False Positives and Overdiagnosis: The Trade Offs of Screening

You got a call back after a mammogram, or a CT scan showed a "spot," and you spent days waiting for more tests. It turned out to be nothing. Or you wonder why your clinician talks about "harms" of a cancer test. This guide explains the hidden side of screening: false positives, overdiagnosis, and how to read the numbers.

What screening is meant to do

Screening means testing people who feel well, to find disease early. The goal is to prevent serious illness or death. Good examples include colon cancer screening, which can find and remove polyps, and blood pressure checks.

But screening also tests many people who will never get the disease. Even a small rate of error, across millions of people, adds up.

The main harms of screening

False positives

A false positive is a result that suggests disease when there is none. It leads to more tests, such as extra imaging or a biopsy, before the all-clear.

False positives are common, especially when screening is repeated over years.

• Mammograms: in a large US study, after 10 years of screening every 2 years starting at 40, about 42 percent of women had at least one false-positive recall. With yearly screening it was about 61 percent [2].
• Lung CT: in the US National Lung Screening Trial, about 24 percent of people had at least one positive scan over 3 rounds, and about 96 percent of those were not cancer [3].
• PSA: most men who have a biopsy because of a raised PSA do not have cancer.

False positives can cause worry, time off work, costs, and sometimes complications from biopsies.

False negatives

A false negative is a normal result when disease is present. It can give false reassurance. This is why new symptoms always need checking, even after a normal screen.

Overdiagnosis

Overdiagnosis means finding a real disease that would never have caused symptoms or death in your lifetime [4]. Some cancers grow so slowly that a person dies of something else first.

The problem is that, for one person, doctors usually cannot tell which cancers are overdiagnosed. So many are treated. That treatment, called overtreatment, brings side effects without benefit.

Estimates of overdiagnosis vary widely by study method. One review estimated that about 25 percent of breast cancers found by mammography and about 60 percent of prostate cancers found by PSA testing in trials were overdiagnosed [4]. Other researchers estimate lower numbers. Most experts agree that overdiagnosis is real, and that it is more common with prostate, thyroid, breast, and lung screening than with colon screening.

Other harms

• Complications from follow-up tests, such as bleeding after a biopsy.
• Radiation from some imaging tests.
• Cost and time.

What doctors look at: reading the numbers

The most useful numbers show what happens to a group of people who are screened compared with a similar group who are not [1].

Relative versus absolute benefit

Here is a made-up example of the same result told two ways.

• Relative reduction: "Screening lowers cancer deaths by 20 percent."
• Absolute reduction: "Screening prevents about 3 cancer deaths for every 1,000 people screened."

Both can describe the same result. Absolute numbers are easier to compare with harms. For example, in a European prostate screening trial, PSA screening lowered prostate cancer deaths by 13 percent over 23 years. In absolute terms, about 1 death was prevented for every 456 men invited [5]. Screening also led to more prostate cancer diagnoses.

Why survival rates can mislead

Five-year survival often looks better with screening, even when screening does not help anyone live longer [1].

• Lead time: if a cancer is found earlier, survival time is counted from an earlier date. People may seem to live longer with the disease without actually living longer.
• Overdiagnosis: adding slow, harmless cancers to the count makes survival look better.

The best measure of benefit is the death rate from the disease in the whole screened group, compared with an unscreened group [1].

Canada and the US

Both countries' guideline bodies now pay close attention to harms.

• The Canadian Task Force on Preventive Health Care built "1,000-person" tools that show benefits and harms in absolute numbers. Its 2024 draft breast screening advice said information should be shared in absolute numbers so people can make their own choice [6]. The Task Force ended its mandate in March 2026, and its tools remain posted for reference.
• The US Preventive Services Task Force grades each service by the balance of benefits and harms. Some services, such as PSA screening for men 55 to 69, are graded as an individual choice because the balance depends on personal values.

Questions to ask your clinician

• Out of 1,000 people like me who get this test, how many avoid serious illness or death?
• How many will have a false positive?
• How many might be overdiagnosed and treated without benefit?
• What happens if my result is abnormal?
• How do my age, family history, and health change the balance?
• What happens if I choose not to screen, or to screen less often?

When to get care quickly

Screening is for people without symptoms. If you have new or worsening symptoms, such as a lump, bleeding, unexplained weight loss, or a lasting cough, see your clinician soon, even if a recent screen was normal. Call 911 for chest pain, severe breathing trouble, or heavy bleeding.

Bottom line

Screening can save lives, but it is not free of harm. False positives are common, and overdiagnosis can lead to treatment you did not need. Absolute numbers give the clearest picture of the trade-offs. Ask your clinician for those numbers, and choose the screening plan that fits your risk and values.

Primary AI can show you the guideline and study sources behind this topic in plain language, so you can bring better questions to your appointment.