Only a fraction of women who need mammograms actually get them every year as recommended, according to a new study. But skipping this important screening test raises a woman's chances of dying from breast cancer, researchers from Massachusetts General Hospital and Harvard Medical School report.

They studied more than 72,000 women who received mammograms at Massachusetts General Hospital between 1985 and 2002. Just 6% of the women who got screened in 1992 returned for another mammogram each of the following 10 years. Among women who got screened in 1996, only 16% returned for another test in each of the next 5 years.

"The medical community has done a great job communicating that it's important to get a mammogram," said lead researcher James Michaelson, PhD, "but it hasn't done as great a job communicating that it's important to return promptly [for subsequent mammograms]."

DON'T FORGET: Use the ACS mammogram reminder to help you remember your next screening

Women who had already been treated for breast cancer had the highest rates of screening, but even they didn't get checked every year. Minority women, those with low incomes, those with no health insurance, and women who didn't speak English had lower rates of screening than other women.

The findings were reported in the June 21 online edition of the journal Cancer, and will appear in the August 1 print edition.

Although mammograms don't always find every case of breast cancer, they are the most reliable breast screening tool available today. Regular mammograms can help spot a breast tumor while it is still small and easier to treat, boosting a woman's chances of surviving the disease. For this reason, the American Cancer Society recommends that women get a mammogram every year, starting at age 40.

The number of women following that recommendation was "disappointingly low," Michaelson said. And it could have a dramatic impact on survival.

Using a computer simulation, Michaelson and his colleagues estimated that among women who develop breast cancer, about 12% of those who get screened every year would eventually die from the disease. But if these women were to get screened every other year, about 16% would eventually die of breast cancer, the researchers predicted. The outlook would be even worse for women who develop breast cancer and were screened only once every 5 years: more than 25% of these women would die, according to the model. One-fifth of the women in the study got screened at this rate.

The researchers could not determine how many women may have gone somewhere else to get a mammogram. But they noted that Massachusetts General Hospital has "an aggressive reminder program" to help women keep their appointments for breast cancer screening. Some 90% of women who make appointments for a mammogram at the hospital keep the date.

Still, the study results suggest too few women are scheduling those crucial appointments. Better reminder systems -- like those many dentists use to keep patients on track with their cleanings -- could help, Michaelson said.

"I personally have a great interest in computer systems for calling people and reminding them [to make an appointment] and I think we could make a big impact in that realm," he said.

But reminder systems are just one part of what should be a larger examination of breast cancer screening processes, he added.

"Mammography needs to be studied like any large enterprise," Michaelson said. That means experts need to find ways to make the process more efficient -- What's the best way to organize the appointment-making system? How much time should be allotted for each appointment? -- and more pleasant for women.

"If we could pay more attention to the management side of screening, we could really make the experience better and save a lot of lives," Michaelson said.