This report was produced in partnership with the McGraw Center for Business Journalism at the CUNY Graduate School of Journalism.

For the first time in John R. Brown’s two-decade quest for the right medication to make his brain behave, his psychiatrist offered him hope in the form of what seemed like rock-solid science.

By collecting the former editor’s DNA with a quick cheek swab, a new genetic test retailing at $3,800 would reveal which depression medications Brown would likely respond well to and those he would not.

The GeneSight test result helped prompt Brown’s psychiatrist to switch him to a new drug in June, making Brown optimistic that there might be a path back to a job, an apartment and even a new girlfriend. But three weeks later Brown, 40, was suicidal, checking into a mental hospital at his therapist’s urging.

“I can’t help but wonder if I should have stuck with the devil I know,” Brown now says.

Genetic tests to identify the most effective psychiatry drugs are the hot new technology in the race to create personalized treatments based on people’s DNA. More than 600,000 of these tests likely have been administered in the last three years, based on company websites and research data, to better treat conditions ranging from depression to attention deficit disorder to anxiety. In a nod to the tests’ growing acceptance, the federal Medicare program agreed last year for the first time to pay for the GeneSight test for some depressed patients.

But a review by the New England Center for Investigative Reporting has found that virtually all the evidence that these psychiatric tests work is based on limited studies funded by the companies themselves or researchers they fund, including all five studies used to promote GeneSight on the company’s website.


Reporter Beth Daley discusses genetic mental health tests on WGBH News’s Morning Edition.


The federal government doesn’t require companies to prove thousands of tests are accurate before marketing and selling them. Unlike drugs, the Food and Drug Administration does not regulate them, although the agency has announced plans to.

Some of the testing firms’ research practices also raise questions about potential conflicts of interest. For example, California-based Proove Biosciences provides an employee to administer and run a study on the company’s test at a Littleton, Mass., mental health clinic. It pays clinicians for time spent related to the study, a relationship that could give clinicians an incentive to steer patients to the test — and give the company access to insurance reimbursement for those patients.

“I can’t help but wonder if I should have stuck with the devil I know.”

Patients also cannot find out for themselves if their doctor stands to gain financially by recommending a personalized medicine test. A public federal database created in 2014 to disclose financial relationships between the healthcare industry and physicians excludes most genetic tests.

“It is the Wild West,” said Mildred Cho, associate director at the Stanford Center for Biomedical Ethics, who says there is little published evidence on the effectiveness of drug response tests.

Eric Lander, a leader of the Human Genome Project and founding director of the Broad Institute in Cambridge, goes further, saying in a New England Journal of Medicine article earlier this year that the mental health testing company claims are “not harmless and may be quite dangerous.”

Graphic: Cory Dawson for NECIR
Graphic: Cory Dawson for NECIR


In psychiatry, a field hampered by a lack of definitive tests for conditions and treatments alike, the idea of a genetic test to identify the drugs most likely to be effective is particularly appealing as a way to cut down on the trial-and-error approach to medications that many patients face.

And the tests do hold promise. The basic science behind them is built on years of solid research — analyzing an individual’s genetic make-up to determine how they may respond to drugs. GeneSight’s test was developed using technology from the respected Mayo Clinic and Cincinnati Children’s Hospital Medical Center. Officials at Assurex Health, the Ohio-based company that makes GeneSight, stand behind their product. They make no guarantee that GeneSight will lead doctors to pick the right medicine for patients, but point to company studies that show it more than doubles the odds they will. Plus, the officials note, the test underwent close review by a Medicare contractor before the agency agreed to pay for the tests.

“Patients do better,” said Bryan M. Dechairo, senior vice president of medical affairs and clinical development for Assurex. “We continue to do more studies,” to show the test’s value.

Psychiatric conditions affect 1 in 4 adults every year and 1 in 5 teens


The amount of money at stake is enormous. Psychiatric conditions affect one in four adults every year and one in five teens, according to the National Alliance on Mental Illness. Finding a test to more precisely target treatment could help troubled people find help more quickly, cut down on side effects and save money. It can also make hundreds of millions for the test makers, which helps explain why at least 15 companies are now ramping up mental health testing.

“It is so much more than just getting the medications right,” said Rachel Scott, vice president of clinical research and operations for Pennsylvania-based test maker Genomind. “It is giving patients the opportunity to step outside the stigma of mental illness, for them to see this is their biology — and give their clinician some objective information.”

