Case study: Bryan is a 57 year old male who is employed as a graphic designer. He has Bipolar Affective Disorder and has frequent episodes. “Being bipolar is like jumping out of an airplane knowing you don’t have a parachute on. You know you’re going to be hurt, but the high is so euphoric that it’s worth the risk. You can deal with the consequences later.”
Question: Can technology help this patient stay healthy and out of hospital?
(Because of firewall issues, this link may not work. The article follows.)
This week’s Reading isn’t from a peer-reviewed journal, it’s from Wired.
The piece considers uses of technology to help mentally ill patients, a topic we’ve considered in past Readings.
The article discusses an app called Priori, developed at the University of Michigan, that “runs constantly in the background, using the phone’s microphone to analyze his voice and track when he is, and isn’t, speaking. Mania is typically marked by speech that’s loud and rapid, often with erratic leaps from topic to topic. Longer pauses or breaks can indicate depression.
At the moment, the app only collects data on his behavior. But the hope is that it eventually will use this information to warn Bryan and his doctor to an impending bipolar episode.
The idea is, of course, very seductive. Consider how little time we spend with our patients. Even an appointment that stretches for an hour or two represents a small fraction of that day — a snapshot into someone’s life. And yet the phone that many of our patients carry could be used to gather so much information. Priori analyzes voice patterns, one of the things we look at when considering a possible emerging manic or depressive episode. Cool.
The Wired piece discusses a data set and its small, though the article is enthusiastic — not quite meeting the standard of a peer-reviewed journal.
Still, Priori gives us a possible glimpse into the future, where technology helps us help patients better.
A few thoughts:
1. On Technology
As I’ve noted before: Technology has transformed our lives. This morning, you (nearly instantly) texted photos to a childhood friend who lives 10 time zones away, cashed a cheque into your bank account, and got live updates on Ferguson — and did all of these things with your smartphone without leaving your kitchen or your morning coffee.
2. On Technology and Medicine
Technology is also transforming health care. Think not just of MRIs and clot-busting enzymes manufactured in labs, but think of the way patients can increasingly empower themselves.
A BMC Research Notes paper looked at Twitter and cancer patients in Japan.
Whereas, not that long ago, a cancer support group was heavily restricted by geography and (often) cancer type, the Twitter-verse isn’t bound by such things. Today, patients offer each other support across cities and regions and countries, regardless of cancer types.
Relationships between users. Correlation diagram centered on user0. The users connected by the arrows mutually sent one or more replies. The search was conducted to incorporate friends’ friends. userXX(outlined) : cancer patients. Users who listed their specific type of cancer in their profiles. (B): breast cancer, (L): lung cancer, (O): ovarian cancer, (U): cancer of the uterus. userXX#: Users who are believed to be cancer patients judging from their tweets, although no disease names were described in the profiles (because of descriptions of terms such as anticancer drugs, routine examinations, CT, contrast dyes, bone scintigrams). userXX: Other users.
Source: BMC Research Notes. 2012 5:699.
3. On Technology and Psychiatry
More and more literature supports psychological interventions for psychiatric illness. As you know, there is as much evidence for Cognitive Behavioural Therapy for mild and moderate Depression as there is for medications — but patients are often unable to access CBT. Internet-assisted CBT? There’s an app for that. Or more than an app. (According to a recent paper by the Ivey Business School International Centre for Health Innovation, there are some 97,000 health apps — and more than a handful offer psychological interventions.) My favs? That’s for a future Reading of the Week.
I note that we will be hearing more about e-therapies in the future. It’s not just consumer demand. The just-released Ontario government’s “Expanded Mental Health and Additions Strategy” makes specific commitment to: “Using virtual applications to enable people with mental illness and addictions to access services.”
Of course, this Wired article maps out a more ambitious future than anything you can find in the app store or a government news release, where the smart-phone is a major tool, not just to inform and educate and introduce CBT concepts, but literally to keep track of symptoms and emerging illness.
That’s in the future.
The reality now? Technology is slowly reshaping our field.
Further reading: Find the Expanded Mental Health Strategy here.
Reading of the Week. Every week I pick a reading — often an article or a paper — from the world of Psychiatry.
How Smartphone Apps Can Treat Bipolar Disorder and Schizophrenia
By Davey Alba 11.20.14
Bryan Timlin always carries an iPhone and an Android phone.
