Last week, a friend who knows my views on excessive medical tests and treatments sent me a link to a People magazine cover story about Amy Robach, the correspondent for ABC's Good Morning America. After receiving an on-air mammogram in October, Robach was diagnosed with breast cancer and underwent a double mastectomy. The cover of People quotes Robach saying, in big yellow letters, "I'm lucky to be alive."

I wrote about Robach's case last month. I'm returning to it now because I fear that Robach, whose courage and candor I admire, is misinforming women about mammograms. For example, she tells People that she regrets not receiving a mammogram earlier. "A year could have made a big difference," she says.

Robach is 40. The National Cancer Institute, a federal research agency, states that "studies to date have not shown a benefit from regular screening mammography in women under age 40." Indeed, for women of any age, the benefits of mammograms are questionable and the risks considerable.

Gemma Jacklyn and Alexandra Barratt, public health experts at the University of Sydney, note that Robach and other celebrities publicizing their mammograms "have a simple message that’s easy to embrace--mammography screening saves lives. Sadly, it’s not that simple."

A positive test for breast cancer, Jacklyn and Barratt assert, is more likely to represent a case of "overdiagnosis" than to be "life-saving." Overdiagnosis" involves detecting "small cancers that, if left alone, would not cause any symptoms or death," they explain.

A 2012 report in Lancet by the Independent U.K. Panel on Breast Cancer Screening concluded that women who undergo screening every two years for 20 years are three times more likely to be over diagnosed than have their life saved. "We’re used to thinking of breast cancer as uniformly lethal if left untreated," Jacklyn and Barratt say. "But studies increasingly show that finding tiny cancers doesn't necessarily translate into saving lives."

Some studies of mammograms show even higher rates of overdiagnosis. A 2006 report in the British Journal of Medicine estimates that screening 2,000 women for 10 years will prevent one woman from dying of breast cancer. Meanwhile, "10 healthy women, who would not have been diagnosed without screening, will have breast cancer diagnosed and be treated unnecessarily; 4 of these will have a breast removed, 6 will receive breast conserving surgery, and most will receive radiotherapy."

In a 2011 study in Archives of Internal Medicine, H. Gilbert Welch, a physician at Dartmouth and authority on medical testing, and a co-author concur that a majority of women with screen-detected breast cancer "have not had their life saved by screening. They are instead either diagnosed early (with no effect on their mortality) or overdiagnosed."

Writing about Amy Robach for CNN, Welch notes that "powerful survivor stories that appear regularly in the media" are too often interpreted "as evidence of the benefit of mammograms. Unfortunately, the more likely interpretation is that they represent evidence of harm: unnecessary surgery, chemotherapy and/or radiation."

When I have discussed these data with female friends, many say that they will continue to receive mammograms; they are willing to endure what may be unnecessary, unpleasant tests and treatments for even a slight reduction of their chances of dying of cancer.

But of course, surgery, chemotherapy and radiation can also adversely affect health. I thus emailed the following question to Welch: "Has anyone estimated how many women die prematurely because of receiving unnecessary treatment for cancer?"

The short answer is no," Welch replied, "or at least not to my knowledge. The near-term death risks from these treatments are not zero, but they are low. My take on the trade-off is less about death vs. death, more the very small chance of avoidance of cancer death (unknown effect on all-cause mortality) vs. a larger chance of unnecessary diagnosis and treatment and a much larger chance of false alarm with increased sense of vulnerability (not cancer, but not normal). No right answer, but clearly a choice--not a public health imperative."

In his CNN column on Robach, Welch notes that "no one wants to dispute the interpretation of a well-meaning cancer patient who is trying to help people… But news stories about health--particularly on television--are too driven by powerful personal anecdotes. The public deserves more nuance."