But psychiatric genetics remains in its infancy. Aggressive marketing and a lack of regulation of tests that have a thin scientific foundation have raised deep concerns for some bioethicists.

“The claims of these companies are weak,” said Robert Klitzman, a psychiatrist and bioethicist at Columbia University who had a commentary published last week in a mental health journal calling for vigilance in the marketing of genetic mental health tests.

“Conflicts of interest in the field are major problems,” he said.


The Littleton, Mass., building that houses the New England Center for Mental Health. (Lauren Owens for NECIR)
The Littleton, Mass., building that houses the New England Center for Mental Health. (Lauren Owens for NECIR)


Potential conflicts of interest

Last fall, the owners of the New England Center for Mental Health in Littleton, an adult and child clinic, were asked to be part of a research study.

California-based Proove Biosciences wanted the center to join a 31-location national study, aimed in part at better understanding how its genetic tests benefited adults diagnosed with mental health conditions.

“There is a big set of ethical concerns here.”

The company would provide the center with a research assistant stationed in an office behind the reception desk. If a clinician and patient decided a Proove drug metabolism test, which analyzes 69 genetic variants, would be useful, the assistant would take the patient’s DNA with a cheek swab, collect insurance information for Proove to be reimbursed, and administer extensive questionnaires to track how the test worked out over time.

“We wanted to offer our patients a test that could really help them,” said Jessica X. Hennessey, director of operations and management at the Center for Mental Health.

The center received a one-time payment of $600 from Proove for training and is allowed to bill $150 for each hour of “clinical research service” and “services rendered” by Littleton staff members related to the study, according to Ernie Corrigan, a spokesman for the center. The Center for Mental Health has billed Proove a little more than $5,600 since then, Corrigan said.

Proove officials and Hennessey said clinicians are never compensated based on the type or number of tests they order — and clinicians do not promote the swab test in any way.

“The tests are only ordered if they are medically necessary,” said Brian Meshkin, Proove’s founder and CEO.

But the close arrangement raises eyebrows among bioethicists because it can lead to the appearance of clinicians ordering the test to receive compensation for the study.

“There is a big set of ethical concerns here,” said Columbia’s Klitzman, after being told about the Littleton study. He said those concerns are especially pronounced if doctors have financial incentives that could have them “order tests that have no clear proven benefit.”

“We wanted to offer our patients a test that could really help them.”

Federal officials are growing increasingly concerned about such research-physician arrangements in the genetic testing field. Late last year, a genetic test study that included psychiatry drug response by New Orleans-based Renaissance RX was shut down after Medicare halted payments and began a review. A Rhode Island doctor also sued the company, charging, in part, that Renaissance RX enrolled ineligible people in order to bill insurance.

Meanwhile, the Office of the Inspector General for the U.S. Department Health and Human Services last year singled out compensation agreements related to research studies in a special fraud alert, noting there were growing cases of doctors receiving improper incentives regarding referrals.


John Brown (Sofia Adams for NECIR)
John Brown (Sofia Adams for NECIR)


‘Why not try it?’

The GeneSight test has gotten the most outside scrutiny among the new tests because it faced review by Medicare before the agency agreed to cover it for certain depressed patients.

Medicare contractors were swayed by company-funded studies showing that when patients with depression took the medications recommended by the test, they had fewer symptoms after eight weeks compared with patients who did not use the test.

“Use of the test to aid in (drug selection) has improved patient responses to treatment by 73 percent on average … these findings support the value of the GeneSight test,” according to the coverage decision.

Excitement built in the mental health world that maybe there was finally a scientific test that could really help patients.

Brown heard about the test from his psychiatrist, Susan Gerretson, of the Community Health Centers of the Rutland Region and found out his federal insurance would pay for it.

While his medications of lithium for bipolar disorder and the anti-depressant Effexor were holding him steady, he was hardly in a happy place.

“I figured, Why not try it?” he said.


John Brown has spent 22 years searching for the best treatment to help manage his mental health. He has been diagnosed with bipolar and depression, which cause him to have manic episodes. In this video, Brown talks about a manic episode he had in 2014. (Video: Sofia Adams for NECIR)


Of the five medications the test suggested he would respond best to, Brown had never tried two of them. After several months of discussion, Gerretson put him on one of those drugs, Fetzima.

Even without the new drug, Brown says it was a challenging time: He turned 40, his old girlfriend called, and he got in a fight with his mother.

