Table 2. Relative risk associated with the use of progestagens according to duration, time since first use and time since last use, compared with never use (n=73 664). E3N cohort study (1990–2002).

	PYa	Cases	RRa,b	95% CIa
Never use	443 802	1510	1.00
Ever use	224 231	880	1.01	0.93–1.11
Duration (years) c
<1	83 449	272	0.90	0.79–1.03
1–2.5	63 932	272	1.00	0.85–1.17
2.5–4.5	43 712	174	1.10	0.96–1.26
>4.5	33 138	162	1.13	0.96–1.33
Trend (per year of use)			1.03	1.01–1.06
Interval since first use (years) c
<4	70 591	235	1.03	0.89–1.18
4–7.5	61 804	235	1.04	0.90–1.20
7.5–11.5	48 026	209	1.04	0.90–1.21
>11.5	43 810	201	0.95	0.81–1.11
Trend (per year since first use)			0.99	0.98–1.01
Interval since last use (years) c
Current use	68 697	235	1.14	0.97–1.33
<3	68 108	265	0.96	0.84–1.10
3–6	40 004	175	1.07	0.91–1.26
6–9.5	25 511	111	0.97	0.80–1.19
>9.5	21 863	94	0.92	0.75–1.14
Trend (per year since last use)			0.99	0.98–1.01

^aPY=person-year; RR=relative risk; CI=confidence interval.

^bAdjusted for BMI before and after menopause (<22/22-25/25-30/>=30), menopausal status°(premenopausal/artificial menopause/natural menopause), age at menopause (<48/48–52/>52), parity and age at first FTP (nulliparous/first FTP at age < 30/first FTP at age ⩾30, num=1/first FTP at age ⩾−30, num>1), age at menarche (<13/13–15/>15), familial history of breast cancer in sisters, mother, children (no/1/more than 1), familial history of breast cancer in other relatives (yes/no), personal history of benign breast disease °(yes/no), personal history of benign uterin or ovarian disease (yes/no), use of oral contraceptive (never/current or < 5 years after stop/ > 5 years after stop), use of HRT °(No/oestrogen alone/oestrogen+progestin/oestrogen+progesterone/others) and previous mammography (yes/no) °Time-dependent variables.

^cCategories correspond to quartiles.