Screening: How overdiagnosis and other harms can undermine the benefits

“Screening saves lives.”

We’ve all seen that seductive message.

The idea behind screening — catching disease early when it’s easiest to treat — sounds good, but it can be misleading.

Screening means testing people who don’t have signs or symptoms of the disease. It might save some people from a fate such as dying of cancer, if it detects treatable cases early enough to make a difference in the outcome.

But it almost surely will subject many more to an array of physical, emotional, and financial harms. The trade-offs between potential benefits and potential harms is a numbers game that is often difficult to win.

It’s also important to note that most screening tests do not help in preventing disease. They can only identify potential disease or risk factors.

So maybe instead of “Screening saves lives,” a more apt assessment would be this from Sir Muir Gray of the UK: “All screening programmes do harm; some do good as well.”

That is a far more accurate and helpful perspective for consumers and journalists to keep in mind.

Nevertheless, the idea that screening saves lives has been drummed into the public psyche, thanks to deceptive slogans that exaggerate the benefits of screening. Here are a few we’ve documented:

What’s does the evidence actually indicate?

The U.S. Preventive Services Task Force (USPSTF) — a body of independent experts in evidence-based medicine — strongly recommends broad-based screening for just a handful of conditions.

The panel’s “grade A” recommendations — deemed to have a high certainty of substantial net benefit — include screening for high blood pressure, HIV infections, cervical cancer, and colorectal cancer in adults ages 50 to 75.

There are also areas where screening is recommended, but where the potential benefit, or evidence of a potential benefit, isn’t as strong. An example of a “grade B” recommendation is breast cancer screening for women ages 50 to 74.

Prostate cancer is a “grade C” recommendation, meaning the net benefit appears small and men should talk to their doctor before choosing to be screened.

For most conditions the task force has studied, either there’s insufficient evidence to recommend screening, or evidence shows benefits don’t outweigh harms.

So why is screening so popular?

While clinicians genuinely hope screening prevents suffering from horrible diseases, there are also medical device and drug companies, advocacy groups, hospitals, and medical specialty groups that benefit from screening. Those vested interests sometimes promote imbalanced pro-screening messages that don’t get scrutinized in news stories.

In some cases, medical specialists have pushed back against evidence-based guidelines. The American College of Radiology once equated the USPSTF’s breast cancer screening recommendations to rationing care, and some urologists slammed the USPSTF’s 2012 recommendation against prostate cancer screening (which has since been amended) because there wasn’t a urologist on the panel.

Defenders of the task force’s work have argued that primary care doctors, who comprise most of its members, are best positioned to weigh the merits of screening, since screening decisions typically occur in primary care settings.

It’s interesting to note that some groups that advocated early detection have recently adopted more cautious stances about screening. In 2013 the American Urological Association brought its own guidelines in closer alignment with those of the USPSTF, and in 2015 the American Cancer Society relaxed its recommendations on screening mammograms.

What are some harms of screening?

Screening means testing millions of people, the vast majority of whom won’t have life-threatening disease but do face potential harms.

Screening entails cost, time away from work and other activities, and sometimes physical discomfort. There can also be harms from the screening procedure itself. For example, x-rays expose people to radiation, and colonoscopies can cause bleeding or in rare cases serious injury.

Another common harm is a false alarm — a screening test result that’s positive but doesn’t turn out to be the disease. False alarms can lead to follow-up tests, including invasive biopsies, as well as emotional fallout.

The more screens you have, the more likely it is you’ll experience a false alarm. About half of women who get annual mammograms over a 10-year period will have a false-positive finding.

While it might be tempting to discount the psychological toll of false alarms, studies suggest they can lead to significant anxiety. Even three years after being cleared of breast cancer, women who experienced a false positive have been found to have more psychological symptoms such as trouble sleeping and feeling less attractive than women with only normal results.

In addition, screening can offer false reassurance, since even people who are screened can go on to develop the disease. In fact, long-term studies have shown that mortality rates from some cancers have not changed despite the introduction of screening.

(More information about false positives and false negatives is available in our toolkit on understanding medical tests.)

What is overdiagnosis?

The most important harm of screening is that it can cause you to be diagnosed and treated for something that will never cause symptoms. Finding an abnormality that doesn’t matter is called overdiagnosis, and it’s extremely common.

In fact, many of us live with what some experts call “turtles” — slow-growing cancers that don’t need to be treated. Autopsy studies have found that many people harbor small thyroid cancers, breast cancers, skin cancers, and prostate cancers, to cite a few examples.

In fact, trivial lesions are so problematic that there has been a call to stop labeling them as cancer.

Turtles contrast with other cancers such as “birds,” which grow too quickly to benefit from screening, and “bears,” which might be detected with screening.

What happens when people are overdiagnosed?

Doctors often can’t tell which cancers will progress and which will not, so the tendency is to treat them all. But all treatments come with potential harms.

