.32)
							≥ 5	1.03 (0.60–1.79)
	[46]	1991	MCC	927	2,616	USA/Wales/Japan	All ages		Adjusted for age, study center, parity, age at first birth, age at menarche, height, body mass index, maternal age at birth, and menopausal status
							1	Referent
							2	0.91 (0.73–1.02)
							3	1.11 (0.87–1.27)
							≥ 4	1.09 (0.81–1.18)
							Premenopausal
							1	Referent
							≥ 2	0.76 (0.60–0.96)
	[12]	1992	LCC	458	1,197	Sweden	1	Referent	Adjusted for age and birth date
							≥ 2	1.00 (0.76–1.32)
	[47]	1994	PCC	2,414	9,138	USA	1	Referent	Adjusted for age at first birth and number of children
							2	0.90 (0.78–1.03)
							3	0.98 (0.84–1.14)
							4	0.86 (0.73–1.02)
							5	0.93 (0.78–1.11)
							6	1.02 (0.84–1.23)
							7	0.91 (0.73–1.14)
							≥ 8	0.88 (0.75–1.04)
	[14]	1996	PCC	1,129	1,393	USA	1	Referent	Adjusted for age, menopausal status, and maternal smoking; P for trend = 0.06 among both groups
							2	1.0 (0.7–1.4)
							≥ 3	0.8 (0.6–1.1)
	[32]	1997	NCC	1,068	2,727	Sweden	1	Referent	Adjusted for maternal age, socioeconomic status, parity, and preeclampsia or eclampsia, neonatal jaundice, severe prematurity, and twinship
							2	1.01 (0.83–1.22)
							≥ 3	1.01 (0.81–1.26)
	[15]	2000	LCC	481	2,863	USA	1	1.07 (0.84–1.35)	Crude ORs
							2–3	Referent
							4–5	1.06 (0.81–1.38)
							≥ 6	1.50 (1.06–2.13)
	[35]	2002	PCC	1,555	1,539	USA	1	Referent	Adjusted for age and residential regions (states)
							2	1.07 (0.88–1.30)
							3	1.07 (0.85–1.35)
							4	1.01 (0.77–1.31)
							5	0.66 (0.48–0.92)
							≥ 6	0.81 (0.62–1.08)
	[18]	2003	LCC	881	3,423	Denmark	1	Referent	Adjusted for mother's marital status, maternal age, and birth order
							≥ 2	1.01 (0.83–1.12)
	[36]	2004	PCC	854	785	USA	All subjects		Adjusted for age, race and sampling fractions, body mass index, hosehold income, maternal age
							1	Referent
							2–4	0.9 (0.7–1.1)
							≥ 5	1.0 (0.8–1.3)
							Born ≥ 1948
							1	Referent
							2–4	0.9 (0.6–1.4)
							≥ 5	0.6 (0.3–1.3)
	[48]^a	2005	MCC	24	34	Nigeria	≤ 3	Referent	Crude ORs
							≥ 4	1.50 (0.25–8.98)
	[20]	2005	PCC	1642	1,713	Poland	1	Referent	Adjusted for age, education, age at menarche, menopausal status and age at menopause, age at first full-term pregnancy, number of full-term pregnancies, family history of breast cancer among first-degree relatives, mammography screening, and current body mass index
							2	1.07 (0.91–1.24)
							3–5	0.99 (0.85–1.15)
							≥ 6	0.81 (0.61–1.06)
								P for trend = 0.81
	[37]	2006	PCC	1,166	2,105	USA	1	Referent	Adjusted for age (years), education (years), race, body mass index, history of breast benign disease, family history of breast cancer, lactation (months), age at menarche (years), age at first full-term pregnancy (years), age at menopause (years), parity
							≥ 2	1.27 (0.88–1.85)
Cohort studies	[31]	2006	Cohort	209	1,024	USA	1	Referent	Hazard ratio for breast cancer mortality: adjusted for age at diagnosis, diagnosis year, stage at diagnosis, and birth order, with exception of birth order, which is adjusted for maternal age
							2	0.2 (0.2–0.3)
							≥ 3	0.2 (0.2–0.3)
								P for trend < 0.01
	[49]^a	2001	Cohort	-	-	Sweden	Continuous scale	1.05 (1.01–1.10)

^aWe did not include these studies in the meta-analysis because they employed different categories or a continuous scale, or they did not provide the numbers of cases and controls in the original article. Cohort, cohort study; LCC, case-control study with linkage with population and cancer registry data; LCC-D, case-control study with linkage with population and cancer death certification data; MCC, multicenter case-control study; NCC, nested case-control study in cohort; PCC, population-based case-control study.
Park et al. Breast Cancer Research 2008 10:R8 doi:10.1186/bcr1850

Table 3