A Canadian study that many experts say has major flaws has revived debate about the value of mammograms. The research suggests that the screening X-rays do not lower the risk of dying of breast cancer while finding many tumors that do not need treatment.

The study gives longer follow-up on nearly 90,000 women who had annual breast exams by a nurse to check for lumps plus a mammogram, or the nurse's breast exam alone. After more than two decades, breast cancer death rates were similar in the two groups, suggesting little benefit from mammograms.

The study did not compare mammograms to no screening at all, as most other research on this topic has done.

Critics of the Canadian study also say it used outdated equipment and poor methods that made mammograms look unfairly ineffective.

The study was published Wednesday in the British journal BMJ.

Breast cancer is the leading type of cancer and cause of cancer deaths in women worldwide. Nearly 1.4 million cases are diagnosed each year. Many studies have found that mammography saves lives, but how many and for what age groups is debatable. It also causes false alarms and treatment of cancers never destined to become life-threatening.

In the United States, a government-appointed task force that gives screening advice does not back mammograms until age 50 and then only every other year. The American Cancer Society recommends them every year starting at age 40. Other countries screen less aggressively. In Britain, for example, mammograms are usually offered only every three years.

Most pessimistic study

The Canadian study has long been the most pessimistic on the value of mammograms. It reported that after five years of screening, 666 cancers were found among women given mammograms plus breast exams vs. 524 cancers among those given exams alone.

After 25 years of follow-up, about 500 in each group died, suggesting mammograms were not saving lives. The similarity in the death rates suggests that the 142 "extra" cancers caught by mammograms represent overdiagnosis — tumors not destined to prove fatal.

The work was immediately criticized. The American College of Radiology and Society of Breast Imaging called it "an incredibly misleading analysis based on the deeply flawed and widely discredited" study. Mammograms typically find far more cancers than this study did, suggesting the quality was poor, the groups said.

In a letter posted by the journal, Dr. Daniel Kopans, a radiologist at Harvard Medical School, described outdated machines and methods he saw in 1990, when he was asked to review the quality of mammograms used in the study.

"I can personally attest to the fact that the quality was poor," he wrote. "To save money they used secondhand mammography machines" that gave poor images, failed to properly position breasts and did not train radiologists on how to interpret the scans, he wrote.

The study leader, Dr. Anthony Miller of the University of Toronto, said it was "completely untrue" that inferior machines or methods were used.

Still, the study highlights the fact that mammograms are an imperfect tool that lead to many false alarms, needless biopsies and treatment of many tumors that would never threaten a woman's life.

Overdiagnosis documented

"Overdiagnosis is not an anomaly in the study from Canada. This has been compellingly demonstrated in research from the U.S. and Europe," said another study leader, Dr. Cornelia Baines of the University of Toronto.

Dr. H. Gilbert Welch, a professor of medicine at the Dartmouth Institute for Health Policy and Clinical Practice in New Hampshire, spoke on the issue at a symposium in December. "Screening is a choice, not a public health imperative. There are trade-offs here," he said.

"The people who stand to gain the most from screening are the people at greatest risk of the disease" — older women who are more likely to have breast cancer and those not too old that they are likely to die of something else, he said.

Death rates from breast cancer have fallen mostly because of dramatic improvement in treatments, doctors have said.

"The better we are at treating clinically evident disease, the less important it becomes to find it early," Welch said.