Table 3.

Cox proportional hazards regression models of the effect of reproductive and cancer-specific prognostic factors on risk of death after epithelial ovarian cancer diagnosis

CharacteristicHR* (95% CI)
LOC
    Low1.00 (reference)
    Medium1.09 (0.76-1.56)
    High1.67 (1.20-2.33)
    P trend0.002
Parity
    Nulliparous1.00 (reference)
    1-20.87 (0.61-1.23)
    ≥30.92 (0.64-1.33)
    P trend0.733
Use of oral contraceptives
    Yes0.87 (0.65-1.17)
    No1.00 (reference)
Breast-feeding history (mo)
    Nulliparous1.00 (reference)
    Never breast-fed0.97 (0.68-1.37)
    1-30.76 (0.48-1.19)
    ≥40.84 (0.54-1.30)
    P trend§0.336
Age at menarche (y)
    <121.51 (1.02-2.24)
    12-131.18 (0.86-1.64)
    ≥141.00 (reference)
    P trend0.042
Menopause status
    Premenopausal1.00 (reference)
    Postmenopausal1.17 (0.84-1.65)
Age at first birth (y)
    Nulliparous1.00 (reference)
    ≤190.82 (0.53-1.29)
    20-240.86 (0.60-1.24)
    25-290.85 (0.55-1.31)
    ≥301.91 (1.01-3.61)
    P trend§0.163
Hysterectomy
    Yes0.80 (0.51-1.25)
    No1.00 (reference)
Tubal sterilization
    Yes1.35 (0.91-2.01)
    No1.00 (reference)
Age at diagnosis (y)
    20-401.00 (reference)
    41-501.76 (1.22-2.55)
    51-542.54 (1.72-3.75)
    P trend0.001
Stage of disease
    Localized1.00 (reference)
    Regional/distant3.73 (2.63-5.29)
Histology
    Serous1.00 (reference)
    Mucinous0.98 (0.52-1.81)
    Endometriod1.40 (1.00-1.97)
    Other1.34 (0.94-1.93)
  • NOTE: Risk of death is defined as death from any cause after ovarian cancer diagnosis, through December 31, 1997; data are from the CASH study, 1980 to 1982.

  • * Adjusted for age at diagnosis and stage of cancer.

  • Age-specific cut points for low, medium, and high LOC (respectively) were as follows: ages 20 to 40: ≤174, 175 to 247, and 248+; ages 41 to 50: ≤377, 378 to 419, and 420+; ages 51 to 54: ≤424, 425 to 451, and 452+.

  • Yes = ≥3 consecutive mo; no = never used or used <3 consecutive mo.

  • § Among purous women.

  • Other includes other epithelial (n = 8), clear cell tumors (n = 35), and tumors for which no review was done by the CASH panel (n = 41).