Asking more "unpleasant questions about mammography screening"

Posted By


There’s been surprisingly little coverage of an analysis by the Nordic Cochrane group in this week’s BMJ that concludes:

“We were unable to find an effect of the Danish screening programme on breast cancer mortality. The reductions in breast cancer mortality we observed in screening regions were similar or less than those in non-screened areas and in age groups too young to benefit from screening, and are more likely explained by changes in risk factors and improved treatment than by screening mammography.”

Of the stories I could find, Andre Picard of The Globe and Mail had one of the better reports, including a quote from one of the researchers:

“We have to start asking unpleasant questions about mammography screening.”

You might also like

Comments (12)

Please note, comments are no longer published through this website. All previously made comments are still archived and available for viewing through select posts.


March 24, 2010 at 3:49 pm

I would suggest you close your blog to comments NOW! Recall what happened last fall with the USPSTF recommendations regarding mammos in women under 50.
Hell hath no fury…


March 25, 2010 at 6:48 am

There was an article in Politiken yesterday.
“This claim isn’t supported”, says prof. at University of Copenhagen’s Institute for Public Health Science, Elsebeth Lynge.
But EL and her two colleagues Sisse Njor, Uni. of Cop. and Anne Helene Olsen, University of Tromsø do not think the Cochrane researchers are correct in their conclusion.
It’s the analysis itself that doesn’t hold up according to the three university researchers.
“The authors first of all use very coarsegrained data, secondly they use an analysis that is inappropriate for the given problem. I call their numbers ‘polluted'”, says Elsebeth Lynge.
Among other sthings EL, SN and AHO emphasise that screening can only affect the mortality from breast cancer in women who have not already been diagnosed with it before screening is introduced. [I think they mean the screening programme, not the scans in the individuals.]
“When analysing the effect of screening it’s thus necessary to remove women who already have breast cancer. This is technically termed ‘refined mortality’. Jørgensen & al do not use ‘ref mor'”, according to the three researchers.
[They] have previously demonstrated that 10 years after the introduction of screening the mortality from breastcancer had dropped by 25% in Copenhagen.
Their study was performed in 2005 and was – like the Cochrane-researchers – published in British Medical Journal.

Hope that helps (h/t Goldacre)

jeffrey dach md

March 26, 2010 at 11:20 am

Excellent article on the topic of screening mammography. I was actually impressed at the level of the information. Many of the points have been raised over the years by other critics of screening mammography, and in my opinion, valid criticisms.
Especially disturbing is that mammography screening generates large numbers of procedures, biopsies and surgeries, with little impact on the numbers of advanced breast cancers. This point was raised
by Laura Esserman in her JAMA article. This new study by Karsten Juhl Jørgensen, M.D., of the University of Copenhagen in Denmark is very supportive of Dr Esserman’s conclusions.
One looming issue is the large numbers of DCIS detected by mammography as small calcifications. Although DCIS is treated aggressively as an invasive cancer, it is really very indolent. DCIS has a 98% -5 year survival with no treatment, and pathologists have been thinking about changing the nomenclature by removing the word “cancer” from its name.
For more:
jeffrey dach md

Elaine Schattner, M.D.

March 26, 2010 at 2:11 pm

I’m glad you didn’t close the comments, as I hope the world’s medical community won’t shut the door on solid data supporting mammography’s use in breast cancer screening.
I find the latest BMJ article to be deeply flawed, a poor example of statistics and circular reasoning. My biggest concern is that some readers, doctors, patients and others will see this as more “evidence” that mammography is ineffective.
Evidence-based medicine can only be good as the data that support it. There’s a real danger here – the term “evidence” can be coercive.
I’d like to write more on this later.

Greg Pawelski

March 28, 2010 at 11:33 am

It is interesting that more and more evidence that mammography is ineffective in this subset group keeps coming up. The BMJ study about mammograms was not brand new information based on whatever critique anyone wants to give it. The newer recommendations were based on research that experts have known about for some time. It seemed like it wasn’t politically feasible to expose it before.
Many years ago, the NCI tried to convince us all to not screen women younger than 50, but the institute was given such a tongue lashing by Congress, they withdrew their recommendation. As one former ACS executive stated, “Powerful forces stood in the way of widespread dissemination while millions of dollars were poured into the Mammogram campaign.”
Likewise, the ACS also avoids looking clearly at the data and continues to recommend screening for younger women. But mammography is not the answer for these women, the executive goes on, “The unfortunate side effect of this delusion (that screening and early detection is the answer) is that we avoid the hard choices like healthy life styles and avoiding cancer-causing drugs such as hormone-replacement treatment.”

Elaine Schattner, M.D.

March 29, 2010 at 10:50 am

Of course preventing cancer is better than detecting it. But treating early-stage breast cancer is far less expensive than dealing with metastatic disease, which is incurable as things stand.
The Task Force analyzed and published data that were decades old and did not include any findings or recommendations regarding digital mammograms – the current, preferred technology for pre-menopausal women in their forties and early fifties.
Between 1992 and 2006, mortality from breast cancer among U.S. women under the age of 50 declined by roughly 35 percent. Oncologists attribute this improvement to two factors: earlier detection and better treatment of the disease once found.
Unfortunately, this favorable trend may reverse if screening in pre-menopausal women comes to a halt.