By Deanna Alexis Carere

from GeneWatch 28-3 | Oct-Dec 2015

The state of the direct-to-consumer genetic testing industry has changed dramatically since it was last highlighted in GeneWatch in Fall 2010. At that time, direct-to-consumer (DTC) genetic testing had recently been named as the "Invention of the Year" by Time magazine, and DTC genetic testing companies were expanding both in number and in their offerings. At the same time, criticisms of the industry were growing, and the Food and Drug Administration and Government Accountability Office were each formally investigating these services. Little regulatory action had been taken, however, with the exception of the FDA blocking over-the-counter sales of DTC genetic tests from one company, Pathway Genomics.

It was in this context that the Impact of Personal Genomics (PGen) Study was funded by the National Human Genome Research Institute (NHGRI) to provide the first "before and after" survey of DTC genetic testing customers. The PGen Study, a longitudinal survey of customers from two DTC genetic testing companies, was designed as a collaboration between academic researchers (at Brigham and Women's Hospital / Harvard Medical School and the University of Michigan School of Public Health) and research scientists in the private sector (at 23andMe and Pathway Genomics). Other studies had previously investigated DTC-type genetic testing, but often lacked baseline (pre-testing) data and access to individual genetic results, were conducted in clinical settings with pre- and/or post-testing genetic counseling, or used test packages that did not reflect current commercial offerings. The PGen Study was designed to address these shortcomings. Participants were recruited from among actual customers of 23andMe and Pathway Genomics, surveys were administered prior to customers' receipt of results and in two post-testing follow-ups, and company collaboration enabled direct linking of complete genetic results with survey data.

Although it was not anticipated at its initiation, the PGen Study - conducted between July 2012 and early 2013 - would provide a glimpse into a DTC genetic testing experience that no longer exists. In general, many DTC genetic testing companies have since closed up shop altogether. Pathway Genomics has shifted to a physician-mediated genetic testing model, and 23andMe launched a new health-related service in October 2015 (two years after being blocked by FDA from returning health-related results directly to consumers) that includes only carrier testing for recessive conditions. Nonetheless, insights from the PGen Study remain relevant to the future of this industry and the kind of testing that likely lies ahead. 23andMe has publicly stated its intentions to seek FDA approval for an expanded test menu, within a context of "direct-to-consumer healthcare," and FDA regulation notwithstanding, DTC-GT offerings are expected only to grow.[1]

What, then, has the PGen Study revealed so far, and how might its findings be instructive to the future of DTC genetic testing (and direct-to-consumer healthcare more generally)? The first substantive findings from the PGen Study pertained to consumer understanding of genetics, comprehension of their results,[2] and confidence in making use of them.[3] Here, the vast majority of DTC genetic testing customers (a self-selected, generally well-educated but still diverse group) showed good understanding of basic concepts in medical genetics, and comprehension was high across all types of results (i.e., pharmacogenomics, carrier screening, and disease risk). Nonetheless, lower educational attainment, lower genetic literacy/numeracy, and older age were associated with lower comprehension of results, suggesting that comprehension may be improved by the tailoring of results reports to individual consumer characteristics or preferences. Consumer confidence was a different story: while understanding of genetics concepts was high both prior to and following testing, perceived self-efficacy with these concepts (i.e., confidence in one's ability to apply one's genetics knowledge) showed significant decreases 6 months after results were returned. Importantly, this decrease in self-efficacy post-testing was associated with lower probability of health-care provider consultation, lower perceived value of DTC genetic testing, and greater regret regarding the decision to pursue testing.

Lowered genetics self-efficacy following DTC genetic testing may reflect an appropriate reevaluation by consumers of their facility with genetic information in response to receiving complex reports, which span dozens of conditions with variable environmental contributions and modes of inheritance. Nonetheless, this finding suggests that DTC genetic testing companies have work to do in improving the consumer experience - and motivation to do so, given the observed relationship between self-efficacy, perceived genetic testing value, and decision regret. On this front, we have reason to be optimistic about the "new customer experience" recently launched by 23andMe: beyond revising the scope of testing available, the company created and evaluated new carrier testing reports in large, population-based samples across the United States to demonstrate user comprehension.[4] These new reports (viewable at feature less text and more white space, highlight the most important pieces of information including intended uses and limitations, and employ a logical step-by-step method of revealing increasingly complex details. What remains to be seen is how these changes will be translated to future reports for complex disease risks and pharmacogenomics results, where consumer comprehension of what actions ought or ought not to be taken in response to results (e.g., cancer screening, medication changes) is of greater clinical significance than in carrier testing.