The 57-year-old is an app and graphic designer with a Michigan company called OptHub, but he doesn’t carry two phones for work. He carries the iPhone because that’s what he likes, and he carries the Android because it’s what he needs.
The Android phone monitors his behavior. Five years ago, Timlin was diagnosed with rapid-cycling bipolar disorder, a mental illness characterized by four or more manic or depressive episodes a year. Some episodes, he says, can last as long as eight weeks. “Being bipolar is like jumping out of an airplane knowing you don’t have a parachute on,” he says. “You know you’re going to be hurt, but the high is so euphoric that it’s worth the risk. You can deal with the consequences later.” With his Android phone, he hopes to deal with these moments in other ways.
The phone, provided by researchers at the University of Michigan, includes an app called Priori that runs constantly in the background, using the phone’s microphone to analyze his voice and track when he is, and isn’t, speaking. Mania is typically marked by speech that’s loud and rapid, often with erratic leaps from topic to topic. Longer pauses or breaks can indicate depression.
At the moment, the app only collects data on his behavior. But the hope is that it eventually will use this information to warn Timlin and his doctor to an impending bipolar episode.
Priori is one of many efforts to address mental health through smartphone apps. Tools gestating within startups, academic institutions, and research clinics aim to help people manage everything from severe depression to bipolar disorder and schizophrenia. Through the discreet and continuous recording of social and physical behavior, these apps can detect changes in mental well-being, deliver micro-interventions when and where needed, and give patients a new awareness of their own illnesses. In the long run, they may even diminish the stigma attached to mental health disorders.
“The question isn’t whether or not this technology is going to be used in healthcare and monitoring individuals with psychiatric illnesses,” says University of Michigan psychiatrist Melvin McInnis, who developed Priori alongside computer scientists at the university’s College of Engineering. “The question is really: How?”
Most of these apps—which include CrossCheck, from Dartmouth Psychiatric Research Center, and Companion, from a Boston-based startup called Cogito—aren’t yet publicly available. But some projects have completed trials with small groups of patients, larger trials are underway, and preliminary results are encouraging. These apps are based on objective, contextual data, and they require little work on the part of patients.
But, certainly, there are many hurdles to overcome—most notably the potential for these tools to mislead patients and compromise their privacy. Finding ways of regulating such apps is as important as refining their technology.
“I think this will have a liberating effect, and will extend the boundaries of healthcare in a really enormous way,” says Dr. Jeffrey Lieberman, psychiatrist in chief at the New York-Presbyterian Hospital/Columbia University Medical Center. “But there are also ethical and legal principles that will need to be established.”
A Look Into the Future
In April 2013, when a bomb exploded at the Boston Marathon, Cogito was running a clinical trial of its mental health app Companion that included 100 people who happened to be in the area. “It was a look into the future of how a tool like this could sense the effect of a traumatic event in a population,” says Joshua Feast, CEO of Cogito.
Whereas Priori tracks only speech patterns, Companion taps a smartphone’s GPS location tools, accelerometer, and light and sound sensors to record movements, social interactions, the frequency and timing of phone usage, and speech behavior. It uses this data to create a profile that seeks to describe all sorts of things a patient may be dealing with. This might include physical or social isolation, sleep or mood issues, or, indeed, extreme trauma.
Like Priori, Companion does not yet use the data it collects to trigger a response, but the marathon bombing, Feast says, showed how well such a thing could work. “We don’t know for sure if someone has a certain condition,” Feast says. “But we know what the symptoms associated with those conditions are, and we can see when they’re going up and down.”
Dartmouth professor Dror Ben-Zeev says much the same thing in discussing his app, CrossCheck. He aims to equip the app with an alert system that will actively notify patients and their medical team when they experience certain episodes related to schizophrenia.
It too uses a wide array of smartphone sensors to track patient behavior over time, and the plan is to correlate this info with behavior reported by doctors and patients. “When there’s a relapse event—meaning, if they either self-report that their symptoms are getting worse, or if they wind up in the hospital—we track back and look at the sensor data for two to three weeks before that event,” Ben-Zeev explains. “We try to see: is there something that happened in those weeks that was different from their datastream up until that point?”
These markers constitute a patient’s unique “relapse signature,” and eventually, the system should recognize these signatures ahead of time, notifying the patient and the patient’s medical team before a major episode happens. Ben-Zeev’s ultimate goal is to not only reduce the number of relapses, but also prolong the period between relapses. “We can raise the red flag,” he says.
