Table 2

Studies assessing the association of birth weight and the risk for breast cancer
Type of study	Ref.	Year	Design	Cases	Controls (or cohort)	Country/place of study	Birthweight (g)	OR (95% CI)	Comments

Case-control studies	[42]	1988	PCC	153	461	USA	1,162–2,948	Referent	Matched analysis; P for trend = 0.41
							2,949–3,340	0.65 (0.33–1.26)
							3,341–4,451	0.76 (0.41–1.43)
	[12]^a	1992	LCC	458	1,197	Sweden	<2,500	1.18 (0.60–2.33)	Adjusted for age and birth date
							2,500–2,999	Referent
							3,000–3,499	1.29 (0.90–1.91)
							3,500–3,999	1.47 (1.00–2.18)
							≥ 4,000	1.23 (0.80–2.00)
	[13]	1996	NCC	550	1,478	USA	<2,500	0.56 (0.34–0.93)	Adjusted for age
							2,500–2,999	0.68 (0.47–0.99)
							3,000–3,499	0.71 (0.50–0.99)
							3,500–3,999	0.85 (0.59–1.22)
							≥ 4,000	Referent
	[14]	1996	PCC	922	1,194	USA	Age 21–45 years:		Adjusted for age, menopausal status, and maternal smoking; P for trend = 0.06 among both groups. The OR (95% CI) for birth weight ≥ 4,000 g among patients with early-onset breast cancer (≤ 30 years old) was 3.3 (1.0–11.0)
							<2,500	1.3 (0.9–2.0)
							2,500–2,999	Referent
							3,000–3,499	1.3 (1.0–1.7)
							3,500–3,999	1.2 (0.8–1.6)
							≥ 4,000	1.7 (1.1–2.5)
							Age 50–64 years:
							<2,500	0.9 (0.5–1.7)
							2,500–2,999	Referent
							3,000–3,499	1.1 (0.7–1.7)
							3,500–3,999	0.8 (0.4–1.3)
							≥ 4,000	0.6 (0.3–1.1)
	[32]	1997	NCC	1068	2,027	Sweden	<2,500	0.80 (0.50–1.26)	Adjusted for maternal age, socioeconomic status, parity, and pre-eclampsia or eclampsia, neonatal jaundice, severe prematurity, and twinship
							2,500–2,999	Referent
							3,000–3,499	1.00 (0.79–1.28)
							3,500–3,999	0.99 (0.77–1.26)
							≥ 4,000	1.04 (0.77–1.41)
	[33]	1998	PCC	510	436	USA	<2,500	1.2 (0.7–2.1)	Crude ORs
							2,500–2,999	Referent
							3,000–3,499	1.0 (0.7–1.5)
							3,500–3,999	1.0 (0.7–1.5)
							≥ 4,000	1.3 (0.7–2.3)
	[15]	2000	LCC	484	2,870	USA	<1,500	1.59 (0.61–4.11)	Crude ORs
							1,500–2,499	1.33 (0.94–1.90)
							2,500–3,499	Referent
							3,500–4,499	1.08 (0.87–1.34)
							≥ 4,500	3.29 (1.37–7.92)
	[34]	2001	LTCC	87	87	Sweden	<1,999	Referent	Matched analysis by conditional logistic regression
							2,000–2,499	1.6 (0.6–4.0)
							2,599–2,999	2.4 (0.9–6.2)
							≥ 3,000	1.6 (0.4–5.6)
								(P trend = 0.05)
	[43]	2001	LCC	319	768	USA	<2,500	1.4 (0.55–3.4)	Crude ORs. Higher birth weight (≥ 3,500 g) carried a marginal significantly higher risk for breast cancer (OR 1.76 [95% CI 0.90–3.35]) relative to lower birth weight (<3,500 g)
							2,500–3,750	Referent
							≥ 3,750	0.9 (0.50–1.6)
	[16]	2001	LTCC	90	90	Sweden	≤ 2,000	Referent	Crude ORs. Study subjects were women with opposite-sexed pair twins
							2,001–2,500	3.2 (0.8–12.6)
							2,501–3,000	3.5 (1–13)
							3,001–3,500	5.8 (1.3–25.7)
							≥ 3,501	12.1 (1.1–138.8)
	[35]	2002	PCC	2,088	2,187	USA	<2,500	1.10 (0.90–1.35)	Adjusted for age and residential regions (states)
							2,500–2,999	0.90 (0.70–1.10)
							3,000–3,499	Referent
							3,500–3,999	1.07 (0.90–1.30)
							4,000–4,499	0.89 (0.70–1.14)
							≥ 4,500	1.18 (0.90–1.51)
	[44]	2002	PCC	288	350	China	<2,500	0.9 (0.4–2.0)	Adjusted for age income, family history of breast cancer in first-degree relative, history of fibroadenoma, age at menarche, parity, and age at first live birth.
