Early Mammograms – New Study Misses the Mark

mammogram2A recent study has concluded that women with breast cancer who failed to get annual mammograms are more likely to die from their disease than those who had annual mammograms, and argues that more frequent mammograms are warranted in women under age 50. Unfortunately, despite all the media attention this study is getting, I don’t think the researcher’s conclusions are supported by the study results.

The researchers did a retrospective medical record review on deaths that occurred among breast cancer patients receiving care at Mass General or Brigham & Women’s Hospitals in Boston between 1990-1999 and followed until 2007. They call this a Failure Analysis.

Invasive breast cancer failure analysis defined 7301 patients between 1990 and 1999, with 1705 documented deaths from breast cancer (n = 609) or other causes (n = 905). Among 609 confirmed breast cancer deaths, 29% were among women who had been screened (19% screen-detected and 10% interval cancers), whereas 71% were among unscreened women, including > 2 years since last mammogram (6%), or never screened (65%). Overall, 29% of cancer deaths were screened, whereas 71% were unscreened. Median age at diagnosis of fatal cancers was 49 years; in deaths not from breast cancer, median age at diagnosis was 72 years

The authors concluded that because most deaths from breast cancer occur in un-screened women under age 50, initiation of regular mammograms before age 50 years should be encouraged.

Where this Failure Analysis Fails

Despite its strongly worded conclusions, the study raises more questions than it answers, and has a number of severe limitations.

  • The study fails to tell us what percent of women who did not die got annual vs not annual mammograms. This is akin to reporting that 80% of auto accident deaths occur among those who started their trip at home vs a public parking garage, without telling you what percentage of all car trips originate from home.
  • The study did not compare breast cancer treatments between women who died and those who did not die. The researchers just assumed that all women got standard of care at their medical center for their cancer. That’s a huge assumption to make without any proof.  It would have been actually quite easy to review a statistical sampling of charts to determine if this assumption was correct, but the researchers did not do this.
  • Women who don’t get regular mammograms may differ from those who do in other ways that increase the risk for death from breast cancer death, including low socioeconomic status, lack of health insurance, and distrust of medical treatments.
  • The researchers try to make the point that among those who died of their cancer, those who had not had mammograms prior to diagnosis had later stage cancers. Given that this analysis was confined to patients died of their cancer, I’m not sure stage at diagnosis mattered.
  • The entire analysis is conducted among women who died, either from breast cancer or from other causes. Since death from non-cancer causes is rare in women under age 50, breast cancer deaths will be over-represented in younger women in the sample. Older women not getting mammograms may be not getting screening because they are ill from other causes and are also more likely to die from these other causes during the follow up period, making breast cancer deaths less common in this group.  Who knows which way the data ultimately skewed as a result of these biases, but regardless, it is skewing every which way as far as I’m concerned. All of which muddies the conclusions.
  • The study was conducted at a Mass General and Brigham and Women’s Hospital using records from their breast cancer registry. Both these hospitals are referral centers likely to attract younger women with more aggressive cancers for treatment, who may not be representative of the general population of women presenting for mammogram screening or who are diagnosed with breast cancer. Indeed, the study population was over 90% white and of high socioeconomic status, pretty standard for a referral population if I ever saw one.

One thing that is evident is that breast cancers in younger women tend to be more aggressive than those in older women, an idea that would support more aggressive screening in younger women since each life saved carries more years of life saved. However, this is countermanded by the argument that breast cancer, despite being more aggressive, occurs much less frequently in younger than older women. Add in that mammograms are much better at detecting slower growing, less fatal breast cancers than the more aggressive cancers, and that screening is less effective in the denser breasts of younger women, and you have a sense of the screening conundrum we face for this cancer that claims so many women’s lives each year.

Unfortunately, this retrospective analysis is not going to solve the issue.


Webb, M. L., Cady, B., Michaelson, J. S., Bush, D. M., Calvillo, K. Z., Kopans, D. B. and Smith, B. L. (2013), A failure analysis of invasive breast cancer. Cancer. doi: 10.1002/cncr.28199

More on mammograms –

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.