Table 1.
Invasive breast cancer incidence and mortality by estrogen plus progestin use at baseline
| Clinical event Cox regression model | Estrogen plus progestin | No hormone therapy | HR (95% CI) | P |
|---|---|---|---|---|
| No. (%) | No. (%) | |||
| Invasive breast cancer incidence | ||||
| M1*: Multivariable adjusted | 1097 (0.60%) | 1139 (0.42%) | 1.55 (1.41 to 1.70) | <.001 |
| M2†: M1 censoring for a >2-year interval without a mammogram | 840 (0.63%) | 702 (0.39%) | 1.61 (1.46 to 1.77) | <.001 |
| M3‡: M1 censoring for adherence | 586 (0.75%) | 919 (0.41%) | 1.99 (1.68 to 2.37) | <.001 |
| Deaths from breast cancer | ||||
| M1*: Multivariable adjusted | 54 (0.03%) | 85 (0.03%) | 1.32 (0.90 to 1.93) | .15 |
| M2†: M1 censoring for a >2-year interval without a mammogram | 45 (0.03%) | 55 (0.03%) | 1.31 (0.81 to 2.11) | .27 |
| M3‡: M1 censoring for adherence | 30 (0.04%) | 74 (0.03%) | 1.41 (0.89 to 2.23) | .15 |
| Deaths after breast cancer | ||||
| M1*: Multivariable adjusted | 141 (0.07%) | 192 (0.06%) | 1.65 (1.29 to 2.12) | <.001 |
| M2†: M1 censoring for a >2-year interval without a mammogram | 123 (0.08%) | 120 (0.06%) | 1.87 (1.37 to 2.54) | <.001 |
| M3‡: M1 censoring for adherence | 82 (0.10%) | 169 (0.07%) | 1.62 (1.09 to 2.40) | .02 |
* Analyses were adjusted for age, race or ethnic group, body mass index, education, smoking status, alcohol use, self-reported health, level of physical activity, presence or absence of a family history of breast cancer, estimated breast-cancer risk based on the Gail model, and bilateral oophorectomy and stratified by baseline age group. All statistical tests were two-sided.
† In addition to M1 adjustments, Cox regression model included time-varying weights, which were inversely proportional to the estimated probability of continued mammogram use to adjust for changes in the distribution of sample characteristics during follow-up. All statistical tests were two-sided.
‡ In addition to MI adjustments, Cox regression model included time-varying weights, which were inversely proportional to the estimated probability of adherence (conformation to baseline study group through ongoing use of estrogen plus progestin or no hormone therapy during follow-up) to adjust for changes in the distribution of sample characteristics during follow-up. All statistical tests were two-sided.