I get e-mail alerts every time a bipolar friend has a manic episode. The alerts come from mutual acquaintances who send links to "insane" Facebook posts and social-media utterances. "He's been posting messages at a strange rate on Facebook," one wrote. "He was posting SoundClouds at 4AM," another texted. One sent screenshots of suspicious tweets. "Is he manic again?" another asked. "He is being insane on 4square, but sometimes I can't tell if everyone is just more active on social media than I am."
"Someone needs to redefine 'bystander effect' for not-quite-friends of unwell people on Twitter," I replied, somewhat cattily, during a recent episode. I was irritated because I knew it was true; I'd seen my friend's mania manifested in hyperactive Twitter sprees before. When I called him, he confirmed he was ill and asked for help seeking care. The armchair psychiatrists had been right — even though some had not spoken to my friend in years, they knew his social-media patterns well enough to recognize an aberration.
As human interaction moves online, so do social judgments and armchair diagnoses. This is, sometimes, charming. "Is your son sick?" one co-worker recently asked another. "You didn't Instagram any videos of him this weekend." If I go more than two days without tweeting, my mother calls to make sure I'm okay — the long-distance version of when she would peek into my room while I slept as a child, just to make sure I was there and breathing.
But the digital diagnostic process is also unnerving. Watching an erratic mind express itself can be engrossing, so the mentally ill may attract rubberneckers. That was the case for actress Amanda Bynes, whose pre-hospitalization Twitter feed seemed for a while like it could have been performance art. Rubberneckers also circled around Charlie Sheen in 2011, when his Twitter rants inspired psychiatric speculation, memes, and profitable gimmicks alike. And now the disordered limelight falls on "porn professor" Hugo Schwyzer, whose live-tweeted manic meltdown has been featured in the Daily Mail, the New York Daily News, and the Daily Beast. (The Cut interviewed Schwyzer a week before his self-described "manic break.") Between tweets about his anti-psychotic medications and suicidal thoughts, Schwyzer at one point scoffed, "You can't 5150 me from Twitter." 5150 refers to the involuntary psychiatric hold that landed Bynes in the hospital last month, putting an end to her tweets.
But can mental illness be diagnosed online? We posed the question to American Psychiatric Association president Dr. Jeff Lieberman, who is also chair of the department of psychiatry at Columbia University's medical school. He said social-media utterances, like any other type of utterance, can play a part in psychiatry.
"Individuals who are manic, depressed, psychotic, anxious, addicted — if they're utilizing social media for communication, they will utilize it in ways that reflect their symptoms or their behavior dysfunction," Lieberman said. "This is not surprising in the least. It will depend, of course, on the degree to which the person has integrated these forms of communication into their modes of behavior."
"It's no different than the person who is sitting there talking gibberish or acting bizarre. This is another manifestation — it's coming through this new communicative form, social media, but it's data." The communication age "poses a real opportunity, but also a real challenge, for physicians and health-care providers." Changes in information-gathering techniques open doctors to legal risks, while privacy regulations may limit avenues for communication. The greatest value, he argued, is not in diagnosing illness but monitoring it.
Lieberman pointed me to Dr. David Kimhy, director of Columbia's experimental psychopathology lab. Kimhy's lab uses custom PDAs to monitor psychosis. He programs Palm Pilots to beep at intervals during the day, presenting schizophrenic patients with questions about their symptoms — how they feel, if they are hearing voices, etc. Doctors may tailor therapy to the patient's patterns while reducing memory biases in patients' self-reporting. Kimhy also has experimented with using PDAs for "rudimentary" therapeutic purposes — intervening at moments of high anxiety with reassurance or a reminder to distinguish reality from delusion, for example. (Ill people may seek reassurance and reality checks online, too; think of Bynes seeking compliments about her appearance.) A team at Dartmouth is doing similar work with smartphone therapy for schizophrenics, while researchers at MIT have developed voice-analysis software that can identify depression from changes in vocal patterns identifiable by cell phone.
"The technology and computational mathematics is there," said Dr. Jon Morgenstern, a member of Columbia's department of psychiatry working on iPhone apps to monitor heavy drinking with a combination of interactive questionnaires, voice analysis, and motion sensors. "Eventually we think phones can be a little like a lab test, or like wearing a heart monitor around your chest."
In the infinitely less precise world of social media, signs of distress and appropriate responses to them are less clear. "If you receive messages from a friend that sound bizarre or distraught, I would encourage you to inquire as to how your friend is doing and verify whether those messages accurately describe their state of mind," Lieberman advised. "And if so, I would suggest bringing it to the attention of a significant other or family member, or perhaps suggesting your friend seek help."
But what of the people we don't actually know, the Byneses, Sheens, and Schwyzers? "I suppose it could be a scenario for a Hitchcock movie, like Rear Window or something," Lieberman chuckled, noting that bystanders' dilemmas tend to be ethical, not medical, dilemmas. "We're in uncharted territory."
Of course, many social media users are not just bystanders — they're trolls. Could using social media actually worsen a sick person's symptoms, when ill-intentioned acquaintances or strangers lob insults or suggestions of suicide, as many Twitter users did with Schwyzer during his meltdown? "Yes, because when a person is in the throes of a manic episode, they are hyper-reactive to stimuli, sensory and intellectual. Receiving messages online could have an agitating effect on them." The feedback loop could push an unstable person to dangerous behavior like self-harm; ditto for depression and other mood disorders. "It doesn't necessarily mean their illness is going to have a more progressive degenerative effect on the brain, but it could increase the severity of the symptoms," he concluded.
Morgenstern noted that social networks and social feedback are significant for addiction therapy. Changes in social networks "change people's perceptions of social norms, so you begin to think to yourself, What I am doing is actually fairly similar to what other people are doing."
Two months after his episode, my bipolar friend reflected on his period of hyper-communication. "Facebook thought I got hacked and shut down my account. There was this red box when I tried to log in that said my account had been temporarily disabled. I remembering thinking it was a game." Apparently his mind incorporated technology into his delusions, a fact that made him laugh in retrospect. "I thought I was an Apple Genius, but not like the one in the store. Like it's a special honor they bestow, like a Nobel Prize. I shit you not. I thought I had special Apple Store access, a lounge in the back, like an airport Admiral's Club."
He hasn’t looked back at the digital record of his episode, though. “It's hard enough to move forward after the end of a manic episode. I barely had the motivation to get out of bed, much less revisit all the embarrassing things I said or did online." He recalled that I told him to stay off Facebook and said his family did, too. "At the time I wondered if they weren't saving themselves from embarrassment."
What does he think now? "I'm not sure, but I don't dwell on it. Everyone was inevitably trying to do what was best," he said. "That part was hard, too, actually. I feel bad saying this because I'm lucky to have people who care about me, but you can only deal with so many interventions."
When it comes to interventions, Lieberman noted the risk of miscommunication. "The other day I was texting with a patient and got back a response that was garbled. And I'm thinking, What is he saying? What does that mean? Is there a problem?" A moment later the patient wrote back. "Ignore. Pocket text."