Table 3. Relative risk of premenopausal breast cancer in relation to (A) perceived body fatness at age 10 years and (B) participation in sports at young ages, with and without adjustment for body fatness, participation in sports at different ages and menstrual characteristics.

	Thin		Average		Plump
(A) Model/perceived body fatness at age 10 years	RRa	95% FCIb	RRa	95% FCIb	RRa	95% FCIb
Multivariate-adjustedc	1.05	0.94–1.17	1.00	0.89–1.12	0.85	0.72–0.99
Multivariatec+participation in sports at all young ages	1.05	0.93–1.17	1.00	0.89–1.12	0.84	0.72–0.99
Multivariatec+participation in sports at diagnosis/pseudodiagnosis	1.04	0.93–1.17	1.00	0.89–1.12	0.84	0.72–0.99
Multivariatec+menstrual characteristicsd	1.06	0.95–1.20	1.00	0.89–1.12	0.84	0.72–0.99

	0–1 h week−1		2–3 h week−1		4+ h week−1
(B) Model/average duration of sports participation at ages 12–30 years	RRe	95% FCIb	RRe	95% FCIb	RRe	95% FCIb
Multivariate-adjustedc	1.00	0.90–1.13	0.99	0.89–1.11	1.01	0.88–1.16
Multivariatec+body fatness at age 10	1.00	0.88–1.13	0.99	0.88–1.11	1.00	0.87–1.16
Multivariatec+body mass index (BMI) at age 20 years	1.00	0.88–1.13	0.98	0.87–1.10	1.00	0.86–1.14
Multivariatec+BMI at diagnosis/pseudodiagnosis	1.00	0.88–1.13	0.99	0.88–1.11	1.01	0.87–1.16
Multivariatec+menstrual characteristicsd	1.00	0.88–1.13	0.98	0.88–1.10	1.00	0.87–1.15

^aRelative risk, compared with women who reported to be of average body fatness at age 10 years.

^bFloated confidence intervals (see Material and Methods).

^cStratified by age and recruitment region, adjusted for parity, age at first birth, height, use of oral contraceptives and alcohol consumption.

^dAdditional adjustments for age at menarche, time from menarche to onset of regular cycles and irregular menstrual cycles at ages 20–30 years.

^eRelative risk, compared with women who reported to participate 1 h or less per week in sports at young ages.