Too much of a good thing: Overdiagnosis to get its due

Increasingly sophisticated imaging and lab tests, along with campaigns to encourage screening tests, have meant that cancers and many other conditions are discovered much earlier in the disease process. Catching a disease early can mean more effective and less costly treatment.

But there’s growing recognition that moving disease recognition upstream can result in overdiagnosis—finding an asymptomatic condition that would not harm a person’s health—and the related problem of overtreatment—treatment of conditions that are overdiagnosed.

Experts have been debating overdiagnosis and over­treatment for years. But the issue seems to be cresting heading into 2016. The American Cancer Society and United States Preventive Services Task Force (USPSTF) issued new mammo­graphy guidelines this year that reflect, in part, concern about overdiagnosis. Women are now being advised to start getting mammograms when they are older (at 45, says the cancer society, and at 50 says the USPSTF) and less often. A study published in JAMA showed large decreases in prostate cancer screening and diagnosis after the USPSTF came out against widespread PSA screening in 2012.

The problem with overdiagnosis is that it leads to unnecessary treatment and spending and, in some cases, real harm to patients because of the risks and side effects of treatment, says H. Gilbert Welch, MD, a professor of medicine at the Dartmouth Institute for Health Policy and Clinical Practice and a leading voice on the topic. Welch is the coauthor of a 2011 book titled Overdiagnosis. He came out with another book this year, Less Medicine, More Health that builds on the overdiagnosis critique.

The work of Welch and others is gaining traction. Consumer ReportsBMJ, and Dartmouth, among other organizations, sponsor an annual conference called Preventing Overdiagnosis.

In 2016, look for more debate about overdiagnosis, particularly with respect to cancer. Prostate cancer screening might continue to make headlines. The USPSTF collected comments on a draft research plan for prostate cancer screening this year and one of the questions posed is whether new tests (genomic or urine testing) might do a better job of identifying prostate cancer that is more likely to cause symptoms or lead to advanced disease. Another development to keep an eye on: proposals that would rename low-risk cancers, such as ductal in situ carcinomas. The idea is to take cancer out of their names, so patients and their doctors might be less inclined to treat them aggressively.

Welch says American health care is in the midst of a change that recognizes that early detection comes at a cost. “While it comes with the potential to benefit a few, it also comes with the side effect of intervening on many who cannot benefit and harming some of them in the process,” he says. “To be clear, it’s a course correction that will take years.”