That's because when doctors screen for cancer in women who don't have symptoms, it's impossible for them to tell whether a tumor picked up by mammography will grow quickly into advanced cancer or will only progress slowly or not at all, said lead author Dr. Mette Kalager.
And although women are well-versed in the benefits of mammography, they aren't always warned about the possible harms of so-called overdiagnosis and overtreatment, according to Kalager, a breast cancer surgeon and a visiting scientist at the Harvard School of Public Health in Boston.
"You have to really consider the benefit and the harm against each other, and really think through: what is my risk of dying from breast cancer, and what is my risk of being overdiagnosed?" she said.
The problem with treating tumors that would never cause any symptoms or cut women's lives short is that women only experience harms with no health benefit, Kalager added.
"These women undergo treatment, that is surgery, radiotherapy, chemotherapy or hormone treatment for breast cancer, that they don't really need," she told Reuters Health.
"It's not only the distress of being a cancer patient, but really the harms of treatment."
Radiation therapy itself is linked to a slightly increased risk of cancer and chemotherapy comes with a range of side effects, from nausea and fatigue to a higher risk of infections.
"For years I think we've exaggerated the benefits (of mammography) and we've sort of downplayed or minimized the harms," said Dr. H. Gilbert Welch, who studies cancer screening at the Dartmouth Institute for Health Policy & Clinical Practice in Lebanon, New Hampshire.
"The issue is no longer whether overdiagnosis occurs, it's how often," said Welch, the author of Overdiagnosed: Making People Sick in the Pursuit of Health, who wasn't part of the new study team.
For the research, Kalager and her colleagues compared breast cancer diagnoses in counties in Norway that had or hadn't started widespread screening between 1996 and 2005 -- as well as in the ten years prior when none of the counties had a screening program in place, for comparison.
Starting in four out of 19 counties, mammography was offered to all women between the ages of 50 and 69.
In the United States, the government-backed U.S. Preventive Services Task Force recommends screening women age 50 to 74 every other year, although some other organizations still call for regular mammograms for all women starting at 40.
In counties that didn't start offering screening until after the study period, there was an 18 per cent increase in the number of breast cancers diagnosed between 1986 to 1995 and 1996 to 2005 -- possibly because of increased use of menopause drugs that have been tied to cancer, the researchers explained.
In comparison, 47 per cent more breast cancers were diagnosed in counties that rolled out screening programs during that time period, Kalager's team reported Monday in the Annals of Internal Medicine.
When they followed women to see whether the number of cancers in the no-screening group caught up with the number spotted during mammograms, the researchers found that it didn't.
From the difference in cancer rates between women who were screened and those who weren't, they concluded that 15 to 25 per cent of the cancers caught by screening wouldn't have become symptomatic or life-threatening.
They calculated that for every 2,500 women invited to get screened over ten years, there would be 20 cases of life-threatening breast cancer caught and one breast cancer death prevented. Among that same group of women, between six and 10 would be diagnosed and treated for a tumor that never would have caused them any problems.
The findings don't mean that mammography is never worth it, Kalager said, and there's evidence it does save some lives.
Dongfeng Wu, who models cancer screening at the University of Louisville, said that her own research has suggested the rate of overdiagnosis is generally lower, at about six to nine per cent instead of up to 25 per cent.
That's based on estimates of how accurate mammography is, as well as how long it typically takes breast tumors to become symptomatic.
Even with a lower chance of unnecessary diagnosis and treatment, some women may choose to opt out of mammography, said Wu, who wasn't involved in the new study.
"If a woman thinks she has a higher risk because of her family -- close relatives have breast cancer -- she might want to (get screened) regularly," Wu told Reuters Health.
"Or if she considers herself a low-risk person, she might just do it every few years, or she might just skip it."
The researchers agreed that once women have all the relevant information, they should be allowed to make their own decision on screening without judgment.
"Women are told about the benefits, but not about the harms," Kalager said. "They are really misinformed."
The harms of cancer screening are the subject of debate. They can range from anxiety to further invasive and expensive testing and treatment.
The authors of a linked commentary wrote that doctors "have an ethical responsibility to alert women to this phenomenon."
Dr. Joann Elmore from the University of Washington School of Medicine and Dr. Suzanne Fletcher of Harvard Medical School added, "Most patient-education aids do not even mention overdiagnosis, and most women are not aware of its possibility."
"There's no right answer here except: tell people the truth," Welch told Reuters Health.
"Rational women can look at the data on mammography and come to different conclusions. I think there's a genuine choice here."