False Positives in CT Lung Scans

A “false positive” means that a test indicates that a person has a problem, such as lung cancer, when actually they do not. A 21 to 33% false positive rate is extremely high, even a 15% rate is very high. 

This story comes from NPR’s Health Blog

By Richard Knox

Somewhere north of 90 million Americans are at risk of lung cancer because they’re current or former smokers. Many are tempted to get a CT lung scan, costing $300 to $1,000, to see if something’s lurking.

Cars and pedestrians.A CT scan of the lungs picks up signals of cancer. (iStockphoto.com)

The idea has obvious appeal. After all, up to 30 percent of lung malignancies have already spread by the time they’re diagnosed. And CT scans can detect tumors smaller than a pea. Small and early should mean curable, right?

Well, it’s not so simple. An analysis of lung cancer screening results in 3,200 people from the National Cancer Institute finds that 21 to 33 percent of the suspicious nodules found by CT scans are false alarms. They’re not really cancer.

These “false positives” cause a good deal of anxiety, of course. To clear up the mystery, seven percent of patients undergobronchoscopy — a procedure involving a tube threaded into the lungs — or have needles stuck into their lungs to retrieve a biopsy sample. Or they undergo exploratory lung surgery.

Patients who undergo screening with conventional chest x-rays have false positives too, but less often — 15 percent after two rounds of screening.

These false-positive rates may be under-estimates, say NCI officials, because in the study only lung nodules larger than 4 millimeters across (about an 0.16 inches) are followed up. Smaller ones are thought to pose minimal risk. But in typical medical practice, all suspicious nodules are usually investigated.

The results appear in the Annals of Internal Medicine

more via NPR’s Health Blog

By the way, if a person repeats a CT scan for lung cancer multiple times, their risk of having a false positive increases dramatically. For example, let us assume that “Bob” has a total of five CT scans for lung cancer and let’s assume the lower 21% false positive rate.

First, we determine the rate of NOT have a false positive (100% – 21%= 79%) the first time. Then we multiple that for each scan 79% X 79% X 79% X 79% X 79% = ~30%.

This means that after 5 scans the chance of NOT having a false positive goes down to around 30%. Of course, this also means that after 5 scans the chance of having a false positive for “Bob” is ~ 70%. Yikes!!

If the false positive rate is 33% then after 5 scans the probability of “Bob” having a false positive is around 86%.


    Leave a Reply

    Fill in your details below or click an icon to log in:

    WordPress.com Logo

    You are commenting using your WordPress.com account. Log Out /  Change )

    Google+ photo

    You are commenting using your Google+ account. Log Out /  Change )

    Twitter picture

    You are commenting using your Twitter account. Log Out /  Change )

    Facebook photo

    You are commenting using your Facebook account. Log Out /  Change )


    Connecting to %s

%d bloggers like this: