HR and 95% CIs for breast cancer risk by categories of the cumulative number of menstrual cycles before FFTP
| No. menstrual cycles | Cancer cases | Person-years | Crude model | Adjusted models | ||
|---|---|---|---|---|---|---|
| HR (95% CI) | HR (95% CI)* | HR (95% CI)† | ||||
| Always irregular | 12 | 5,113 | 0.72 (0.37-1.41) | 0.71 (0.35-1.44) | 0.72 (0.35-1.46) | |
| ≤118 | 30 | 9,127 | 1‡ | 1‡ | 1‡ | |
| 119-147 | 39 | 9,565 | 1.25 (0.78-2.01) | 1.24 (0.77-1.99) | 1.21 (0.75-1.96) | |
| 148-185 | 25 | 6,568 | 1.17 (0.69-1.99) | 1.15 (0.67-1.96) | 1.14 (0.66-1.95) | |
| ≥186 | 38 | 9,343 | 1.25 (0.77-2.01) | 1.23 (0.76-1.98) | 1.27 (0.77-2.08) | |
| Ptrend§ | 0.351 | 0.494 | 0.407 | |||
NOTE: Cox model with follow up as time scale. n = 5,130 (144 breast cancer cases).
↵* Adjusted for age at enrollment, mother or sister with breast cancer (y/n), BMI, more than 1 year of irregular years (y/n), hormone replacement therapy ever (y/n).
↵† Adjusted for age at enrollment, mother or sister with breast cancer (y/n), BMI, more than 1 year of irregular years (y/n), hormone replacement therapy ever (y/n), ever married (y/n), infertility consult ever (y/n), and educational level.
↵‡ Reference category.
↵§ The irregular category was not included when testing for trend.