If the TV ads are at all effective, plenty of people will be getting the gift of their genetic tests this Christmas. These tests frequently allow people to explore their inherited tendencies toward health problems and, in some cases, may suggest lifestyle changes to ward off future problems—although studies have indicated that few people do.
However, DNA test results can also cause issues that wouldn’t otherwise be there. Genetic information can exert a potent placebo effect—or the opposite, the nocebo effect, wherein if you think that something can harm you, it in fact does. And the potency of this effect has not been studied until now.
Some psychologists at Stanford wondered if the perception of genetic risk could actually increase people’s risk, independent of their genetic risk. In other words, could simply learning that you have a genetic propensity for something elicit physiological changes akin to really having that propensity, regardless of whether you have it? The team designed experiments to find out.
First, they recruited a couple hundred research subjects into a “personal health study” in which the participants were told that they would have genetic profiles done and then learn about which exercises and diets were best suited to them. The researchers obtained a sample of the participants’ DNA to find out their true genetic status for two easily measurable physiological parameters: how much they could run on a treadmill or how full they got after eating a meal. But they didn’t tell the participants their results. Researchers then measured that parameter by feeding the participants or putting them on a treadmill, providing a baseline reading.
A week later, the same participants were invited back. At this point, the researchers lied to half of the people about their genetic risk, while telling the controls the truth. They had both groups run on the treadmill or eat the same meal again. Then they told the participants the truth about the genetic results.
If lying about genetic tests sounds ethically… questionable, the researchers know. They dealt with their reservations in a few ways. First, the researchers focused on genetic risk information for obesity; they figured that learning you are at risk would obviously suck, but it’s not like they were telling people they were at risk for cancer or Alzheimer’s. Second, they only allowed the participants to hold the wrong idea about their genes for an hour, just long enough for the researchers to assay the effect of the misinformation on behavior and physiology.
Finally, the team notes that “the potentially iatrogenic [induced inadvertently by medical care] effects of learning one’s genetic risk are already occurring at scale.” One in twenty-five American adults have already purchased personalized genetic test results; we might as well study how this information is impacting them.
Food and fitness
The first genetic variant sequenced was one associated with reduced exercise capacity. Those told that they had the high risk genotype—regardless of whether they did—got off the treadmill quicker and didn’t breathe as efficiently as they had the previous week, before they got their DNA test results. Those told they had the protective variant ran for longer before reporting feeling hot or that the workout was hard, compared to their baseline workout the week before.
To see if these results were generalizable, the researchers did a similar experiment using a genetic risk factor for obesity. In this experiment, those told they had the protective phenotype (whether or not they actually did) got fuller after eating the same meal they had been fed the previous week, as determined by elevated levels of signaling molecules in their gut. In this case, the effect of the perceived genetic risk was “much greater” than that granted by the actual protective genotype. The effects of the perceived genotype did not differ according to which DNA sequence the participants actually had.
The psychologists who ran this study aren’t sure why they found what they did. While genetic information can change behavior, it can’t account for the rapid physiological changes they saw here. Stress doesn’t explain it either, since the perception of a protective genotype generated a physiological increase in hormones that control our eating. They conclude that the causal mechanism, which takes effect pretty instantaneously, is “mindset.”
The results of this ethically challenging study will only lead to more ethically challenging situations. Society is already grappling with deciding who should get information about genetic risk, including for Parkinson’s and Huntington’s diseases, as well as other conditions that are hereditary but for which we have no known cure or even much treatment. Meanwhile, people are getting information about far more subtle genetic influences anyway.
If the very information can affect health, like this study suggests, should doctors ask patients if they know their genetic risk profiles, much like they would ask patients which pills they’re taking? And if the mere knowledge of risk itself compounds that risk—if such knowledge becomes an additional risk factor that impacts health—could doctors withhold such knowledge from patients? Should they?