GUY RAZ, HOST:
It's the TED Radio Hour from NPR. I'm Guy Raz. And on the show today, ideas about Crisis and Response. So far on the show, we've been talking about how we humans respond to crises. But what if technology could help us respond better and faster?
NANCY LUBLIN: The number is 741741, and you text it like you're texting your best friend.
UNIDENTIFIED MAN #1: I can't take my family.
LUBLIN: There's real pain out there.
RAZ: Nancy Lublin...
LUBLIN: And we're seeing it every day.
RAZ: ...Founded Crisis Text Line.
UNIDENTIFIED WOMAN #1: I'm so nervous, and it's making me nauseous.
RAZ: Crisis Text Line is just what it sounds like, and they get messages from every area code in the U.S. And every day, people use it to connect with counselors to ask for help with anxiety or depression or worse.
UNIDENTIFIED WOMAN #2: I want to die or run away.
LUBLIN: Before starting this interview, I looked down at my phone - 'cause I have the live stats on my phone - and there are 12 suicidal people right now texting with us while you and I are talking.
And Nancy can use every text that comes in...
UNIDENTIFIED MAN #2: I feel completely invisible.
RAZ: ...To better understand moments of crisis...
UNIDENTIFIED WOMAN #3: I feel like all my friends and leaving me. What do I do?
RAZ: ...What triggers them and how to respond.
UNIDENTIFIED WOMAN #4: I kept saying stop. I don't like this.
LUBLIN: And it's so amazing to see how just someone being there for them and responding and caring and asking questions and validating can take the heat out of that moment.
RAZ: If any text to Crisis Text Line comes from a suicidal person who suggests they might try something, Nancy's counselors trigger something called an active rescue which alerts the local police. And that's what happened to one girl Nancy mentioned in her TED Talk. She sent in a text saying, I want to die. I have a bottle of the pills on the desk in front of me.
(SOUNDBITE OF TED TALK)
LUBLIN: And so the crisis counselor says, how about you put those pills in the drawer while we text? And they go back and forth for a while, and the crisis counselor gets the girl to give her her address because if you're texting a text line, you want help. So she gets the address, and the counselor triggers an active rescue while they're texting back and forth. And then it goes quiet, 23 minutes with no response from this girl. And the next message that comes in says, it's the mom. I had no idea, and I was in the house. We're in an ambulance on our way to the hospital. As a mom, that one just - that just - the next message comes a month later. I just got out of the hospital. I was diagnosed as bipolar, and I think I'm going to be OK.
LUBLIN: Now, I would love to tell you that that's an unusual exchange, but we're doing on average 2.41 active rescues a day. The beautiful thing about Crisis Text Line is that these are strangers counseling other strangers on the most intimate issues. It's exciting. And I will tell you that we have done a total of more than 6.5 million text messages in less than two years.
RAZ: That number is now more than 15 million. And with all those messages coming in, Nancy and her team saw an opportunity. They thought, maybe we could turn those text messages into data, data that could be the difference between life and death.
LUBLIN: We decided that we wanted to take people based on severity and not chronology. So when you call most customer service places or when you call other phone hotlines, you're taken chronologically...
LUBLIN: ...In the order that you come in. But shouldn't it work more like a hospital emergency room...
LUBLIN: ...Where the gunshot wound is taken before, you know, the kid with a funny rash...
LUBLIN: ...Or the sprained ankle? Well, that's how we operate. If you text in, I want to die, I want to kill myself, we code you orange, and you are number one in the queue.
RAZ: So what Nancy and her team can do is feed every text into a database and then a computer algorithm can start to make sense of things that at first seem nonsensical, like a few months ago when the algorithm started identifying people as suicidal when they texted in #KMS.
LUBLIN: I had no idea what that was. Apparently, it became shorthand for kill myself, and people have been using that on the Internet. The algorithm picked that up and made those people number one in the queue.
LUBLIN: And so those people are then getting a human eye and a human response in 1.8 minutes.
(SOUNDBITE OF TED TALK)
LUBLIN: We have the data to know what makes a great counselor. We know that if you text the words numbs and sleeve, there's a 99 percent match for cutting. We know that if you text in the words MG and rubber band, there's a 99 percent match for substance abuse. And we know that if you text in sex, oral and Mormon you're questioning if you're gay. Now, that's interesting information that a counselor could figure out. But that algorithm in our hands means that an automatic pop-up says, 99 percent match for cutting. Try asking one of these questions to prompt the counselor. Or 99 percent match for substance abuse - here are three drug clinics near the texter. It makes us more accurate. I can tell you that the worst day of the week for eating disorders - Monday. The worst time of day for substance abuse - 5 a.m. And that Montana is a beautiful place to visit but you do not want to live there because it is the number one state for suicidal ideation. And we've made this data public and free and open. We've pulled all the personally identifiable information, and it's in a place called CrisisTrends.org because I want schools to be able see...
LUBLIN: ...I want schools to be able to see that Monday is the worst day for eating disorders, so that they can plan meals and guidance counselors to be there on Mondays. And I want families to see that substance abuse questions spike at 5 a.m. And I want somebody to take care of those Native American reservations in Montana.
RAZ: I'm amazed, like, that you've been able to sort of find out, like, these moments - right? - like 5 a.m. is the worst time for substance abuse, and, you know, depending on the day of the week people can behave - are more likely to behave in certain ways.
LUBLIN: Yeah. It's illuminating. And you can just go down a rabbit hole on that website and it - because it's all real time, it changes. So in October, I turned to our chief data scientist when some of the anti-Muslim rhetoric was really heating up in the presidential campaign. And I said, you know what? Just pull conversations where someone indicates that they're Muslim and see what's happened since this. And there was a 600 percent increase in October, and I think it was a 900 percent increase in November in people saying that they were Muslim and either were experiencing anxiety, bullying, depression, any number of those issues.
It's just really exciting to have the data sets and put them out there and ask the questions. But there are people who will then research and find out, why is this or change policies based on this data. You know, information can save lives.
RAZ: Nancy Lublin is the founder of the Crisis Text Line. You can see her full talk at ted.com. And if you need help, the Crisis Text Line number is 741741.
NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.