|
Association between exposure categories and progesterone receptor expression |
|||||
| Explanatory Variable |
Controls |
Cases with tumors expressing PR* in 0 to 25% of cells |
Cases with tumors expressing PR* in = order 25 to 100% of cells |
||
|
|
|||||
| n |
n |
OR (95% CI)† |
n |
OR (95% CI)† |
|
|
|
|||||
| OC* use |
|||||
| Never |
103 |
15 |
Referent |
33 |
Referent |
| Ever |
183 |
31 |
3.2 (1.3 – 8.0) |
64 |
1.1 (0.6 – 1.9) |
| HRT* use in post-menopausal women |
|||||
| Never |
58 |
15 |
Referent |
23 |
Referent |
| Ever |
114 |
22 |
0.8 (0.4 – 1.6) |
41 |
0.9 (0.5 – 1.6) |
| Current menstrual status |
|||||
| Menstruating |
114 |
9 |
Referent |
33 |
Referent |
| Post-menopausal |
172 |
37 |
2.7 (1.2 – 5.8) |
64 |
1.3 (0.8 – 2.1) |
| Menstrual status 10 years before reference date |
|||||
| Menstruating |
158 |
21 |
Referent |
62 |
Referent |
| Post-menopausal |
126 |
23 |
0.5 (0.2 – 1.5) |
35 |
0.5 (0.2 – 1.0) |
| Current menstrual status |
|||||
| Menstruating |
72 |
4 |
Referent |
17 |
Referent |
| Post-menopausal and not using HRT |
58 |
15 |
2.4 (1.2 – 5.8) |
23 |
3.1 (1.1 – 8.6) |
| Menstrual status 10 years before reference date |
|||||
| Menstruating |
83 |
8 |
Referent |
27 |
Referent |
| Post-menopausal and not using HRT |
46 |
10 |
0.3 (0.1 – 2.5) |
13 |
0.8 (0.5 – 1.3) |
| Smoking status 10 years before reference date |
|||||
| Non-smoker |
144 |
29 |
Referent |
48 |
Referent |
| Smoker |
142 |
17 |
0.5 (0.3 – 1.0) |
49 |
1.0 (0.6 – 1.5) |
| Body Mass Index on reference date |
|||||
| < 25 |
156 |
23 |
Referent |
45 |
Referent |
| 25 – 30 |
66 |
12 |
1.2 (0.5 – 2.5) |
22 |
1.1 (0.5 – 2.4) |
| ≥ 30 |
64 |
11 |
1.1 (0.6 – 2.0) |
29 |
1.5 (0.8 – 2.6) |
| Number of live births |
|||||
| None |
58 |
2 |
Referent |
16 |
Referent |
| 1 |
28 |
5 |
4.0 (0.7 – 22.3) |
17 |
2.1 (0.9 – 4.9) |
| ≥ 2 |
200 |
39 |
4.3 (1.0 – 18.8) |
64 |
1.1 (0.6 – 2.2) |
|
*PR, progesterone receptor; OC, oral contraceptive; HRT, hormone replacement therapy † Odds ratio (OR) and 95% confidence intervals (CI) estimated from multivariable polytomous logistic regression including continuous age and three categories of education (high school or less, some college or trade school, and college graduate or graduate school) as confounding variables. None of the tests for linear trend was significant. | |||||
Custer et al. BMC Cancer 2006 6:152 doi:10.1186/1471-2407-6-152 |
|||||