Skip to main content
. Author manuscript; available in PMC: 2012 Nov 28.
Published in final edited form as: Cancer Epidemiol Biomarkers Prev. 2011 Apr 5;20(5):900–911. doi: 10.1158/1055-9965.EPI-10-1325

Table 4.

Associations Between Selected Reproductive and Menopausal Risk Factors and Lung Cancer by Smoking Status Among 185,017 Women, NIH-AARP Diet and Health Study, 1995–2006

Characteristic # Lung Cancers (n=276) Never smokers
# Lung Cancers (n=1,509) Former smokers
# Lung Cancers (n=1,727) Current smokers
RR* 95% CI RR* 95% CI RR* 95% CI
Age at menarche (years)
 <11 23 1.00 (referent) 99 1.00 (referent) 128 1.00 (referent)
 11–12 121 0.76 0.48 1.19 639 0.94 0.76 1.17 731 0.92 0.76 1.11
 13–14 109 0.66 0.42 1.04 620 0.89 0.72 1.10 694 0.87 0.72 1.05
 15+ 21 0.55 0.30 1.00 143 0.91 0.70 1.18 167 0.84 0.67 1.06
 Missing 2 0.95 0.21 4.29 8 1.15 0.55 2.42 7 0.95 0.44 2.09
p-value for trend 0.03 0.26 0.08
Age at first live birth among parous women (years)
 <20 38 1.00 (referent) 295 1.00 (referent) 432 1.00 (referent)
 20–24 126 0.98 0.74 1.30 677 0.99 0.88 1.12 762 0.97 0.87 1.08
 25–29 53 0.90 0.63 1.28 266 1.00 0.86 1.17 227 0.87 0.74 1.01
 30+ 10 0.51 0.26 0.99 69 0.83 0.64 1.07 75 0.96 0.75 1.22
 Missing 2 0.49 0.12 2.08 20 1.25 0.78 2.00 19 1.02 0.64 1.63
p-value for trend 0.22 0.18 0.04
Type of and age at menopause
Natural menopause
 <45 years 17 1.01 0.60 1.72 143 1.37 1.13 1.67 212 1.29 1.09 1.52
 45–49 years 34 0.87 0.58 1.30 270 1.20 1.02 1.40 337 1.12 0.96 1.29
 50–54 years 78 1.00 (referent) 362 1.00 (referent) 377 1.00 (referent)
 55+ years 23 1.11 0.70 1.77 76 0.89 0.69 1.14 63 1.00 0.77 1.31
p-value for trend 0.56 <0.0001 <0.01
Surgical menopause, bilateral oophorectomy
 <40 years 14 0.94 0.53 1.67 146 1.64 1.35 2.00 157 1.18 0.98 1.42
 40–44 years 13 0.96 0.53 1.73 95 1.36 1.08 1.70 90 1.09 0.86 1.37
 45–49 years 19 1.30 0.79 2.15 71 1.04 0.81 1.34 70 1.05 0.81 1.35
 50+ years 14 1.39 0.79 2.46 42 1.02 0.74 1.40 29 0.92 0.63 1.35
p-value for trend 0.21 <0.001 0.18
*

Relative risks (RR) and 95% confidence intervals (CI) from Cox models adjusted for age at entry into cohort, race/ethnicity, education, body mass index, emphysema, smoking status and dose, age at menarche, and type and age at menopause.

Trends were calculated within menopause subgroups, using the youngest age as the referent, despite the use of a different referent group for the calculation of the RRs.