Trends in breast cancer incidence associated with reductions in the use of hormone replacement therapy

Background: The publication of the Women's Health Initiative Study in 2002 spurred reductions in hormone replacement therapy (HRT) use worldwide. Subsequent drops in breast cancer incidence have since been reported. We used person-level data from a large Israeli health plan to study the relationship between HRT use and breast cancer incidence.

Methods: We accessed pharmacy, mammography and cancer registry data for women ages 50 and over. Using Chi-square analysis, we examined changes over time in HRT and mammography utilization and breast cancer incidence [invasive and ductal carcinoma in situ (DCIS)]. Using calendar year-specific Cox proportional hazards analyses, we estimated the hazard of recent HRT use on cancer incidence. We used locally weighted scatterplot smoothing (LOWESS) and autoregressive integrated moving average (ARIMA) modeling to further examine this association.

Results: The study population ranged from 118,724 in 2000 to 154,447 in 2007. Thirteen percent of study women purchased HRT in 2000, vs. 4.0% in 2007 (p<0.001). Breast cancer incidence rose from 3.8/1000 to 4.3/1000 in 2005; the proportion of women undergoing mammography rose from 8% in 2000 to a peak of 34.1% in 2005 (p<0.001). Hazard ratios for the association of HRT with breast cancer ranged from 1.2 to 2.4. ARIMA demonstrated the association of both mammography (in the same month) and HRT use (in the preceding month) with cancer incidence.

Conclusion: This study demonstrates the competing impacts of HRT and mammography on cancer incidence. Our findings support the conclusion that changes in HRT utilization are linked to reductions in breast cancer incidence.