WASHINGTON (AP) _ In the seesawing debate over mammograms, the government issued an affirmation Thursday: Women age 40 and older should get one every year or two.
An international uproar has arisen over whether routine mammograms really save lives. Danish scientists analyzed three decades' worth of mammography studies last fall and concluded they're so flawed, it's impossible to tell whether the X-rays help.
An independent panel assigned by the U.S. government to investigate concluded Thursday there is ``fair evidence'' that getting one regularly can reduce the risk of dying from breast cancer by about 20 percent. That evidence is strongest for women over 50, but mammograms likely will benefit 40-somethings, too, concluded the U.S. Preventive Services Task Force.
``Our bottom line is that mammography reduces deaths,'' said task force vice chair Janet Allan.
But there's no firm age at which to start getting them. Discuss that with your doctor based on your individual risk of cancer, she advised. Risks include having a mother or sister with breast cancer, a first child after age 30, or a previous biopsy that found unusual cell growth.
Nor is there proof that an annual mammogram is better than one every other year.
And, a finding that may surprise some women, there's no evidence yet that a breast self-exam, or one by a doctor in addition to mammography, saves lives, the task force said.
American women undergo some 30 million mammograms a year. The breast X-ray long has been considered the best way to detect breast cancer, which strikes 190,000 women a year and kills 40,000 of them.
But the Danish analysis, published in the medical journal Lancet last fall, caused a backlash as scientists, doctors and patients debated the tests' value.
Last month, advisers to the National Cancer Institute agreed that the Danish work raised serious doubts. The question: whether by the time a mammogram spots a breast tumor, even a tiny one, it has been there long enough that the woman's chance of survival is largely predetermined, explained Donald Berry of the M.D. Anderson Cancer Center, one of the NCI advisers.
So the NCI and the task force _ an independent panel charged by the government to evaluate the latest research in deciding what routine medical screening average Americans need _ investigated.
Some of the original mammography studies, done in the 1970s and '80s, had flaws but not drastic ones, they concluded.
For example, the Danes disregarded a Swedish study partly because it concluded that women who got mammograms died ``with'' breast cancer while women who didn't died ``of'' it, said NCI cancer prevention chief Dr. Peter Greenwald. In fact, finding breast cancer early enough that women survive it long enough to die of something else is mammography's point, he said.
U.S. breast cancer deaths did decline in the 1990s, due partly to mammograms finding early tumors, he said.
Thursday's announcement is the government's final word, said Health and Human Services Secretary Tommy Thompson.
``Women are confused,'' Thompson said. The new guidelines send ``a very powerful and clear message. ... Please take these recommendations to heart.''
But the controversy isn't settled. The NCI advisory panel that last month aired concern about mammograms also controls what information is posted on a main NCI Web page for consumers seeking cancer advice. And it still intends to update that Web page to say ``there is uncertainty regarding the benefits of mammography,'' Berry said Thursday.
Mammograms aren't perfect, Allan acknowledged. They miss some cancer. They also too often flag benign lumps, causing unnecessary anxiety and additional testing, even biopsies. Over 10 years of screening, about 23 percent of women will have an abnormal mammogram and about 80 percent of those will be benign.
Also, more powerful mammograms are uncovering more tiny tumors called ``ductal carcinoma in situ'' that aren't invasive cancer and don't always become life-threatening. Many experts say such patients are overtreated, but there's no way to tell in advance who will be endangered.
Thursday's decision updates the task force's last recommendation, in 1996, that set age 50 as mammography's starting age. Separately, the NCI already was urging testing of 40-somethings. Neither group decides if insurers pay for the test.