In prostate cancer, a study published in 2009 estimated that since a screening tool called the prostate-specific antigen (PSA) blood test was introduced in 1986, more than 1 million men were treated for a disease that likely would never have harmed them.

Those men had surgery and radiation therapy that led to terrible complications. According to data compiled by the USPSTF:

  • About 63% of men who have surgery or radiation treatment for prostate cancer become impotent.
  • About 19% of men who have surgery or radiation treatment for prostate cancer lose bladder control.

Similarly, an estimated one in five women diagnosed with breast cancer after a mammogram has a cancer that would not affect her in her lifetime, resulting in tens of thousands of women annually in the U.S. receiving non-beneficial treatments such as surgery, radiation, and chemotherapy.

Potential harms of those treatments include breast removal, pain and scarring, hair loss, nausea, and skin burns, wrote Karuna Jaggar, executive director of Breast Cancer Action.

She noted a host of other consequences:

“Less commonly discussed, many women experience a range of long-term effects that include physical and health harms such as disability, neuropathy, lymphedema, heart disease, infertility, and secondary cancers; financial consequences from medical debt to un- and under-employment; the psychological toll of having been diagnosed with cancer; and other quality of life impacts of treatment such as issues with sexuality, “chemobrain,” and others.”

Why do the harms of screening get overlooked?

Disturbingly, people who are overdiagnosed and overtreated don’t know it; they are unaware that they are the big losers of screening. That’s because it’s not always possible to tell whether a cancer found via screening would have progressed.

Yet many people prefer to believe the more reassuring message that they were “saved” from a deadly cancer.

It’s that hopeful framing that often captures media attention. Take comedian Ben Stiller’s assertion that a PSA test prevented him from dying of prostate cancer, or the testimonial of an Oklahoma mayor that breast cancer screening with 3-D mammography “saved” her life.

This result is a “popularity paradox” in which people who have been harmed by screening actually wind up being the ones advocating for it. As researchers Gray and Angela Raffle, MD, put it: “The greater the harm through overdiagnosis and overtreatment from screening, the more people there are who believe they owe their health, or even their life, to the programme.”

Why can make screening programs look better than they are?

Screening can create an illusion that people with the disease are living longer.

This works in two ways. First, screening can detect a lot of trivial cancers that won’t ever lead to death. Second, it can generate earlier diagnoses that extend the period between cancer detection and death without necessarily prolonging life, which is called lead time bias.

Here’s a graphic used by publisher Gary Schwitzer that explains how lead time bias works:

Because of this, it’s important to be wary of five-year survival data, which is not a useful measure.

A famous example of that deception was Rudy Giuliani’s faulty comparison of five-year prostate cancer survival rates in the U.S. and Britain, in which he claimed the higher American survival rate was attributable to better care in the U.S. versus European “socialized medicine.” In fact, the difference was attributable to the higher number of small, inconsequential cancers being detected — and overtreated — as a result of aggressive prostate screening in the U.S.

What about shared decision-making?

More and more, health authorities urge patients to make their own informed decisions about screening, with support from their doctors.

But that’s a challenge. Such discussions are time-consuming, doctors aren’t always equipped with appropriate data, and physicians are often incentivized to write screening referrals, not to discuss options.

Take lung cancer screening. Experts advise physicians and patients to discuss the potential risks and benefits of lung cancer screening with heavy smokers, but researchers who analyzed office visit discussions between doctors and patients found the quality of the conversation about lung cancer screening was “poor” and discussion of the potential harms of screening was “virtually nonexistent.” Doctors spent less than a minute, on average, discussing the issue.

What should journalists report?

News reporting should strive to show harms as well as benefits of screening and reflect the balance between the two.

A good example is a lengthy New York Times story on lung cancer screening that noted lower mortality rates with earlier diagnosis and treatment, but also described the “considerable pain” of a 76-year-old woman who had a diagnostic surgery after a CT scan found a “hot spot” on her lung. No cancer was found.

The story said close to 40% of participants in a national trial had positive results from at least one of three CT scans, “but more than 96 percent of these nodules weren’t cancerous.”

It also reported that false positives usually require additional scans over several years to determine whether the nodules are malignant, leaving patients in a state of uncertainty and potentially leading to invasive biopsies that might trigger dangerous complications such as a collapsed lung or bleeding.

Another Times story on misdiagnoses in breast cancer told the story of a woman whose breast was partially removed following an incorrect breast biopsy result.

That type of critical reporting can help to offset the ubiquitous sloganeering that “screening saves lives.”


Harvard University video:  Mammography: How to make the right decision for you

Journal of the American Medical Association: The Harms of Screening

National Institutes of Health: To Screen or Not to Screen?

National Cancer Institute: What is overdiagnosis?

New York TImes Magazine: The Perils of Prevention by Shannon Brownlee

U.S. Preventive Services Task Force recommendations