Add another high-profile departure to the list of people leaving Verily, the Google-spawned health science company: Thomas Insel, a neuroscientist and former head of the National Institutes of Mental Health who was leading Verily’s mental health initiatives. After barely a year and a half, Insel is joining a startup called Mindstrong, aiming to come up with technologies to diagnose and treat mental disorders.
(The semantic web of Google companies: Verily, which used to be called Google Life Sciences, is partially owned by Alphabet, which used to be called Google and now owns Google.)
At first look, Insel’s new gig sounds a lot like his old gig. So what could be going on at Verily—or with Insel—that would lead him to bail on Google-sized money, freedom, ambition, and datasets?
“I’ve been asking myself that question a lot in the last few weeks. It was a tough decision,” Insel says. “Google has phenomenal resources, and Verily is a really exciting place to work. I just developed this kind of entrepreneurial itch. That’s what drove me to leave government, and even within Alphabet and Verily I felt like I had to scratch it at some point.”
Insel’s hopes for research have always been outsized. You might remember the hypothesis—somewhat controversial—that the hormone oxytocin is a key to human attraction and relationship-formation. That was his work. Insel brought to NIMH a commitment to finding neurochemical and other quantifiable markers for psychiatric disorders rather than relying on behavior and self-reported feelings. He even wanted to throw out the Diagnostic and Statistical Manual of Mental Disorders, the bible of psychiatric symptomatology. For the past few years, well before moving to Verily in 2015, Insel was writing about using biomarkers and mobile technologies to create “precision medicine for psychiatry.”
Apparently it didn’t work. “I spent 13 years at NIMH really pushing on the neuroscience and genetics of mental disorders, and when I look back on that I realize that while I think I succeeded at getting lots of really cool papers published by cool scientists at fairly large costs—I think $20 billion—I don’t think we moved the needle in reducing suicide, reducing hospitalizations, improving recovery for the tens of millions of people who have mental illness,” Insel says. “I hold myself accountable for that.”
If biomarkers can’t diagnose mental health issues, maybe a “digital phenotype” can. In 2015, Insel told Technology Review that this was why he was jumping to Alphabet: the idea that a combination of your medical records (electronically stored, duh) and how you use your gadgets—tracking of activity correlating with depression or future self-harm, let’s say—could be a Big Data bonanza for predicting and treating health issues. “For a bipolar patient whose mania is manifested in rapid, uninterruptible speech or hypergraphia, their disease could be characterized by the frequency, length, and content of participation in social media,” write the researchers who defined the term in Nature Biotechnology. In fact, mood is one of the things that Verily’s $100 million Baseline study will track among its 10,000 eventual participants.
At Mindstrong, one of the first tests of the concept will be a study of how 600 people use their mobile phones, attempting to correlate keyboard use patterns with outcomes like depression, psychosis, or mania. “The complication is developing the behavioral features that are actionable and informative,” Insel says. “Looking at speed, looking at latency or keystrokes, looking at error—all of those kinds of things could prove to be interesting.” Maybe in five years digital phenotypes will have gone the way of neurobiology and genetics in mental health, Insel acknowledges, but for now the tech ideas are worth a look.
You can understand how investors might be drawn to a startup working on the problem, especially one that boasts not only the former head of the NIMH but also Richard Klausner, the former head of the National Cancer Institute. “At some level, people with psychiatric things going on in the background are much more likely to have other health care problems in the background, too,” says Jim Tananbaum, whose Foresite Capital is a major investor in Mindstrong. “I’m sorry to be oblique, but if you were to look at avoidable hospitalizations, depression tracks with those.” In other words: Mindstrong offers the potential for money-saving predictive mechanisms.
And you can understand why the health care world would be open to that kind of technology, too. “Those kind of applications are the ones I like most, because they’re one-to-many,” says Joseph Kvedar, who runs the Connected Health program at Partners Health Care in Boston. “They enable us to look at a population, screen in the background for challenges, and then intervene when necessary with a qualified professional to help. What we do now is wait for you to decide you’re depressed and come to us.”
But it would have been a worthy problem at Verily, too. The company has seen a series of departures, some back to the Google mothership and some out of the Googleplex altogether. An article in Stat blamed Verily’s CEO, Andrew Conrad, saying he’s mercurial in setting priorities. The Baseline study has launched, but other Verily projects, like a Tricorder-like medical scanner and a glucose-sensing contact lens, haven’t worked out yet. Neither has the company announced any specific mental health-oriented initiatives. “I have a tremendous fondness for the company and for the CEO, and feel strongly that he’s gotten a pretty raw deal in the press coverage,” Insel says. “It was really difficult for me to leave because of my attachment to Andy Conrad, who I think has been a really visionary leader there … If anything, Andy’s frustration was that we couldn’t move faster and go farther, especially on the mental health team.”
As to whether Mindstrong will move fast or far enough for Insel, he’ll have the first hint in six months or so—when the data from that 600-person study starts coming in.