Skip to Main Content

One of the shinier entrants to have emerged in the world of mental health startups abruptly announced last week it would wind down, right in the middle of an ongoing crisis in mental health care. Mindstrong, which had raised a total of $160 million from a who’s-who of blue-chip investors, and was led for a while by a former National Institute of Mental Health director, simply couldn’t find a way to make money delivering the low-cost, high-quality care it had promised.

Mindstrong had started out as a high-tech biomarker company trying to apply artificial intelligence and passive sensors to track mental health symptoms, but eventually shifted to providing app-based mental health care. And at that point the company discovered a fundamental truth of the U.S. health system: Americans value mental health extremely highly until they have to pay for it.


Psychiatrists and other mental health clinicians often use an approach called motivational interviewing to encourage people to change their behaviors. As part of this technique, we highlight the difference between what individuals say they want and what their behaviors reflect about their priorities. By making this gap clear, we try to help people align their goals with their behaviors.

If I could do some motivational interviewing with leaders in the federal government and across the health care ecosystem, I might ask why, when they are finally willing to talk about mental health, do their behaviors suggest that they don’t care as much about it as they say they do? What led Mindstrong, which had every structural advantage, to conclude it couldn’t make money providing a service people are clamoring for?

To begin with, people have relied for too long on magical thinking about the role of technology in mental health. Apps, artificial intelligence, chatbots, and telehealth should absolutely help provide better, more efficient health care, even in psychiatry. Part of my work at Massachusetts General Hospital involves ensuring that researchers have access to the large-scale clinical datasets needed to build the artificial intelligence so frequently touted in the media. But there’s a reason psychiatrists, psychologists, and other clinicians are needed to provide care, just as planes that can technically fly themselves still have pilots in the cockpit.


The talk may be about technology because we don’t want to talk about a less comfortable topic: money. A byzantine coding system developed to favor procedures and specialty care enables payers to starve mental health services of the payment they need to survive, just as they do primary care. When I led clinical services in a mood disorder treatment program, what insurance companies reimbursed us for treatment did not fully cover the cost of a receptionist, much less enable us to pay for the care management services many of our patients needed. When I asked my clinical leadership about obtaining additional resources, I was encouraged to write a grant.

An underappreciated consequence of the focus on profit in the health care industry, even by health systems that are supposed to be nonprofit, is continued underinvestment in mental health services. No one questions the need for these services, but when they lose money on every patient they simply cannot make it up in volume. Little wonder that even health systems nationally ranked for the quality of their mental health care try to outsource their own employees’ care to a tech-enabled virtual mental health provider, touting coaching and mindfulness rather than psychiatric care to control costs.

Medical leaders, in and out of government, must confront the reality of reimbursement, not just the magic of technology. I hope the next National Institute of Mental Health director will be a clinician-scientist, not a bench scientist — and that she will be willing to be a vocal advocate for payment reform. Likewise, while the U.S. Surgeon General deserves tremendous credit for a thoughtful and comprehensive report on child mental health care, translating that into practice against stiff lobbying headwinds is difficult, and doesn’t lend itself to headlines. As long as insurers are allowed to persist with woefully inadequate compensation for mental health care services, Americans will continue to get woefully inadequate mental health care services.

The irony is that the cost-effectiveness of treating mental health issues is unequivocal: treating anxiety and depression lowers the cost of just about every chronic illness imaginable. But for payers, who shrewdly calculate that those savings may accrue over years, the game is to avoid paying right now and ideally to get someone else to pay down the road. (If this sounds eerily familiar, recall that it took multiple rounds of federal legislation to get insurers to cover smoking cessation, with a similar positive cost-effectiveness profile as treating mental health disorders.)

There will be plenty of new companies to replace Mindstrong, and I hope some of them will get traction. Mental health care needs all the new ideas it can get. But as someone trying both to develop these technologies and to care for people with these illnesses, I can tell you: Technology can improve the care of people with mental illness, it just can’t replace actual care. And when it comes to care, you get what you pay for.

Roy Perlis is a psychiatrist and associate chief for research in the Department of Psychiatry at Massachusetts General Hospital in Boston and a professor of psychiatry at Harvard Medical School.

Create a display name to comment

This name will appear with your comment

There was an error saving your display name. Please check and try again.