							2,500–2,999	Referent
							3,000–3,499	1.1 (0.8–1.6)
							3,500–3,999	0.8 (0.4–1.4)
							≥ 4,000	0.7 (0.4–1.4)
	[17]	2002	LCC	373	1,150	USA	<3,090	Referent	Adjusted for parity and age at first birth. P for trend = 0.02
							3,090–3,410	1.1 (0.8–1.5)
							3,420–3,720	1.2 (0.9–1.6)
							≥ 3,630	1.4 (1.1–1.9)
	[18]	2003	LCC	881	3,423	Denmark	<2,500	1.66 (1.00–2.51)	Adjusted for mother's marital status, maternal age, and birth order
							2,500–2,999	0.83 (0.60–1.10)
							3,000–3,499	Referent
							3,500–3,999	0.98 (0.80–1.17)
							≥ 4,000	1.25 (1.00–1.55)
	[19]	2004	NCC	89	238	Sweden	100 g increase	1.06 (1.00–1.12)	Adjusted for gestational age, birth year, and maternal hypertension/proteinuria
	[45]	2004	LCC	2471	9801	USA	<1,500	0.64 (0.40–1.11)	Adjusted for age and maternal age at first birth
							1,500–1,999	1.05 (0.70–1.68)
							2,000–2,499	1.02 (0.80–1.31)
							2,500–3,499	Referent
							3,500–3,999	0.97 (0.90–1.08)
							4,000–4,499	0.93 (0.80–1.11)
							≥ 4,500	0.69 (0.40–1.09)
	[36]	2004	PCC	196	167	USA	All subjects:		Adjusted for age, race and sampling fractions, body mass index, household income, and maternal age. Tertiles are race specific with cutpoints derived from controls. White women: <3,062, 3,062–3,458, >3,458 g; black women: <3,146, 3,146–3,488, >3,488 g. Restricted data using birth weight measured in pounds and ounces and participant delivered in a medical facility by a physician
							Lower tertile	1.0 (0.6–1.7)
							Central tertile	Referent
							Upper tertile	0.7 (0.4–1.2)
							White, restricted
							data:
							Lower tertile	1.1 (0.5–2.4)
							Central tertile	Referent
							Upper tertile	1.4 (0.6–2.0)
	[20]	2006	PCC	2,386	2,502	Poland	<2,500	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. Lower birth weight (<2,500 g) carries greater risk than birth weight of 2,500–4,000 g among women under 45 years old
							2,500–4,000	1.22 (0.92–1.62)
							>4,000	1.54 (1.08–2.19)
								(p-trend = 0.01)
	[37]	2006	PCC	1,166	2,105	USA	<2,495	1.19 (0.85–1.66)	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
							2,495–3,130	Referent
							3,131–3,855	0.97 (0.75–1.25)
							>3,855	1.03 (0.74–1.44)

Cohort studies	[21]	1999	LCohort	57	152,590	Sweden	<2,500	Referent	Standardization for sex, age, and age-specific incidence rate
							2,500–3,999	1.3 (0.6–2.4)
							4,000–4,499	1.2 (0.0–6.7)
							≥ 4,500	1.3 (0.7–2.3)
	[22]	2000	Cohort	37	2,221	UK	All ages		Adjusted for age. P for trend = 0.03 among premenopausal women
							<3,000	Referent
							3,000–3,499	1.05 (0.41–2.71)
							3,500–3,999	1.76 (0.72–4.33)
							≥ 4,000	2.02 (0.59–6.90)
							Premenopausal ages
							<3,000	Referent
							3,000–3,499	1.99 (0.40–9.86)
							3,500–3,999	3.26 (0.69–15.36)
							≥ 4,000	5.65 (0.95–33.84)
	[38]	2001	LCohort	177	3,447	Sweden	≤ 2,000	Referent	Crude hazard ratios
							2,001–2,500	1.4 (0.6–3.4)
							2,501–3,000	1.9 (0.8–4.3)
							3,001–3,500	1.5 (0.6–3.5)
							≥ 3,501	1.9 (0.7–5.0)
	[39]	2001	Cohort	62	1260	Sweden	≤ 3,000	Referent	Singleton only; adjusted for gestational age and cohort membership
							3,010–3,349	1.16 (0.47–2.87)
