Examining Cancer Screenings
Early detection of cancer is the best chance of cure, right? Maybe not.
A growing body of counterintuitive evidence suggests that some cancer-screening tests — including those for breast, prostate, colon, lung and cervical cancer — may be unnecessary and, worse, can do more harm than good.
“Most consumers have the belief that more screening is better, but screenings result in many false positives, which can lead to unnecessary worry, more testing, excessive costs and, worse, actual harm, even death,” said Dr. Gil Welch, a professor of medicine at Dartmouth College and author of Overdiagnosed: Making People Sick in the Pursuit of Health.
“There are two sides to this discussion, but only one gets heard. The other side is that you have to involve many people and harm some to help one,” Welch said.
The collateral damage is what people ignore, Welch said: “We’re overstating the benefit and understating the harm.”
Dr. Otis Brawley, chief medical officer of the American Cancer Society, agrees.
“We have a group that gets too much health care and receives interventions that are not necessary, are of no benefit and can cause harm,” said Brawley, an oncologist and author of How We Do Harm: A Doctor Breaks Ranks About Being Sick in America.
They make a very good point, said Dr. Clarence Brown, emeritus president of Florida’s MD Anderson Cancer Center Orlando, who now heads oncology development for the Orlando Health Foundation.
“It’s a fine line between whether you’re doing good or harm,” Brown said. Still, he considers himself “a proponent of screening.”
What’s important, said Brown, is to put risks in perspective and to look at an individual’s chances of developing a particular cancer.
Consumers have been taught, and logically assume, that early screenings lead to early detection, which leads to better outcomes.
“We try to find more cancers because we think if treated early that will improve outcomes,” said Dr. Kenny Lin, associate professor of family medicine at Georgetown University School of Medicine. “But there is no evidence whatsoever that just because you find more cancer, that will lead to improved outcomes. It often leads to worse outcomes, especially when screenings find cancers that aren’t lethal.”
The problem, experts agree, is knowing which cancers will kill you and which you can live with. Some cancers don’t matter. Some regress.
Much excess testing is done because patients insist, Brawley said. Plus, screenings are good for business. “Patients need to be wary of the interests involved,” Welch said. “Health centers need to find ways to pay for the expensive diagnostic equipment, and screenings are a good way to capture business.”
Brown agrees that physicians probably do over-treat because “we have a fear of being sued.”
However, he said, “over the 35 years I have watched cancer screening evolve, I know we have caught cancers through screening. When you’re on the firing line, it’s better to say, ‘Have that test.’ “
Finding cancers earlier also leads to what researchers call lead-time bias, which skews survival rates. Detecting a cancer early that a patient dies from anyway doesn’t mean he lived longer, only that he lived longer with knowledge of the disease, experts say.
The problem with the guidelines is that one size doesn’t fit all, Brown said. “Some of this is about how medicine is an art versus a science. We need to use a lot of intuition and conversation instead of just screening and treatment.”
Which Tests to Get
Consider these points for five common cancer screenings:
Proponents of regular mammography say mortality rates from breast cancer have dropped 30 percent in 30 years because of better screening.
It’s not screening but better treatment that is saving lives, said Welch.
“It’s both,” said Brown.
Welch doubts mammography screening is helping anybody right now, after factoring in the effects of chemo, radiation, surgery, anxiety and downtime. “The data suggest that, at most, 13 percent of those diagnosed with breast cancer have been helped. That means 87 percent have not,” Welch said.
Brown advocates for mammography but not for annual screenings of every woman 40 or older. But that’s what the American Cancer Society recommends, as well as MD Anderson. Brown advises women who have no risk or history of breast cancer, including his own daughters, to have a baseline at age 40. If all is normal, have another at age 45. Have another at age 50 and every other year after that until 69.
“After age 70, a mammogram is not necessary. Continuing to screen those women is not ever right,” Brown said.