But Fetzima “was a factor” in his breakdown, he believes. He paused during an interview at the Brattleboro Retreat, a mental health treatment center, this summer and gave a wry smile. “It sure didn’t help.”

Gerretson declined to talk specifically about Brown’s case, citing privacy concerns. But she said in an email that she believes GeneSight can be helpful as long as a doctor fully understands a patient’s history and conducts his or her own research.

“I never believe what sales representatives try to sell me,” she wrote.

Yet critics of these tests say it’s impossible for doctors — or anyone — to truly evaluate many of the tests. GeneSight’s proprietary formula for interpreting genetic information has never been independently assessed. The test examines genes well-known to play a role in antidepressant response, but it adds in others that do not have the same level of evidence, some researchers say.


John Brown of Rutland, Vt., talks about his struggle with depression and bipolar and how the several medication regimes he has been on for the past 22 years have not truly brought him relief. (Video: Sofia Adams for NECIR)


Robert Howland, associate professor of psychiatry at the University of Pittsburgh, wrote a paper last year concluding that the makers of the GeneSight test have overstated the findings of their studies, making outside review more urgent.

“Independent assessment … of the GeneSight test is necessary to overcome the appearance of bias,” said Howland.

But GeneSight’s Dechairo said, if the company didn’t do the research, no one would. The company has 12 ongoing studies not only for depression, he said, but for bipolar disorder, anxiety, schizophrenia and others. “We aren’t stopping … we are collecting more data,” said DeChairo.

“I never believe what sales representatives try to sell me.”

To reach patients, the company aggressively markets its tests to the psychiatry world, in part by paying psychiatrists and nurses to talk favorably about their test to peers. The practice is hardly new in medicine, but, unlike the drug industry, most genetic test makers do not have to disclose which doctors they pay because they’re not FDA regulated.

Catherine Poulos, a psychiatric nurse practitioner at East End Mental Health in Hampton Bays, N.Y., advertises the GeneSight test on her website, saying it allows her practice “to use the right drug at the right dose.” But nowhere does it say on the site she is paid to speak favorably to peers about the test; a patient would have to look at her Linkedin profile to see that. Poulos declined to speak on the record to NECIR.

But some consultants who use the tests say they do so because they believe in it – not because they get paid.

“It adds clarity,” said Chris Bojrab, president of Indiana Health Group and a psychiatrist, a large behavioral health practice and a GeneSight consultant. He discloses his financial relationship with GeneSight on his website.


The main entrance of FDA Building 1 in Silver Spring, Md. (Michael J. Ermarth/FDA)
The main entrance of FDA Building 1 in Silver Spring, Md. (Michael J. Ermarth/FDA)


A regulatory fight

The FDA loophole that allows unregulated tests on the market dates back more than three decades to when the FDA began regulating diagnostic tests. At the time, the agency exempted what was then a small group of tests developed, manufactured and used, in a single lab, such as in a hospital.

Starting in the mid-2000s, for profit companies have used the exemption to develop thousands of complex genetic tests now in the marketplace without FDA review.

The FDA announced plans last year to regulate these tests over the next nine years. However, there is no timeline to finalize the rules, according to an FDA spokeswoman, and the process is likely to take years.

“People are making decisions based on unproven tests.”

Commercial labs and academic institutions are pushing back against regulation, arguing the rules will stifle innovation and delay life-saving tests getting to patients. Regardless of the outcome of the regulatory fight, many academic scientists say this new era in psychiatry testing needs independent study. They question if the tests are adding unnecessary costs to healthcare.

“These companies look at a very small subset of a very large number of factors,” that go into treating patients, said Bruce Cohen, a psychiatry professor at Harvard Medical School and director of the Program for Neuropsychiatric research at McLean Hospital in Belmont. “People are making decisions based on unproven tests.”

Brown’s decision was to go back to old medications. For him, the GeneSight test felt more like “a cool piece of science fiction” than a breakthrough. Still living in his mother’s house, he remains as determined as before to rebuild his life.

“I’m going to have to figure it out another way,” he said.


Editor: Clara Germani Digital producer: Joshua Eaton Contributing interns: Marta Craviotto, Rachel Riley, Meggie Quackenbush, Sofia Adams and Cory Dawson

Have a tip?


Also in this series:

Live chat: What should you know about psychiatric genetic testing?
Are genetic mental health tests causing more harm than good? (+documentation)