Published online 12 January 2011 | Nature | doi:10.1038/news.2011.12


People prove impervious to anxiety from genetic tests

The results of direct-to-consumer tests relating to the risk of developing diseases seem easy to swallow.

swabbed mouthMany were concerned that high disease risk determined by direct-to-consumer genetic tests would trigger anxiety.Peter Dazeley/Getty Images

A vial of saliva harbours a wealth of genetic information, and companies are mining this treasure trove to provide the public with personal disease-risk profiles. Some experts have questioned whether people might misinterpret such complex information and become anxious, but a study published today in the New England Journal of Medicine1 seems to debunk that idea. Test results also failed to prompt changes in diet and exercise.

Direct-to-consumer genetic tests, such as those offered by personal-genomics companies 23andMe in Mountain View, California, and Navigenics in Foster City, California, scan an individual's DNA for a whole host of disease-related genetic variants. "The concern was if people hear that they're at risk of developing a scary disease, they would be terrified," says Robert Green, co-director of the Alzheimer's Disease Clinical and Research Program at Boston University in Massachusetts, who was not involved in the study. Spurred by anxiety, people could then ask for unnecessary screening — pushing up health-care costs.

A previous study found that genetic testing for Alzheimer's diesease had no impact on anxiety2. But a team led by Eric Topol, a professor of translational genomics at the Scripps Translational Science Institute in La Jolla, California, has now looked at the effect on psychology and behaviour of commercially available genome-wide scans focused on a range of diseases. "This is an area that has been in the dark matter of our knowledge base," says Topol.

Topol and his colleagues recruited roughly 3,600 participants from Scripps Health, a non-profit health network in San Diego, California, and other large employers in the area including Microsoft and Sempra Energy. Participants completed a lengthy online health questionnaire before submitting a vial of saliva to Navigenics for testing. Participants received a big discount on the cost, normally around US$2,000, for the test, which analyses about 500,000 common genetic variants.

Orange for danger

The Navigenics test results included risk information for 23 medical conditions including Crohn's disease, Alzheimer's disease, diabetes and psoriasis. For each of the conditions, the test results showed a comparison between the general population's average lifetime risk and the participant's estimated lifetime risk, expressed as percentages, in a box. For example, the average risk that a woman will develop breast cancer is 13%, but a participant's estimated risk may be 9%. If a subject's overall risk for a condition was greater than 25%, or if their risk was 20% above average, the box was coloured orange.

About six months after the participants received their test results, Topol and his colleagues asked them to fill out a second questionnaire. The researchers then compared participants' baseline data with the answers on the follow-up questionnaire to determine whether people had become anxious or distressed, or changed their diet and exercise habits. The team also asked participants whether they had been screened for any diseases since receiving the results, or if they planned to be screened.

Of the 2,037 people who completed the study, 90% showed no sign of distress related to their test results. Nor did the results seem to increase anxiety. The test results also indicated that there had been no effect on the subjects' fat intake or exercise habits. But participants who had a higher risk for certain conditions were more likely to report that they planned to be screened for those conditions in the future.


"That was the one change that was noteworthy," says Topol. Regarding the more than 40% of the participants who dropped out, the authors note that it is possible those subjects may have experienced more anxiety than the group completing the study.

The researchers plan at least one more follow-up with the study subjects.

"This is a great first step," says Green. He points out, however, that Topol and his colleagues did not investigate whether participants understood the data they received. Green and his colleagues have a grant from the US National Human Genome Research Institute to conduct an in-depth before-and-after survey of people who buy direct-to-consumer genetic tests. "We're hoping to build on these data," he says, "and really understand what people get out of this information." 

Commenting is now closed.