Adding a newer test to digital mammograms can increase the detection rate for breast cancer and decrease nerve-racking false alarms, in which suspicious findings lead women to get extra scans that turn out normal, a study finds.

Millions of women will get the newer test, tomosynthesis, this year. The procedure is nearly identical to a routine mammogram, except that in mammography the machine is stationary, but in tomosynthesis it moves around the breast. Sometimes called 3-D mammography, the test takes many X-rays at different angles to create a three-dimensional image of the breast. It was approved in the United States in 2011.

The verdict is still out on the long-term worth of this new technology. The new results are promising but not definitive, according to experts not associated with the study, published Tuesday in The Journal of the American Medical Association. Tomosynthesis has not been around long enough to determine whether it saves lives or misses tumors.

Even so, more and more mammography centers are buying the equipment — which is far more costly than a standard mammography unit — and marketing the test to patients as a more sensitive and accurate type of screening. It has come on the scene at a time when the value of breast cancer screening and the rising costs of health care are increasingly debated.

Some 1,100 of about 13,500 mammography units in the United States perform tomosynthesis, according to Jim Culley, a spokesman for Hologic, the company that makes the only tomosynthesis equipment approved by the Food and Drug Administration. He estimated that more than 6 million U.S. women will undergo tomosynthesis this year.

"It is very much taking hold in the breast-imaging world," said Dr. Sarah M. Friedewald, the section chief of breast imaging at Advocate Lutheran General Hospital in Park Ridge, Illinois, and the first author of the new study.

Recent studies have suggested that the benefits of mammography have been overstated and its potential harms understated, but many health groups still recommend it. More than 38 million mammograms are performed each year in the United States, at a cost of about $8 billion.

The new information on tomosynthesis is "not going to resolve the ongoing discussions about the overall utility of mammography," said Dr. Barnett S. Kramer, director of the Division of Cancer Prevention at the National Cancer Institute. He added: "In a nutshell, it shows sufficient promise that the thought leaders are interested and so are the people who deal with imaging and screening at the National Cancer Institute." He was not part of the study, but the institute helped support it by providing a research grant to one of its authors.

Dr. Etta D. Pisano, a mammography expert and dean of the Medical University of South Carolina, called tomosynthesis "extremely promising." But in an editorial with the journal article Pisano, who was not involved in the study, said it was still not clear whether women should seek it or clinics should adopt it. She said more research was needed. One concern she raised about the study was that it simply looked back at records instead of using the more rigorous method of picking patients at random to compare types of screening.

Hologic, based in Bedford, Massachusetts, paid for the study and had the right to review the journal article before publication but could not change it, the researchers said.

In an interview, Pisano said, "We need to be cautious as we adopt a new technology."

She said that tomosynthesis units cost about half a million dollars, twice the price of a digital mammography machine, and that her university could not afford one.

"Wealthy communities can afford it," she said.

Hologic declined to provide pricing information for its equipment.

At this point, Pisano said, she did not think the evidence was strong enough to encourage women to go out of their way to have tomosynthesis. She said she herself as a patient would not choose it at this point.

The tomosynthesis test costs more than digital mammography and not all insurers cover it. Culley said that when coverage is denied, some clinics charge patients $50 or so for the extra test in addition to the usual mammography fee.

The study analyzed the records from 13 U.S. mammography centers before and after they added tomosynthesis. The researchers compared the cancer detection rate, how often women had to be called back for more scans to check on suspicious findings, and what proportion of the callbacks and biopsies actually found cancer.

The project involved 454,850 screenings, including 281,187 with digital mammography alone and 173,663 that combined it with tomosynthesis. The exams were done from March 2010 through December 2012.

Tomosynthesis improved cancer detection. Cancer was found in 4.2 of every 1,000 scans with digital mammography alone, but in 5.4 when tomosynthesis was added.

The researchers also analyzed the results further, to separate the more dangerous, invasive cancers from growths called ductal carcinoma in situ, or DCIS, which are less likely to progress and become life-threatening. Again, performance improved with tomosynthesis: combined testing found 4.1 invasive cancers per 1000 scans, whereas digital mammography found 2.9. For in situ growths, the rate was the same, 1.4.

Adding tomosynthesis lowered recall rates, sparing some women an experience that many find frightening. With digital mammography alone, for every 1,000 women, 107 were called back for more X-rays, but with tomosynthesis added, the figure was 91.

With tomosynthesis, there were more biopsies: 19.3 per 1,000 scans, versus 18.1 with digital mammography.

Other findings were expressed in percentages. With digital mammography alone, 4.3 percent of the women called back turned out to have cancer, but the rate was 6.4 with the tests combined.

Although tomosynthesis resulted in more biopsies, more of them were positive for cancer: 29.2 percent versus 24.2 percent with digital mammography alone.

"We found an increase in invasive cancers, the ones we worry about, that could be lethal," Friedewald said. She added: "We're picking up the ones we want to be picking up. Overall, it's very encouraging. We're also reducing the number of people who have to come back."

She said it was good news that the test did not increase the detection of in situ growths. Although these can progress to invasive cancer, most do not, and many doctors think the growths are overtreated.

The study results were the average for the 13 participating mammography centers. Their performance varied. With tomosynthesis, most had fewer recalls but some had more, and while some had markedly fewer unneeded biopsies, others had more. The variation from clinic to clinic means that an individual patient's results might not match those in the study.

"So the balance of benefits and harms may differ," Kramer said.

Tomosynthesis uses more radiation than mammography alone, but the dose is still low and well within limits considered safe, doctors say. Newer tomosynthesis software, already in use at some centers, can cut down on the radiation dose, but the results of that technique were not part of the study.

Other countries began performing tomosynthesis in 2008, Culley, the Hologic spokesman, said, and more than 50 now use it. But many restrict it to patients with breast problems, rather than using it to screen healthy women.

Companies, including General Electric and Siemens, which market tomosynthesis equipment overseas, are expected to introduce it in the United States.