IdeaFestival 2016: Data shows overdiagnosis is making people less healthy

Medical technology in diagnosing and treating illness has reached great heights, but new risks have come to light. At IdeaFestival 2016, experts discussed the consequences of overdiagnosis.


Otis Brawley, chief medical officer for the American Cancer Society

Image: Hope Reese/TechRepublic

"Cancer is far more complicated than we ever thought it was," said Otis Brawley, chief medical officer for the American Cancer Society on Wednesday at IdeaFestival 2016 in Louisville, Kentucky.

The conversation, revolving around overdiagnosis, took place between Brawley and medical ethicist Harald Schmidt of the University of Pennsylvania. The question: What are the ethics around cancer screenings?

Brawley, an oncologist and epidemiologist, has spent his career looking at early-detection methods and education in cancer prevention. Since the US established the National Cancer Act in 1971, he said, cancer rates have dropped. In 1991, cancer was at a peak, 215 deaths per 1000 in the population. Over the next twenty years, up to 2011, there was a 23% drop in cancer deaths. Brawley attributes this to smoking cessation programs, and "appropriate screenings for colorectal and breast cancer." Breast cancer deaths, he said, dropped 35-40% in that period.

But what he is concerned about is unnecessary screening.

Screening, it should be mentioned, only applies when patients display no symptoms. There's an overdiagnosis of cancer as a result of the screenings, Brawley reported.

"A pea-sized lesion may or may not be genetically programmed to grow, spread, and kill," he said. "Many who come in for screenings may have cancers that would not go on to spread, cause symptoms or death," he said.

"It's pathological profiling," said Brawley. Why is this a problem? "There are cancers that can be cured but do not need to be cured," he said.

This is a departure from what many believe: That screenings for cancer will save them.

Schmidt, who researches the role of personal responsibility in healthcare, sees major ethical problems with incentivizing mammograms. He looks at policies that encourage people--employees, for instance--to be screened.

SEE: Mammograms lead to breast cancer "overdiagnosis" in 1 million women, study finds (CBS News)

"When should we reward people?" he asked.

It is not a good idea to incentivize having a mammography, Schmidt said. Instead, it is a good idea to incentivize people to think hard about the decision.

But this is not merely Schmidt's opinion. He looked at the effects screenings had on women, in a controlled study. Many women who had screenings received false positives. And there's a risk attached with that, he said.

Of course, there's a powerful drive for people to do what they can to prevent cancer. But he sees problems with the concept of using "survivors" to illustrate the benefits of screening.

"Our understanding of cancer is not ideal," said Schmidt. "The benefits of screening are overstated."

If you screen 1000 women, he said, you prevent only four deaths from breast cancer. And you have 20 over-diagnoses and over-treatments, he said.

SEE: The self medication debate returns (ZDNet)

What we need to share with women is information about the possible benefits and risks of screening. It isn't ethical, he argued, to hide this information from people considering screening.

Why is this an issue? On top of harmful side effects, medical diagnostic radiation causes 1.5% of all cancers in the US, said Brawley. And there are other interests at play. Insurance companies, pharmaceutical companies, and sometimes doctors, benefit financially from promoting screening and treatment.

Yet incentives for women to receive treatment send the message that they are always a good thing, and "short-circuit decision making," said Schmidt.

"That's worrying," he said, "especially if the financial incentive has more attraction for a lower socioeconomic group."

A big problem, both Schmidt and Brawley said, is the tests involved--they just aren't good enough yet, especially in screening for breast and prostate cancer. In terms of mammograms, "we need a test that works," said Brawley. And prostate screenings, Brawley said, received a grade of "F."

"We need to more judicious in our use of these tests," said Brawley.

Also see