Technology shows promise for psychiatric clinical assessment
NEW YORK — In a session presented here on smartphone apps, sensors and web technology, experts discussed the state of research and the clinical implications of these technologies.
Laura Germine, PhD, technical director of McLean Institute for Technology in Psychiatry, discussed the ways in which technology can enhance the understanding of cognitive function.
“Measurement is hard, especially in mental events. Understanding someone’s brain or understanding their mental function is a difficult thing, and the reason is that we have the observations we make which are based on clinical interview and cognitive assessments, and those are sort of a shadow of reality,” she said. “... I would argue that amongst the fields of medicine and research, we have one of the hardest jobs.”
While traditional clinical neuropsychiatry and cognitive neuroscience afford researchers with the tools to diagnose impairment and measure behavior, smartphone and wearable technology will provide an opportunity to measure these things in more people.
Germine presented three facts about digital behavior:
- In the hierarchy of needs, participating in research tends to be a fairly low priority.
- People like to share information they find interesting with their social networks on social media.
- People love to learn something — pretty much anything — about themselves.
Using these facts, along with other principles like minimizing barriers to participation, sharing data/results with patients and respecting patients’ time, Germine and colleagues developed a digital research laboratory — TestMyBrain.org. The lab offers various kinds of research studies that are freely available to the public. After completion, participants receive their results. According to Germine, this type of research participation is readily available and readily sharable — people will complete these tests, then share their results with their social networks, which brings more participants to the study. Through the lab, she and colleagues have tested over 2 million people from around the world.
“In 10 years we’ll be in a great place, but right now it’s something with a lot of promise,” Germine said of digital assessment. “We should start experimenting with it as a field and thinking about how to implement it, but the best practices and the measures are still being formed, hopefully by communities like this one, towards a future where there is more access.”
Though data demonstrate there is interest in mental health apps, researchers have largely been unable to reach and impact people who could benefit. According to John Torous, MD, codirector of the digital psychiatry program at Beth Israel Deaconess Medical Center, this may be due to issues of privacy.
“In the mental health community, and as clinicians, [privacy is] our core nature: We protect people’s information. ... If we’re worried about where data goes, we don’t deserve to have it,” he said.
Torous and colleagues conducted a study of patients with schizophrenia using the SARDAA Health Storylines app. When they analyzed the number of people who downloaded and engaged with the app, results were disappointing: only 225 people downloaded the app over the course of 1 year, and about half used the medication feature once. Citing other examples of low engagement in mental health apps, Torous explored possible solutions, including doubling down on technology to make apps more engaging, doubling down on the people who use them, and a combination of both through digital therapeutic relationships.
Along with colleagues at Beth Israel Deaconess, Torous has developed the LAMP app, which collects information on patients’ physical activity, environmental stressors and real-time surveys and adds it to the treatment plan. The app allows patients to track what they wish and the symptoms that are important to them. It is currently available in the iTunes and Android app stores but requires a code for download.
In the final presentation, Ipsit Vahia, MD, medical director of geriatric outpatient services and medical director of the Institute for Technology in Psychiatry at McLean Hospital, and assistant professor of psychiatry at Harvard Medical School, discussed a relatively new concept called digital phenotyping.
“When we do psychiatric care, we are essentially trying to capture the phenotype — defined in the dictionary as the outward appearance of an organism. We call it the interview, we call it clinical assessment, we call it symptom monitoring, but it’s really capturing the phenotype. That means that in the end, even the most skilled among us is only really as good as the data we have to work with.”
Vahia described the case of an 81-year-old man referred to specialty consult for brain fog. Using step tracking data from the patient’s wearable device, Vahia determined the patient had neuro-vegetative depression. The lesson learned was that technology can add a dimension of information that is otherwise not accessible in the course of normal care. The new term for what he did is digital phenotyping — moment-by-moment quantification of the individual-level human phenotype in situ using data from personal digital devices.
Working with researchers at MIT, Vahia developed a wall-mounted sensor that tracks movement to better understand the behavior of a patient with Alzheimer’s disease. The device is a work in progress, but it has already validated five of the 12 domains of the neuropsychiatric inventory, the gold standard measurement tool for dementia. Though psychotic symptoms cannot yet be mapped with the tool, Torous and Germine may have tools that can help.
“True phenotyping is going to be multidisciplinary, using a wide array of tools,” Vahia said. “I think we were able to develop this work because we came at this with a clinical problem in mind, not a technology in mind.”
MIT Review featured the technology in an article earlier this year, and Vahia said he is excited that technologists are beginning to look at this clinically, as well.
“In the end, what matters is what you’re trying to achieve with the tool, not the validation, not the data, not the analytics. The tool is simply a means to an end; the patients are what matters,” he said. – by Stacey L. Adams
Vahia I. Beyond the interview: Applying smartphone apps, sensors, and web technology to the process of clinical assessment in psychiatry. Presented at: American Psychiatric Association Annual Meeting; May 5-9, 20178; New York.
Disclosures: Germine reports holding a consultant/advisory board role at 23andMe. Torous and Vahia report no relevant financial disclosures.