Table 3

Multivariate* relative risk (95% CI) of breast cancer by quartile of plasma prolactin concentration among postmenopausal women in the Nurses’ Health Study by invasiveness, type, tumor size, and receptor status

Prolactin ConcentrationsP for trendP for heterogeneity§
≤7.4 ng/mL>7.4–9.4 ng/mL>9.4–12.3 ng/mL>12.3 ng/mL
In situ 1.0 (ref.)0.850.800.960.840.11
(n = 115 cases)(0.49, 1.46)(0.46, 1.39)(0.57, 1.63)
Invasive1.0 (ref.)0.841.271.410.003
(n = 722 cases)(0.63, 1.12)(0.97, 1.67)(1.08, 1.86)
Ductal1.0 (ref.)0.851.191.380.0070.43
(n = 588 cases)(0.62, 1.15)(0.89, 1.60)(1.04, 1.85)
Lobular1.0 (ref.)0.861.691.760.11
(n = 93 cases)(0.42, 1.76)(0.90, 3.14)(0.95, 3.26)
Tumor size ≤2 cm1.0 (ref.)0.871.221.350.010.24
(n = 531 cases)(0.63, 1.19)(0.91, 1.65)(1.00, 1.83)
Tumor size >2 cm1.0 (ref.)0.741.351.660.03
(n = 162 cases)(0.43, 1.28)(0.84, 2.19)(1.04, 2.64)
ER+/PR+1.0 (ref.)0.871.521.78<0.0010.03
(n = 397 cases)(0.60, 1.26)(1.08, 2.13)(1.28, 2.50)
ER−/PR−1.0 (ref.)0.490.730.760.28
(n = 96 cases)(0.26, 0.92)(0.41, 1.28)(0.43, 1.32)
ER+/PR−1.0 (ref.)1.761.741.940.12
(n = 91 cases)(0.90, 3.47)(0.88, 3.43)(0.99, 3.78)
  • Abbreviations: ER, estrogen receptor; PR, progesterone receptor; BMI, body mass index.

  • * Adjusted for BMI at age 18, weight change from age 18 to blood draw, family history of breast cancer, age at menarche, age at first birth/parity, age at menopause, and matching factors.

  • Too few ER−/PR+ cases (n = 18) were available to analyze separately.

  • Determined using continuous, log-transformed prolactin concentrations.

  • § Determined using polytomous logistic regression and the likelihood ratio test, comparing a model constraining relative risks to be the same across all case groups versus a model allowing the relative risks to differ across case groups.