Post by Former NIMH Director Thomas Insel: BITs and BYTEs
One of the upsides of so many snowstorms this winter has been the chance to catch up on reading. I’ve been enjoying Erik Brynjolfsson and Andrew McAfee’s book the Second Machine Age about the social revolution instigated by the digital revolution.1 One story they recount is about the Dutch chess grand master Jan Hein Donner. When the authors ask Donner how he would prepare for a match with a supercomputer, he replies, “I’d bring a hammer.”
Computers are getting faster at an awesome or alarming rate, depending on your point of view. Brynjolfsson and McAfee, who are both awed and alarmed, describe the exponential rate of progress over the past decade. Quoting Hemingway about how a man goes broke (“gradually and then all of a sudden”), they see a recent sudden burst in progress in technology. Whether in cars that drive themselves, Jeopardy-winning computers, or smartphones with SIRI, technological challenges facing profound obstacles only a decade ago are now milestones in the rear view mirror of progress.
All of this brings to mind how technology will change the diagnosis and treatment of mental disorders. On the diagnostic side, we already have sensors built in to many of our devices—smart phones, computers, appliances—that are constantly capturing data on location, movement, and communication and can help to create real-time pictures of functional status. Activity monitors that can monitor sleep and movement have been around for three decades. Phone sensors can “diagnose” depression from changes in voice quality. Some have suggested that credit card companies may be the first to detect the onset of a manic episode, although that information seems largely neglected. Qualcomm has recently announced a $10 million XPRIZE for a “tricorder”—a 5-pound device that can diagnose 15 diseases non-invasively. Imagine how technology could provide the sensors to detect signs of mental disorders. This may not win the XPRIZE, but the potential is real.
Of course, the bigger challenge will be to consider how technology will alter treatment. eHealth technologies like electronic medical records, electronic decision support tools, and the use of telemedicine are applicable to all of medicine and pose relatively little disruption to mental health care. But so-called “behavioral intervention technologies” or BITs are a potential game changer, especially for the delivery of psychotherapy. A recent summary of BIT opportunities and challenges is worth a quick summary.2 David Mohr and colleagues describe four different BIT opportunities:
First, videoconferencing and standard telephone technologies have been well validated as extenders of individual psychotherapy. Indeed, most studies demonstrate that telephone-based treatment is equivalent to face-to-face interaction, with the advantage of broader access. The major limitation is that these tools rely on availability of a therapist to provide the content.
A more radical advance is the use of web-based interventions that do not require a professional to drive the interaction. Some of these sites have more of a conventional self-help feel, but others have begun to personalize and tailor the information for very specific needs. The more successful sites have used a coach to create “supportive accountability” to improve adherence. The evidence is still evolving, along with the technology, but Mohr and colleagues are positive about the current results, especially for insomnia and some anxiety disorders.
The big growth is in their third modality: mobile technologies that move interventions into the real world, providing a continuous and ubiquitous connection between patient and care provider. With 56 percent of Americans now owning smartphones, mobile technologies may offer the greatest reach and impact.3 Beyond using mobile technology for improving adherence or collecting passive data about activity or sleep, the cutting edge of this field uses patient-initiated dialogue (think Siri as psychotherapist) or personalized messages keyed to environmental cues—such as proximity to an anxiety-provoking stimulus detected by a GPS system. Social prosthetics for autism are being developed using Google Glass linked to a smartphone for detecting face emotion and translating this into words.
Finally Mohr and colleagues note the advent of social media and gaming as potential therapeutic opportunities. Obviously, these technologies come with risks as well as promise. Consider the pro-ana sites that encourage weight loss for adolescents with anorexia or the video games that increase social isolation and absorption in fantasy for adolescents at risk for psychosis. Nevertheless, the potential to develop group support through social media and increased resilience via cognitive training games could be one of those areas of technology where we will see exponential growth in the next few years. As with many other areas in the BIT world, we don’t have the data yet to know what is going to be most effective for many of our most challenging issues, but the studies that are emerging demonstrate remarkable effects for cognitive training.3
I was trained to believe that transference was essential for insight and that therapeutic alliance was a prerequisite for behavior change. With a movie currently in theaters imagining a man in love with a computer operating system, perhaps the idea that a device, essentially a robot, could replace a therapist seems less outlandish. Like the chess master, my first instinct is still to look for a hammer. But something that Brynjolfsson and McAfee mentioned in passing really got me thinking. They note that in the world of artificial intelligence, the hard problems are easy and the “easy” problems turn out to be hard. Training a computer to do your taxes or analyze stock quotes turns out to be surprisingly easy, but getting a robot to pick up a button or go up steps has been remarkably difficult. We may be just at the beginning of thinking about how technology can help people change their thinking and their behavior. In time, this may be one of the “easy” problems for technology. But for now, it will take our best minds to figure out how to harness this second machine age to make sure we empower people with technology rather than creating more problems for human therapists to solve.
2 Mohr DC et al. Behavioral intervention technologies: evidence review and recommendations for future research in mental health. Gen Hosp Psychiatry. 2013 Jul-Aug;35(4):332-8. doi: 10.1016/j.genhosppsych.2013.03.008. Epub 2013 May 8.
3 Smith A. Smartphone Ownership—2013 Update. Pew Research Internet Project. http://www.pewinternet.org/2013/06/05/smartphone-ownership-2013/ .
4 Anguera JA et al. Video game training enhances cognitive control in older adults. Nature. 2013 Sep 5;501(7465):97-101. doi: 10.1038/nature12486.