TMI in Medicine: the Problem of Overdiagnosis

You have no symptoms. The doctor suggests running some "routine" tests. Should you go through with them?

Or maybe you feel healthy. Then you're diagnosed with cancer. You might not dream of asking: "What happens if I do nothing?"

But a growing number of experts say these are exactly the type of fundamental questions patients should think more deeply about. "We look harder, we find more," says Dr. H. Gilbert Welch, a professor of medicine at The Geisel School of Medicine at Dartmouth College in Hanover, New Hampshire. "So, what in the past was not a diagnosis, all the sudden became one, and that has very powerful effects on both how much disease we think there is and how well we think we're doing [at treating it]."

Welch, a general internist, has studied and written extensively about excessive medical treatment and overdiagnosis. He sees a downside to all the data gathered by the enhanced medical imaging technology used today, which detects smaller and smaller abnormalities in the human body. Welch explains that the term overdiagnosis refers not to something that was misdiagnosed -- like a stomach cancer mistaken for an ulcer -- but to a broadened, if still technically correct, diagnosis of an issue that would have likely never harmed a patient had it gone undiscovered and untreated.

"For years, pathologists have noticed that people who die from other things harbor small cancers. So that creates a little conundrum," Welch says. "If there are cancers in people who die from something else, what does that really mean if we find cancer?"

This wasn't a question patients or providers wrestled with decades ago when cancer was only detected as tumors grew to the size of baseballs and the disease spread to other parts of the body, he adds. "All those cancers clearly mattered. But the minute we started looking early for cancer, all the sudden we began to identify a new subset that has always been there, but we're now able to detect it," Welch says.

Experts like Welch contend that finding the small spot of cancer that now has a patient freaked out often hurts more than it helps, which runs counter to the indoctrinated notion that early detection is always best.

"It's important to remember that even what appear like harmless little tests can ultimately bring you or your loved ones great harm, as they can open the door to a potential cascade of further unnecessary tests and treatments," writes Dr. Ray Moynihan, a senior research fellow at Bond University in Australia, in an email. Moynihan, who has also studied and written extensively about overdiagnosis, recommends people learn as much as they can about the risks and benefits of a test or treatment, and the evidence for it, before moving forward with it.

Concerns about overdiagnosis span the globe as health technologies advance. Bond is a member of the scientific steering committee planning the third annual Preventing Overdiagnosis international scientific conference to be held next month at the National Institutes of Health in Bethesda, Maryland.

"It's often very hard to know if you have been overdiagnosed or overtreated after the fact. The best approach is to try as much as possible to avoid tests and treatments and diagnoses that you don't really need in the first place," Moynihan adds. "It's often not easy to know what's warranted and what's not, but having a strong, healthy skepticism is very valuable -- and asking as many questions as you can before you are tested, diagnosed or treated."

Welch adds that providers and patients should -- and safely can -- dial down the urgency on treatment in most cases where a patient doesn't have an acute medical issue, or isn't otherwise symptomatic following a diagnosis. "I think maybe one of the most important cautions is to take your time, and get other opinions. We have overstated the need to act fast -- even in cancer," he says.

Dr. Sharon Giordano, chair of the Department of Health Services Research at The University of Texas MD Anderson Cancer Center in Houston, agrees that patients can and should take time to explore their treatment options after a cancer diagnosis.

"There are a few cancers that move very quickly and are very aggressive and it's important to start treatment very quickly," says Giordano, such as some leukemias and certain subtypes of lymphoma. "But for the vast majority of cancer diagnoses, patients really do have the time to make sure the diagnosis is correct and that they're getting the correct treatment."

While some raise concern that early detection frequently leads to overtreatment without meaningfully moving the needle on reducing total cancer deaths, Giordano says that wider screening with enhanced technology can save lives and improve outcomes. "If we find a cancer when it's still -- it's smaller, it's early stage disease, it's easier to treat, requires less intensive treatment and is more curable," she says.

However, she adds that patients -- particularly older ones -- should consider whether they're willing to undergo treatment, such as radiation or surgery, before getting screened for cancer.

The harm-benefit analysis differs by study and cancer type. While much debate swirls around whether getting screened for prostate cancer is worth it, for example, wider support exists for colon cancer screening. Overall, in cancer treatment, much gray area remains.

"If you find a non-invasive breast cancer, still we're going to treat it because we have concerns that could be the cancer that goes on to become an invasive cancer and ultimately could be potentially life-threatening," Giordano says. "But we know that we need better science to try to be able to distinguish which are the cancers that are really going to cause trouble, and which aren't. Because we probably treat a lot that were never going to be troublesome to prevent the few that are going to be."

Though cancer is an anxiety-provoking diagnosis for patients, prompting many to choose aggressive treatment, Giordano notes that some patients, like those with slow-growing prostate cancer, may opt for active surveillance. Essentially, health providers keep an eye on the cancer without treating it.

The problem with cancer tests, Welch argues, is that they're generally really good at finding the slow-growing cancers that aren't likely to spread enough to do harm, while many faster-growing cancers often advance too fast for regular screening to prevent a patient's death.

Not that overdiagnosis concerns stop with cancer.

For example, heart structural abnormalities, like a potentially life-threatening abdominal aortic aneurysm, are easier to detect with today's imaging technology.

But that leads to many operations being performed unnecessarily, Welch asserts. "That's kind of the trade-off in trying to get ahead of diseases. You end up treating some people needlessly in the hopes that you can help others."

Genetic testing has also been slammed for its potential to yield needless diagnoses and treatments, despite the promise of personalized medicine.

Then there's what you can find over-the-counter or through the mail to test yourself. "There's a huge industry ... selling and promoting testing of people who feel well," Welch says, that extends beyond medical offices.

Still, health care minimalists and advocates for a more aggressive approach tend to find some middle ground in asserting that patients need to take the reigns, ask questions and know their options. Even if plenty of disagreement remains over whether they should undergo treatment -- or have gotten that test in the first place.

"I want to be clear, I believe medicine can do some real good for people, and it's most likely to do so when they're acutely sick or injured," Welch says. "But you ought to be a little more skeptical when you're well, or when you're at the very end of life, about what you're trying to do here, and recognize that everything we do has some potential for benefit, but it also has some potential for harm. And so, I think a little caution is in order."

Michael Schroeder is a health editor at U.S. News. You can follow him on Twitter or email him at mschroeder@usnews.com.