These projects remain in the early stages. Researchers have completed a pilot study of Priori involving six people, for instance, and are now testing the app on another 30, including Timlin. They hope to gather data over the next two years.
But many researchers believe these tools can completely change how we examine and treat mental illness. Unlike with other illnesses, there is more of an ebb and flow to how patients experience symptoms, and the only way to capture it accurately is to track someone around the clock. “Otherwise, you’re getting these very blunt and all-encompassing summaries rather than the real flavor of what patients have experienced,” Ben-Zeev says.
More than 60 million American adults—or one in four individuals—live with some form of mental illness.
As Ben-Zeev points out, the rise of the smartphone has—for the first time—provided a reliable means of tracking behavior in full. “The intensity and creativity of these things that are infusing the mHealth field, both the research and private sector, are directly linked to this amazing penetration of mobile phones,” Ben-Zeev says. “That’s what we’re trying to piggyback on.”
At the same time, using smartphones as a sensing tool could reduce stigma. Research has shown that many individuals with mental illness are uncomfortable using wearable devices that are conspicuous and purpose-built for research. Smartphones can add a veil of privacy.
If Ben-Zeev and other researchers successfully tap the power of the smartphone, the impact could be enormous. More than 60 million American adults—one in four individuals—live with some form of mental illness, according to the World Health Organization. Of those, nearly 14 million suffer from a serious mental illness, including schizophrenia, bipolar disorder, deep depression, or post-traumatic stress disorder. According to one study, these adults die an average of 25 years earlier than the general population.
These illnesses also affect friends, family members, and others around them. In the US, annual healthcare costs associated with these conditions exceed $30 billion.
Thermometer of a Different Kind
Bryan Timlin compares Priori to a thermometer or insulin testers. It merely reads what his body is doing. And because he is aware of those readings, he’s more attuned to how he’s feeling in relation to his illness.
He also likes the app because it’s harder to cheat the thing. When he goes in for a clinical assessment, for instance, he could lie about how he’s doing to get a better evaluation—something he’s been tempted to do on occasion. And when he’s having a manic episode and feeling good, Timlin often stops taking his medication. In the moment, he believes he doesn’t need it. The app can show him that he does.
As it stands, there’s no way to ensure that an app really does what it says it does.
But he also believes the app can eventually serve as an alternative to over-medication in at least some cases. “We don’t need all that we think we need,” he says. Mental illness also can be managed through diet and exercise, and he says Priori could help him find the right balance between this and medication.
But Jeffery Lieberman, of the New York-Presbyterian Hospital/Columbia University Medical Center, is quick to point out the potential for abuse. As it stands, there’s no way to ensure an app really does what it says it does. There’s no “Good Housekeeping Seal,” so to speak, for these mental health apps.
Yes, those coming out of academic institutions or professional healthcare organizations are sure to be rigorously vetting, Lieberman says. But he’s watching the private sector more closely. That’s where all the entrepreneurial energy comes from, after all. “But this is happening kind of piecemeal and ad hoc, and the question is: how can it be done more effectively and systematically?” he says.
Today, there are hundreds of medical-themed apps cluttering Apple’s App Store and in Google Play. Many of their claims are untested, and that has many medical practitioners concerned. We need a means of distinguishing the real tools from the fly-by-night variety.
Last year, the FDA released a document for guidance on medical apps, but it merely stated that the agency would focus its oversight on apps and phone attachments that make a specific medical diagnosis or turn a smartphone into some kind of instrument, like an ECG machine. There is no mention of guidelines specifically for mental health apps.
Just Getting Started
Adam Haim, chief of clinical trials operations and biostatistics at the National Institute of Mental Health, says efforts to test—let alone regulate—these apps are only getting started. “You’re getting a whole host of new data from different sources, and if intervention can be made more efficient, I think practitioners would rally behind that,” he says. “But it would also require additional training, and it would be an iterative process.”
But he also says that these apps will eventually prove fruitful: “All signs are pointing toward the integration of evidence-based technology into mental healthcare.”
Timlin agrees. He’s unswervingly optimistic about Priori, and looks forward to the next six months, when he will continue to use the app as part of its clinical trial. “I’m always going to be bipolar,” he says. “But this condition can be managed.”