A Sharp Decrease In Breast Cancer Incidence In The United States In 2003

Reviewer: John P. Plastaras, MD, PhD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: December 14, 2006

Presenter: P. M. Ravdin
Affiliation: M.D. Anderson Cancer Center, Houston, TX


  • The incidence of breast cancer over time may reflect a differences in etiologic factors and screening practices. For example, after the wide-spread institution of mammography, there was bump in the incidence of breast cancer
  • The incidence of breast cancer has been stable from 1998 to 2002.
  • SEER data on the incidence of cancers is delayed by about 3 years for validation.
  • This study examines a downward trend in breast cancer incidence in 2003
  • There were two important events in breast cancer in 2002:
    • Controversy regarding the value of mammographic screening
    • Results of the Women’s Health Initiative showing that hormone therapy had no cardioprotective effect but increased the incidence of breast cancer by 1.24. This lead to a dramatic decrease in the number of prescriptions for Premarin and Prempro between 2002 and 2004


  • The SEER public use data from 1990 to 2003 was analyzed.


  • Breast cancer incidence in the U.S. increased at about 1.7% per year from 1990 to 1998. Then, the incidence decreased 1% per year from 1998 to 2003. In 2003, however, there was a sharp decrease of 7% in one year.
  • This was analyzed by region and it did not appear to be a statistical artifact.
  • There was a 5.5% decrease in the incidence of in situ cancers and a 7.3% decrease in invasive cancers.
  • The decrease is most consistent with a change in use of hormone therapy:
    • The decline was most evident in patients 50-74 years old, which is the group most likely to have used hormone therapy
    • The decline was 8% for ER+ tumors and 4% in ER- tumors. There was an even greater disparity between ER+/- breast cancer incidence in women aged 50-69 (12% vs. 4%).
  • The decrease was not confined to small stage I tumors, arguing against the hypothesis that the controversy surrounding mammography was responsible for the decrease
    • There was only a 1% decrease in the use of mammography during this time, which doesn’t account for the magnitude of the change. The decrease was 3.2% in women aged 50-64, but even this doesn’t account for the large decrease seen in this age group

Author's Conclusions

  • The sharp 7% decrease in the incidence of breast cancer in 2003 is most likely attributable to decreased use of conjugated estrogen hormonal therapy, a reaction to the results of the Women’s Health Initiative study released in mid 2002.
  • If this is true, the effect occurred too early to have affected tumor initiation, favoring an effect of tumor progression.
  • Results from additional incidence data will determine whether the incidence will continue to decrease

Clinical/Scientific Implications

  • Epidemiologic data like these are certainly intriguing, but do not establish causality of hormonal therapy as the reason for the sharp decrease.
  • Other hypotheses, such as increased use of tamoxifen for prevention, use of raloxifene for osteoporosis, or decreased incidence of second primaries resulting from increased use of tamoxifen in the adjuvant setting were all of too small of a magnitude to account for the decrease
  • This study shows the potential important impact of studies like the Women's Health Initiative in establishing preventative measures. It is a shame that sales of cigarettes have not plummeted in the same way that hormone therapy fell out of fashion.