What is the difference between a screening test and a diagnosis?

A screening test raises a question. A diagnosis answers it. The National Cancer Institute defines cancer screening as "looking for cancer before a person has any symptoms," and is explicit that "screening tests usually do not diagnose cancer. If a screening test result is abnormal, more tests may be done to check for cancer."

NCI's own example is the clearest one. "A screening mammogram may find a lump in the breast. A lump may be cancer or something else. More tests need to be done to find out if the lump is cancer."

What settles it

Diagnosis usually comes down to looking at the cells. NCI notes that "diagnostic tests may include a biopsy, in which cells or tissues are removed so a pathologist can check them under a microscope for signs of cancer."

That is the difference in a sentence. A screening test sees a shadow. A biopsy sees the cells.

Why screening tests raise so many false alarms

The two kinds of test are built for different jobs, and the trade-offs run in opposite directions.

A screening test is given to millions of healthy people. It has to be quick, safe, and affordable, and it leans toward flagging anything questionable rather than missing something real. A diagnostic test is given to a much smaller group with a specific question, and it is allowed to be more invasive in exchange for a firm answer.

That imbalance is why abnormal screening results are common and cancer is not. Two figures from NCI's PSA fact sheet make the gap concrete. About 6% to 7% of men have a false-positive PSA result in any given screening round, and among men who go on to biopsy because of a raised PSA, only about 25% turn out to have prostate cancer.

The two ways screening gets it wrong

NCI names both.

A false positive is "one that shows there is cancer when there really isn't." It "can cause anxiety and is usually followed by more tests and procedures."

A false negative is "one that shows there is no cancer when there really is." Someone who gets one "may delay seeking medical care."

There is a third problem, overdiagnosis, which means finding something real that would never have caused harm in that person's lifetime and treating it anyway.

None of this means screening is a bad idea. It means each test has to earn its place. Lung cancer screening does: NCI's evidence review reports that screening higher-risk people three times, once a year, with low-dose CT reduced lung cancer deaths by 20% compared with chest x-rays. Ovarian cancer screening does not, and the US Preventive Services Task Force recommends against it in women without symptoms.

The rule people most often get wrong

Screening is defined as looking before symptoms appear. So if you already have a symptom, screening is the wrong tool.

A lump, bleeding, a cough that will not clear, or unexplained weight loss calls for a diagnostic workup aimed at that symptom. A normal mammogram three months ago does not answer a new lump today, and saying so to your doctor is the right move rather than waiting for the next scheduled screen.

If a screening result comes back abnormal, ask two things: what the next test is, and when it is booked. Most abnormal screening results are not cancer, and the waiting is usually the hardest part.