							3,350–3,590	1.65 (0.71–3.86)
							3,600–3,960	1.58 (0.67–3.72)
							≥ 4,000	1.57 (0.67–3.64)
	[23]	2003	LCohort	63	5,352	Sweden	<3,000	Referent	Crude ORs;P for trend = 0.01
							3,000–3,499	1.46 (0.60–3.43)
							3,500–3,999	2.09 (0.90–4.85)
							≥ 4,000	2.78 (1.10–7.15)
	[24]	2003	LCohort	2,334	106,504	Denmark	1,000 g increase	9 (0.02–17)%	Adjusted for age and calendar period. Additional adjustment for parity and age at first birth did not indicate confounding
	[25]	2003	LCohort	39	1483	Sweden	500–1,999	1.14 (0.70–1.85)	Standardized incidence ratio (expected/observed)
							2,000–2,999	0.71 (0.40–1.15)
							≥ 3,000	2.55 (1.03–5.25)
	[26]^a	2004	LCohort	2,074	91,601	Denmark	Median of each quintile		Adjusted for age and calendar period. No change in estimates when additionally adjusted for parity and age at first birth
							2.5	Referent
							3.0	0.98 (0.85–1.13)
							3.4	1.06 (0.93–1.20)
							3.6	1.05 (0.87–1.27)
							4.0	1.17 (1.02–1.33)
	[27]	2004	Cohort	59	2,176	UK	<3,000	Referent	Adjusted for age; P for trend = 0.03
							3,000–3,499	1.37 (0.34–5.47)
							3,500–3,999	2.18 (0.58–8.21)
							≥ 4,000	5.03 (1.13–22.47)
	[28]	2005	LCohort	311	16,011	USA	<3,040	Referent	Adjusted for year of birth
							3,040–3,310	1.4 (1.0–2.1)
							3,320–3,550	1.0 (0.6–1.5)
							3,560–3,830	1.3 (0.9–1.9)
							≥ 3,840	1.5 (1.0–2.2)
	[29]^a	2005	LCohort	367	5,346	Sweden	<50 years
							<3,000	Referent
							3,000–3,499	1.81 (0.77–4.26)
							3,500–3,999	2.66 (1.09–6.46)
							≥ 4,000	4.00 (1.49–10.72)
							≥ 50 years
							<3,000	Referent
							3,000–3,499	0.86 (0.62–1.19)
							3,500–3,999	1.06 (1.20–3.34)
							≥ 4,000	0.91 (0.57–1.46)
	[40]	2006	Cohort	97	5,847	USA	<3,000	0.98 (0.61–1.60)	Adjusted for age
							3,000–3,499	Referent
							≥ 3,500	1.09 (0.66–1.80)
	[30]	2006	Cohort	3,140	91,601	USA	Premenopause		Adjusted for age: P for trend = 0.019
							<2,495	0.69 (0.50–0.94)
							2,495–3,130	0.79 (0.64–0.97)
							3,131–3,810	0.76 (0.63–0.93)
							>3,810	Referent
							Postmenopause:		Adjusted for age: P for trend = 0.99
							<2,495	1.04 (0.88–1.23)
							2,495–3,130	1.00 (0.87–1.14)
							3,131–3,855	1.05 (0.93–1.20)
							>3,855	Referent
	[31]	2006	Cohort	209	1,024	USA	<2,500	0.9 (0.5–1.6)	Hazard ratio; adjusted for age at diagnosis, diagnosis year, stage at diagnosis, and birth order, with exception of birth order, which is adjusted for maternal age
							2,500–3,999	Referent
							≥ 4,000	1.8 (1.0–3.1)
								(P trend = 0.1)
	[41]	2007	Cohort	657	38,566	Sweden	<2,500	0.65 (0.43–0.99)	Adjusted for adult body mass index
							2,500–3,000	1.04 (0.86–1.25)
							>3,000	Referent

Cohort, cohort study; LCC, case-control study with linkage with population and cancer registry data; LCohort, cohort study with linkage with population and cancer registry data; LTCCS, twin case-control study by using linkage with birth and cancer registry data; NCC, nested case-control study in cohort; PCC, population-based case-control study. ^aThe numbers of cases and controls were not shown in the original article.
Park et al. Breast Cancer Research 2008 10:R8 doi:10.1186/bcr1850