Patients whose physicians place labels on their test results - such as normal or abnormal, positive or negative - are inclined to make different decisions about their health care than those who receive the same results solely in terms of numbers and other quantifiable measures, according to a new study from the University of Michigan Health System and the VA Ann Arbor Healthcare System.
The study - which looked at interpretive labels used in prenatal genetic screenings - suggests that health care providers should carefully consider the language they use when describing risk levels and test results with patients, the authors say. The study appears in the November issue of the American Journal of Obstetrics and Gynecology.
"It may seem like an innocuous practice to label a test result as positive or negative, abnormal or normal. But our research shows that using these labels can change what a patient will do, even when they have received the numeric information about their level of risk," says lead author Brian J. Zikmund-Fisher, Ph.D., a decision scientist and research assistant professor at the Center for Behavioral and Decision Sciences in Medicine (CBDSM). The center is a joint program between the U-M Health System and the Veterans Affairs Ann Arbor Healthcare System.
Zikmund-Fisher draws an analogy to a practice that consumers encounter regularly at clothing stores and supermarkets. "If you see a shirt that costs $20, it shouldn't matter if the store puts a sticker on it that says, 'low price,' " Zikmund-Fisher says. Yet consumer habits suggest that the "low price" sticker might indeed affect many shoppers' decisions.
The study examined women's perceptions of their risk for fetal chromosomal problems in a hypothetical scenario that involved pregnancy and amniocentesis, a prenatal test that provides health information about the fetus based the extraction of amniotic fluid.
Participants in the study were given the same numerical risk information regarding potential chromosomal problems with the baby, but some were told the results were "positive" or "abnormal," or "negative" or "normal." Those in the positive/abnormal groups reported a higher perception of their risk, greater levels of worry and greater interest in having amniocentesis than the negative/normal groups.
"I think this topic is much broader than simply a study about amniocentesis," says senior author Peter A. Ubel, M.D., director of CBDSM and professor of internal medicine at the U-M Medical School. "This kind of language is common across all clinical settings, and our findings suggest that physicians should be very cautious about the implications of the words they use when presenting information to patients."
The danger, Ubel says, is that "the addition of labels induces bias in people by having them think in terms of broad categories at times when more detailed consideration of the specific test results may be what is required for informed decision-making."
In the study, 1,688 women were asked to imagine being four months pregnant and speaking with their physicians about screenings for fetal chromosomal problems. They completed a hypothetical scenario on the Internet and were randomly assigned into two groups: high risk (defined as a risk level of 12.5 out of 1,000) or low risk (two out of 1,000) of fetal chromosomal problems.
Participants were asked if they were interested in having a blood test, and all who wanted to were informed that the hypothetical test had been performed and indicated a risk of five out of 1,000. In other words, the group initially defined as high risk now had a lower risk, and the initial low-risk group now had an increased risk.
A third of this group was told their risk of fetal chromosomal problems was simply told the risk number: five in 1,000. A second group had their results preceded by an interpretive label - that their results had come back "positive" or "negative," and that this meant they were at increased or decreased risk of fetal chromosomal problems. They did not learn until the next page of the scenario that the risk estimate was five in 1,000. Finally, a third group was given similar results, but with the words "abnormal" or "normal" instead of positive or negative.
Women's reactions were magnified and significantly influenced by the use of labels, the authors report. When told that their baby faced a five in a thousand risk of an abnormality, the risk felt much worse to people who were told the number was the result of an abnormal test. Among that group, 56 percent favored having amniocentesis. That compares with 37 percent who were told it was the result of a normal test, even though the risk levels were exactly the same.
"Although our results are based on a hypothetical scenario, they suggest that obstetricians and genetic counselors should be cautious about providing verbal interpretations of prenatal screening test results to patients," the authors write. It is understandable that medical experts want to help their patients by providing context, the authors note, but experts should be careful not to use language that could significantly affect the patient's decisions.
In addition to Zikmund-Fisher and Ubel, authors of the paper are Angela Fagerlin, Ph.D., of CBDSM, and Kristie Keeton, M.D., of the U-M Department of Obstetrics and Gynecology.
The research was supported in part by the National Institutes of Health, a career development award from the American Cancer Society, and by a Merit Review Entry Program early career award from the Department of Veterans Affairs.
Reference: American Journal of Obstetrics & Gynecology, Vol. 197, Issue 5, p. 528.