The absence of disease risk and pharmacogenomic results from 23andMe's new service marks another important, post-regulation change: the company currently does not provide any genetic information that might be relevant to the immediate tester's health. This means not only that disease risk is not provided, but also that consumers will not be told through carrier testing if they carry two recessive mutations for a condition (and are therefore, at least at the genetic level, "affected"). A notice added to carrier testing reports reads: "This test does not diagnose any health conditions. Please talk to a healthcare professional if this condition runs in your family, [or] you think you might have this condition." Exclusion of affected individuals and those with a relevant family history is significant, and addresses a frequent criticism of DTC genetic testing - namely, that consumers who should be getting a comprehensive clinical genetics evaluation (including a physical examination, assessment of family history and sequencing of relevant genes) may assume that DTC genetic testing services adequately serve their needs. For example, an adjusted lifetime risk estimate for breast cancer based on genotyping of multiple single nucleotide polymorphisms (SNPs) may (setting aside legitimate criticisms of how such estimates are obtained) be relevant to a woman with no personal or family history of breast cancer or related risk factors, but as soon as she has a moderate/high personal- or family-history-based risk of breast cancer, these SNP-based estimates become clinically inappropriate and potentially confusing.

That consumers seek DTC genetic testing not only for its predictive abilities but also for its explanatory power is evidenced by findings from the PGen Study: across all conditions tested, both family medical history and personal medical diagnosis were strongly associated with interest in condition-specific results.[5] Furthermore, an unpublished review of qualitative data on test satisfaction indicates that some consumers were disappointed or confused when their DTC-GT results did not confirm a medical diagnosis (e.g., Ehlers-Danlos), or indicated they were at low genetic risk of a condition present in a close relative (e.g., breast cancer in a sister). Finally, we know from a PGen Study analysis of perceived cancer risk before and after testing that DTC genetic information has a measurable and predictable effect on consumers' perceptions of disease risk.[6] How to responsibly address the fact that consumers with cancer, heart disease, or neurological conditions (or significant family histories of these conditions) will continue to order DTC genetic testing once disease risk estimates are reinstated is something that 23andMe and other companies will need to sort out in order to satisfy FDA.

Preliminary PGen Study data can provide insight into the other component of 23andMe's services halted by the FDA: pharmacogenomic testing. Pharmacogenetics represents one of the greatest opportunities in personalized medicine, with the promise of preventing adverse drug events and targeting medications to those who will benefit most. The potential clinical utility of pharmacogenomics, however, means it may also be one of the least 'benign' types of DTC genomic information, and the FDA detailed specific concerns in its initial Warning Letter about consumers changing their medications without clinician consultation. PGen Study data presented at the 2014 American College of Medical Genetics Annual Meeting revealed that, among 1,003 PGen Study participants, 54 (5%) changed a prescription medication and 70 (7%) changed a non-prescription medication on the basis of their DTC genetic testing results; of these, 8 (0.8% overall) and 28 (3%), respectively, did so without consulting their health care provider. The frequency of unsupervised and inappropriate medication changes thus appears to be quite low, and some of these changes may have been made regardless of the results of genetic testing. (Of course, people make unsupervised changes to their medications all the time based on perceived side effects, unaffordability, or information gleaned from the internet). On the other hand, evidence of even a low rate of inappropriate medication changes may represent sufficient harm for FDA to continue to block such testing, particularly when frequencies of 0.8% and 3% are applied to the entire DTC genetic testing population, which currently numbers over 1 million through 23andMe alone. 

There is still much to be learned about the impact of direct-to-consumer genetic testing - in particular what burden (if any) it places on health care systems and its effects on clinical outcomes - and large-scale empirical studies will be essential to answering these questions. It is fortunate, then, that a likely downstream effect of FDA's regulatory interest in the industry is a renewed focus on rigorous empirical research, both by DTC genetic testing companies in the process of bringing products to market, and by government and academic researchers in post-market evaluations.


Deanna Alexis Carere, ScD, CGC, is a genetic counselor and postdoctoral fellow in Epidemiology at McMaster University in Hamilton, Ontario. As a doctoral student, she studied the evidence for and application of genetic testing using common variants within the Genomes2People research group at Brigham and Women's Hospital in Boston, Massachusetts. She currently researches the molecular connections between cardiovascular genetics and dementia.


[1] Steinmetz, K. "23andMe's CEO says the company will win FDA approval for more tests." Fortune. Accessed online, December 1 2015:

[2] Ostergren JE, et al. "How well do customers of direct-to-consumer personal genomic testing services comprehend genetic test results? Findings from the Impact of Personal Genomics Study." Public Health Genomics. 2015;18(4):216-24.

[3] Carere DA, et al. "Consumers report lower confidence in their genetics knowledge following direct-to-consumer personal genomic testing." Genet Med. 2015; Epub ahead of print.

[4] FDA. "FDA news release: FDA permits marking of first direct-to-consumer genetic carrier test for Bloom syndrome." Accessed online, December 1 2015:

[5] Meisel, SF et al. "Explaining, not just predicting, drives interest in personal genomics. Genome Med. 2015;7:74.

[6] Carere, DA et al. "The impact of direct-to-consumer personal genomic testing on perceived risk of breast, prostate, colorectal, and lung cancer: findings from the PGen Study." BMC Med Genomics. 2015;8:63.

Search: GeneWatch
The purpose of the Genetic Bill of Rights is to introduce a global dialogue on the fundamental values that have been put at risk by new applications of genetics.
View Project
Genetic Testing, Privacy and Discrimination
View Project