More Cancer Tests Mean More False-Positive Results
While that conclusion may seem like common sense, it's not something that patients or doctors often consider, suggest the authors of a study in the May/June issue of the Annals of Family Medicine.
False-positive results from routine cancer screening can cause undue worry and in some cases lead to unnecessary biopsies or treatments, experts note.
In the new study, "after 14 tests total, over half of the people in our study had a false-positive result. No test is perfect, and you would expect to see it go up over time, but how rapidly the risk went up was surprising," said the study's lead author, Dr. Jennifer Croswell, acting director of the office of medical applications of research at the U.S. National Institutes of Health in Bethesda, Md.
"It's important to know ahead of time the risk of false-positives," she said. "Screenings have to be thought of like any other medical intervention and it's important to have the discussion about the risks and the benefits."
Croswell and her colleagues reviewed data from the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial, which included nearly 70,000 participants. The study volunteers were between the ages of 55 and 74, and were randomly selected to receive either normal care or more intensive screening.
Those in the normal-care group were offered screening through their own private physicians as usual. Those in the intervention group were offered a baseline chest X-ray along with a yearly follow-up for two years for non-smokers and three years for smokers to check for lung cancer; a baseline flexible sigmoidoscopy to check for colorectal cancer, along with a three- or five-year follow-up; annual tests for cancer antigen 125 (CA-125) and a transvaginal ultrasound yearly to check for ovarian cancer in women; and an annual digital rectal exam to check for prostate cancer in men, as well as a prostate-specific antigen test.
No information was included on mammography for women in either the intervention group or the usual care group.
For those who had 14 screening tests during the study period, the cumulative risk of having at least one false-positive was 60.4 percent for men and about 49 percent for women. The cumulative risk of having to undergo an invasive diagnostic procedure due to a false-positive test result was almost 29 percent for men and just over 22 percent for women.
Croswell said the researchers weren't sure why the false-positive rate was higher for men, but it likely had something to do with the tests that were studied.
"The fact is that screening tests aren't diagnostic," said Robert Smith, director of cancer screening for the American Cancer Society. "A certain risk of false positives and false negatives are to be expected. You can try to very aggressively reduce the false-positive rate, but then you also cut into cancer detection rates," he explained.
"Most people place a higher priority on finding cancer early than preventing false-positives. I think the public often understands that false-positives happen," said Smith. But, he added, "We really need to do a better job of explaining to adults that there is a plus side and a down side to screening. We need to do a better job of letting people know what to expect, and that screening tests aren't perfect."
In another study in the same issue of the journal, researchers at the University of Auckland, New Zealand, take a different look at medical errors, and focus on the types of errors that patients make. While the study didn't attempt to rank which medical errors patients are most likely to make, the researchers did try to identify the types of errors patients might make.
They found that patient errors could be classified into two broad categories: action or mental errors. Action errors are related to patient behavior and may include coming late to an appointment or not following directions when taking medications. Mental errors are when a patient has compromised thought processes, memory problems or knowledge deficits. Examples of mental errors include forgetting to take medications or a failure to understand the doctor's instructions. The authors suggest that future research should try to account for these errors and find ways for clinicians and patients to work together to reduce errors.