DTC Test Results Don’t Lead to Dramatic Changes in Health Care Use

The U.S. Food and Drug Administration (FDA) has frequently expressed concern about direct-to-consumer (DTC) marketing of genetic testing. For example, the FDA required pre-market approval for 23andMe's Personal Genome Service. One of the FDA's stated concerns is that in the case of DTC genetic tests no physician is involved to provide consumers guidance in utilizing these results and there is a danger that consumers will make their own decisions about treatment or use of prescription medicines that can create risks to their health. Recent studies provide some insight regarding consumers' perceptions of these genetic test results.

Impact on risk perception and decision making
Consumers buying direct-to-consumer personal genome tests (DTC-PGT) want to believe good news, according to an article published in the September issue of Nature Biotechnology. While participants' perception of their personal risk changed after receiving results showing decreased and increased risk, there was an "optimism bias"—with a greater change in perception resulting from good news. Additionally, the researchers found that following receipt of PGT results, consumers primarily seek medical attention in response to large, surprising results.

Despite high-profile regulatory scrutiny of DTC-PGT [see box], there is not much evidence showing how use of these services impacts consumers' medical decision making.

The researchers modeled risk perception as a combination of baseline beliefs and learning in response to genetic news. Health care utilization was assessed as a function of changes in these risk perceptions. Eight conditions were evaluated (Alzheimer's disease, Parkinson's disease, breast cancer, lung cancer, colon cancer, prostate cancer, type II diabetes, and coronary heart disease).

The study included 617 real-world consumers who individually sought out and purchased PGT services from 23andMe before the U.S. Food and Drug Administration (FDA) banned DTC health reports. The participants were enrolled in the Impact of Personal Genomics (PGen) study, a longitudinal study of real DTC genomics customers. Risk perceptions were assessed through surveys at baseline and again six months post results.

The researchers found that participants had a slight "optimistic bias" in perception of baseline risk. This perception varied by condition, but the average level of optimism across the conditions was 42 percent, reflecting below-average risk perception versus an average 19 percent perception of above-average risk.

On average, following receipt of test results, participants updated their risk perceptions. However, the magnitude of this updating was asymmetric, meaning that results showing "good news"—a decreased risk of a condition—led to a significantly bigger drop in risk perception compared to the relative increase in risk perception following the receipt of "bad news" or results showing an increased risk of disease. This pattern in results existed regardless of the participants' initial reason for purchasing the test.

"The significance and modest magnitude of observed risk updating suggests that neither excessive overreaction nor complete disregard for the test results was prevalent in our sample," writes lead author Joshua Krieger, from the Massachusetts Institute of Technology. "A pattern of general and moderately increased concern across several conditions did not appear to spark decisions to use medical services, whereas a single higher-amplitude risk perception change increased the odds of engaging in follow-up medical action."

The authors caution that the six-month results regarding follow-up medical action may be conservative, given the likelihood some patients will wait until a routine check-up to raise the PGT test results with health care providers.

Impact on prescribed therapies
A separate study, also involving the PGen Study Group, found that pharmacogenomic results indicating an atypical drug response are common among patients requesting DTC-PGT. Receipt of these results is associated with prescription medication changes, although less than one percent of DTC consumers report unsupervised changes to their prescription medications six months following testing.

The concern, of course, is that receipt of DTC-PGT results will prompt consumers to "self-manage their treatments," including changing doses or abandoning treatment altogether, without consulting a physician or pharmacist.

This study relied on data from 961 new DTC-PGT customers (23andMe and Pathway Genomics) who were enrolled in 2012 and surveyed prior to the return of results, as well as six months after receipt of results. "Atypical response" was defined as pharmacogenomic results indicating an increase or decrease in risk of an adverse drug event or likelihood of therapeutic benefit, the authors report.

The researchers found that 91.2 percent of consumers received at least one pharmacogenomic result indicating atypical drug metabolism—a number consistent with previous estimates. In response to the PGT results, fewer than six percent of participants changed a prescription medication within six months of testing and less than one percent reported making changes without consulting a health care provider.

Takeaway: Studies show the impact of DTC-PGT results on changes in risk perception and on health care utilization among real-world consumers.