Studies have also found that, because of the low levels of radiation mammograms emit, having many over a lifetime appears to contribute to breast cancer. All this changes if a woman finds a lump. “Then I’d come right in,” Welch said.
∎ PSA test
Prostate cancer is a disease many more men die with, not of, said Brawley.
No good study shows that prostate-cancer screening, known as a PSA blood test, saves lives, Brawley said. “However, much evidence exists to indicate that men who get treated have a high incidence of infection, incontinence and impotence.” They also have a 1.5 percent chance of death from treatment.
A 2011 study compared men 55 to 60 who were screened and not screened during 10 years. The study showed that to save one life, doctors would have to screen 1,000 men for 10 years. Four out of 1,000 screened men died of prostate cancer, compared with five of 1,000 unscreened men, Brawley said. That supports their claim of a 20 percent reduction in mortality, which sounds more impressive than the real numbers reflect.
Last month the American Urological Association raised its age guidelines for PSA tests from 40 to 55. The group recommended that after age 55, men make an “informed decision” about the potential benefits versus the harms of biopsy and surgical side effects.
Statistics show that, although one in 30 men dies of prostate cancer, one in six get diagnosed with it. Those who treat prostate cancer don’t know which tumors will become aggressive, said Dr. Vipul Patel, a urologist at Florida Hospital who treats prostate cancer and disagrees with the new recommendation.
“We understand there is morbidity. But what’s lost in the debate is that we’re taking away a man’s right to know he has prostate cancer,” Patel said.
∎ Lung-Cancer Screening
As the No. 1 cancer killer, lung cancer kills more Americans than the next five cancers combined. Smokers have the highest risk.
The National Lung Cancer Screening Trial, sponsored by the National Cancer Institute, randomly divided 52,000 smokers 55 or older into two groups. Half received a lung-cancer screening using spiral CT every year for three years. The other half were not screened.
During the next 10 years, there were 87 more deaths in the unscreened group. But for every 5.4 lives saved, one person died as a result of the screening, and two were put into intensive care because of complications from follow-up testing.
The American Cancer Society therefore does not recommend routine lung-cancer screenings for Americans at average risk. But it does recommend screenings for those patients ages 55 to 74 who have a 30-year smoking history and “who understand the harms and benefits and want to get screened,” Brawley said.
∎ Colorectal Screening
Because colorectal cancer is the second-leading cause of cancer death among Americans, doctors are increasingly aggressive about screening for it. The U.S. Preventive Services Task Force recommends screening begin at age 50.
The task force also says any of three screening methods will do. That message is lost on Americans, who opt for the most invasive and expensive screening test: colonoscopy, Brawley said. They’re convinced that colonoscopy, in which a physician scopes the colon with a camera, is the king of procedures, Brawley said. Sigmoidoscopy, which gives doctors a view of only part of the colon, is the second choice. Third is the low-tech stool sample.
“We have no studies that show colonoscopy is better than stool testing,” Brawley said. “The doctors who say it’s better tend to be those making money from it.”
Statistically speaking, stool testing is a very good first step, harmless and, at about $30, much cheaper. A colonoscopy costs about $3,000. Patients who have a suspicious result from a stool test should then get the more advanced test.
“All the colorectal screenings work,” Brawley said. “What doesn’t work is not doing them. Pick one and get it done.”
∎ Pap Smears
Although many doctors still tell women to have an annual Pap smear, the cancer society recommends that after age 21, women have the cervical-cancer screening every year for three years. After three normal results, she can drop down to every three years until age 65. And if all remains normal, not at all after that.
Brown adds that women should also have an HPV test at the same time to screen for human papillomavirus, which also causes cervical cancer.
Researchers who have studied the Pap test’s effectiveness found that half of the women who died of cervical cancer had never had a Pap test, and more than 90 percent had not had one in 10 years.
The test is important, but having it every year is not, Brawley said. “Four thousand women a year die of this disease, which is overwhelmingly preventable.”