If a simple, painless test can find the world's deadliest cancer when it is smaller than a pea -- and such a test does indeed exist -- shouldn't people who are most at risk have one?

Surprisingly, the federal government, American Cancer Society and a raft of cancer specialists say the answer is "no."

They are waging an uphill battle as frightened current and former smokers rush to get a special kind of X-ray that other physicians are urging for lung cancer detection but that has not yet conclusively been shown to save lives.

A huge federal study is under way to see if it can, and answers may come as soon as next year. In the meantime, it is generating a classic "can't wait for science" stampede.

The clamor rose earlier this month with the deaths of newsman Peter Jennings and "Dallas" star Barbara Bel Geddes, and the news that Christopher Reeve's widow, Dana, has lung cancer.

Patricia Dowds and her husband, David Byrom, psychologists from Long Island, N.Y., are among the many former smokers who had the $300 test recently.

"We do it for our own peace of mind," Dowds said.

No one disputes that the test, called a spiral or helical CT scan, detects lung abnormalities as small as 5 millimeters -- less than a fifth of an inch.

The argument is over whether that is a good thing.

For every cancer these scans detect, many more "false positives" occur -- harmless bumps and lumps leading to painful, expensive and unnecessary biopsies and surgeries. Complications can include lung collapse, bleeding and infection.

"The concern that we have is false positive rates," which range from 25 percent to as high as 60 percent, said Tom Glynn, the American Cancer Society's director of science and trends. "What we don't want to do is create even more anxiety" by backing a test that is so imprecise, he said.

Even when cancers are found, experts argue about whether that is a benefit or a risk. No one knows how many of these tumors are so slow-growing that they pose less of a health threat than the surgery, radiation and chemotherapy used to treat them.

However, screening proponents say there is no mystery about how deadly the disease is and that survival improves the sooner it is detected. Doctors will diagnose lung cancer in about 172,570 Americans this year, and almost as many, 163,510, will die of it.

The morning after Jennings's death, the chairman of internal medicine at St. Louis University, D. Douglas Miller, contacted the public relations staff, urging them to get out the word on lung CT scans' benefits.

Len Horovitz, a lung specialist at Lenox Hill Hospital in New York City, has also been urging the test. Though there is no proof that it improves survival, "I can't believe it wouldn't," he said. "There are half a dozen people I could bring into my office today who are alive for that reason."

As for the risk of unnecessary follow-up procedures, "You don't have to biopsy everything you see," Horovitz said. "You could also elect to watch it" and do a biopsy only if the lump appears to be growing.

Perhaps the biggest proponent of the scans is Claudia Henschke, director of the lung cancer screening program at New York-Presbyterian Hospital/Weill Cornell Medical Center, who published a landmark paper in 1999 reporting that the scans found far more lung tumors than conventional chest X-rays.

"We feel we have evidence that CT screening saves lives," not just from that 1,000-person study but from follow-up work on 30,000 patients around the world, she said.

But her study is not considered the kind of evidence that insurers, policy-makers and scientists require. That will come from the huge randomized screening study the National Cancer Institute commissioned after Henschke's paper appeared. The cancer society, the American College of Radiology and other professional groups are participating in the enormous effort.

As an indication of how eagerly awaited this study is, doctors allowed themselves two years to recruit 50,000 people for it -- and got 53,000, four months ahead of schedule, said Denise Aberle, the University of California at Los Angeles scientist leading the study. It will compare regular chest X-rays to CT scans and see whether either improves survival.

If it shows benefit, "I would feel great," she said, but she will not recommend it until then, "and a lot of my peers and the scientific community are not."

One is Mary Yagjian, coordinator of lung screening at the University of Pittsburgh Cancer Institute, who advised against testing for her own 82-year-old mother, a former smoker at high risk for lung cancer.

"She's scared to death about getting it," and has a brother just diagnosed with cancer and a friend going through a recurrence now, Yagjian said.

"It's important that we complete the definitive studies and get scientific information," said Christine Berg, chief of the early detection research group at the National Cancer Institute. In the meantime, doctors have to do the best they can to advise each patient, considering that person's smoking history and other risk factors.

"It's the physician-patient relationship that's paramount," she said.

Patricia Dowds undergoes a spiral CT scan, which can detect cancer early on. Some doctors criticize the test for the number of false positives it produces.