Table 6.

Exogenous hormone use and breast cancer risk stratified by hormone receptor expression

First author (reference)Country (study)Cases (n)Risk estimate (95% CI)*
Combination oral contraceptives
ER+ER−
    Althuis (28)United States (WISH)1,3751.6 (0.9-2.8)3.1 (1.6-5.9)
    Stanford (38)United States (CASH-GA)4580.8 (0.6-1.2)1.2 (0.8-1.8)
    McTiernan (39)United States (CASH-WA)2401.2 (0.7-1.9)0.8 (0.5-1.4)
    Cooper (36)Australia3800.9 (0.5-1.5)1.3 (0.7-2.6)
ER+/PR+ER−/PR−
    Cotterchio (24)Canada (ON)3,2760.9 (0.6-1.2)1.2 (0.8-1.9)
    McCredie (23)Melbourne (Australia)6181.1 (0.7-1.9)0.9 (0.5-1.6)
    Britton (29)United States (WISH)1,2121.2 (0.9-1.5)1.5 (1.0-2.1)
    Huang (32)United States (CBCS)7831.5 (0.8-2.7)1.1 (0.6-1.9)
Hormone Replacement Therapy
ER+ER−
    Chlebowski (13)WHI3091.4 (1.1-1.7)CHRT1.5 (0.8-2.9)CHRT
    Cooper (36)Australia3800.9 (0.5-1.4)NS1.8 (0.5-2.2)NS
    Hildreth (42)United States (CT)1480.9 (0.7-1.2)NS1.3 (0.9-1.8)NS
    Stanford (38)United States (CASH-GA)4581.0 (0.7-1.4)NS1.0 (0.7-1.3)NS
PR+PR−
    Chlebowski (13)WHI3091.5 (1.1-2.0)CHRT1.2 (0.8-1.8)CHRT
ER+/PR+ER−/PR−
    Li (25)Western WA9001.1 (0.8-1.5)ERT1.0 (0.6-1.7)ERT
2.0 (1.5-2.7)CHRT0.9 (0.5-1.8)CHRT
    Huang (32)United States (CBCS)7830.9 (0.6-1.2)NS0.6 (0.4-0.9)NS
    Potter (17)IWHS6101.1 (0.9-1.3)NS1.1 (0.7-1.8)NS
DES exposure in utero
ER+ER−
    Palmer (14)United States (DES)411.9 (0.8-4.5)0.4 (0.1-1.9)
  • NOTE: Table excludes Morabia et al. (33), who found that risk associated with oral contraceptive use was similarly elevated for ER+ and ER− breast cancers; however, ORs were not presented in the article.

  • * Ever-users compared with never-users. ERT, estrogen replacement therapy. CHRT, combination hormone replacement therapy; NS, formulation not specified. ORs were the reported risk estimates for case-control studies and RRs for cohort designs.

  • Among women ages 20-34 years. No significant association was found among women ages 35-54 years: ER+, OR, 1.0 (0.8-1.3); ER−, OR, 1.2 (0.9-1.6).

  • HRT modeled as a continuous variable; risk was estimated for every 50 mg/mo of use.