If a suspicious mammogram turns out to be a false alarm, the experience can change the woman's attitude toward breast cancer screening.

A number of studies have found that women follow screening recommendations even more closely after a "false positive" mammogram, though ruling out cancer requires them to return for additional breast X-rays and, occasionally, a biopsy.

But a large new study by Chicago researchers found just the opposite reaction. Compared with women whose mammograms correctly detected no cancer, those with false positives were more likely to delay their next breast-squishing exam -- or to stop showing up.

"The medical literature suggests the experience of a false positive can cause anxiety, worry, and affect the woman's quality of life," said lead author Firas M. Dabbous, an epidemiologist and researcher at Advocate Lutheran General Hospital near Chicago. "That may deter a woman from coming back. She thinks, 'I don't want to go through it again.' But you can also argue the other way around. She thinks, 'I'm happy they found nothing, and I'd go through that again to be sure I don't have cancer.' The hypothesis I went with is that the experience deters women from coming back."

Debra Copit, director of breast imaging at the Einstein Healthcare Network, was "a little surprised" by the results.

"Typically, I see just the opposite," she said. "Many women ask if they should come more frequently after a false positive. My experience over the past 20 years is that they're happy and relieved, and while they have short-term anxiety, that resolves and they come back."

Dabbous and his co-authors analyzed more than 741,000 mammograms of 261,767 women taken between 2001 and 2014. About 12 percent of those X-rays were false positives, close to the industry benchmark of 10 percent. (Women whose mammograms correctly detected cancer were excluded.)

The researchers did not address the fact that screening guidelines -- which have divided the medical community since the 1980s — are inconsistent. This reflects debate over the value of screening, which has been shown to reduce breast cancer deaths by 15 percent to 20 percent.

Doctors in Advocate's 12-hospital network follow the predominant guidance, telling women to get screened annually starting at age 40. Reminders are sent digitally and by mail. The study looked at women who took more than a year to return for screening, or didn't return at all.

However, since 2009, the U.S. Preventive Services Task Force, an independent expert group that influences insurance coverage, has recommended mammograms every other year beginning at age 50. Some medical groups hew to this model.

"No matter where you stand on the recommendations, it's a hot topic, and the back-and-forth over what's right is very confusing for women," said Emily Conant, chief of breast imaging at the University of Pennsylvania.

The new study did highlight another well-known issue: women with certain characteristics are more susceptible to having mammograms that are hard to interpret, so they have more false positives.

"Women with false positives were younger, premenopausal, and were more likely to be experiencing their first screening," said the study in Cancer Epidemiology, Biomarkers and Prevention, a journal of the American Association for Cancer Research. "Also, they were more likely to be black, have denser breasts," and lack a previous mammogram for comparison.

To compensate, the researchers analyzed their data with and without making statistical adjustments for age, race, and other factors.

They found that about 22 percent of women with false positives and 15 percent with true negative  mammograms did not come back. (The study could not say whether some women went to different mammography centers.)  After adjustment, the difference was smaller: 19 percent of women with true negatives didn't return.

The study also found that women with false positives were slower to return for a routine annual mammogram. They took a median of 25 months from the initial screening, compared with 15 months for women with true negative mammograms. After adjustment, women with true negatives delayed a median of 18 months.

The article did not include data on women who developed cancer, or their corresponding screening delays, yet it concluded that "a prior false positive experience may subsequently increase" the risk of being diagnosed with late-stage cancer. Dabbous said he didn't provide the numbers because of journal space limitations.

While the missing data left her guessing, Conant said, "all in all, it's a strong study that suggests an area for  improvement, both for health-care providers and radiologists, in communicating results to patients and discussing their approach to screening."