For instance, three-quarters of the more than 400 doctors surveyed believed better survival rates prove screening is a lifesaver although that's not the case, researchers say.

And nearly half thought early detection translates into saving lives -- another common misperception.

"This is really unfortunate because one of the things we always say is, 'Discuss it with your doctor,'" said Dr. Otis Brawley, chief medical officer of the American Cancer Society. "This is evidence that your doctor doesn't know."

Brawley, who wasn't involved in the new survey, said the lack of statistical sophistication among doctors is part of the reason why prostate cancer screening, now a controversial practice, has become so widespread.

"Misunderstanding of statistics has been fueling a great deal of screening," he told Reuters Health. "Unfortunately we have all been taught that the way to deal with cancer is to find it early. As we have learned more and more about cancer, we are starting to find that that's not true."

The reason is that screening automatically increases survival rates, because finding a tumor early means people live longer with their cancer diagnosis than if they had waited until they had symptoms to see a doctor -- regardless of whether or not anything is done to treat them.

In some cases, such as slow-growing prostate cancers, the tumor might never have bothered them in the first place. That means screening, and the further tests and treatments that might follow, would have led to costs and potential side effects without any benefit to the patient -- a phenomenon called overdiagnosis.

"For helping people understand if screening works, survival rates are misleading," said Dr. Steven Woloshin, of Dartmouth Medical School in Hanover, New Hampshire, who worked on the new survey.

He told Reuters Health that death rates gleaned from clinical trials are the only reliable way to judge if a screening test is effective. But organizations that promote screening, such as the breast cancer charity Susan G. Komen for the Cure, tend to prefer survival rates, which sound more impressive.

"There are a lot of actors here that could do a better job," Woloshin said.

The new findings are based on responses from 412 U.S. primary care physicians surveyed in 2010 and 2011, and were published Monday in the Annals of Internal Medicine.

When asked if a better survival rate proves that screening saves live, less than a quarter of doctors answered correctly that it doesn't, while 76 percent believed it did.

Forty-seven percent of doctors also mistakenly thought early detection of cancers is proof that screening saves lives, while 49 percent correctly answered that that's not the case.

"It shows that there is a lot of confusion out there," said Woloshin. "It is sort of understandable, because I don't think doctors necessarily get a lot of education about statistics."

When Woloshin's team quizzed doctors about two hypothetical screening scenarios, the responses were just as discouraging.

Doctors were three times more likely to recommend a test that increased the (irrelevant) five-year survival rates from 68 percent to 99 percent than to recommend a test that slashed the much more important death rate from 2 in 1,000 people to 1.6 in 1,000.

"Physicians clearly do not understand how to interpret cancer screening statistics themselves -- expecting them to communicate this information to patients is a stretch," Dr. Virginia A. Moyer, who chairs the government-backed U.S. Preventive Services Task Force, said in an editorial on the new poll.

Dr. Elie A. Akl of the University at Buffalo in New York, who has studied risk communication, said better guidelines for how to report medical studies are important.

"The solution might not (just) be further education of physicians but also standardized naming and reporting of statistics in a way that makes intuitive sense, not just for clinicians, but also for journalists and the lay public," he told Reuters Health by email.

In the meantime, patient advocacy groups can help educate consumers about the questions they should be asking their doctors, said Brawley, the author of How We Do Harm: A Doctor Breaks Ranks About Being Sick in America.

"What are the harms of this screening test? What are the potential benefits? If the patient gets an answer of increased five-year survival, that's an indication that their doctor doesn't know what they are talking about," Brawley said.

SOURCE: bit.ly/an7XRm Annals of Internal Medicine, March 5, 2012.

• A Permanent Economic Emergency